Advances in Mobile Robotics

PREFACE

CONFERENCE ORGANISERS

CONFERENCE COMMITTEES

CONFERENCE SPONSORS AND CO-SPONSORS

CONTENTS

No abstract received.

No abstract received.

Climbing robots are now widely accepted as valid options in situations where it is important to move on sloped or vertical structures in order to inspect, paint, clean or perform the required operations. Even if the first applications of climbing robots appeared more than 40 years ago, many new ideas are continuously being proposed in the scientific literature and in the market.

In this work a classification of the different adhesion techniques proposed for climbing robots is proposed and discussed. Adhesion methodologies can be classified as active when they require an external energy supply to support the robot, or passive if no energy is needed (e.g. permanent magnets or suction cups). Another classification can be done on the basis of the nature of the forces required to support the robot: pneumatic, if the adhesion force is generated by a pressure difference; magnetic if the force is magnetic; mechanical if it depends only on mechanical supports, chemical if it is due to some particular glue, or electrostatic. Moreover within each of these categories, different kind of robots have been proposed in the last years, also on the basis of the locomotion architectures: walking with legs, frame walking, with wheels, sliding, jumping, etc.

Recently biologically-inspired gripping methods, trying in many cases to imitate gecko skin, appeared in several research works. However some doubts remain concerning the applicability of such systems in real applications.

Some critical considerations on the different techniques and on their practical advantages and drawbacks will be exposed and an overview of the different climbing robots developed in the last 12 years at University of Catania is also presented.

This work presents investigations into controlling a two-wheeled robotic machine (TWRM) with a payload positioned at different locations along its intermediate body (IB). Two types of control techniques are developed and implemented on the system model. The traditional proportional-derivative (PD) control and fuzzy logic (FL) control. An external disturbance force is applied to the rod that constitutes the IB in order to test the robustness of the developed controllers. The simulation results of both control algorithms are analyzed on a comparative basis for the developed two control techniques.

A control technique is developed to balance a two-wheeled robotic machine (TWRM) with a payload in a sliding motion along its intermediate body (IB). The balancing of the robot has to be achieved during the motion of the vehicle and the payload. An external disturbance force is applied to the rod in order to test the robustness of the developed controller. Investigations are carried out on the effect of changing the level and duration of the disturbance force, and changing the speed of the payload on the system during the balancing mode. The simulation results with two control algorithms are analysed and discussed on a comparative basis.

This paper presents a cooperative climbing robot which follows a welding robot for the inspection of long weld lines simultaneously with the automatic welding process. The robot is designed to climb on ferrous surfaces with strong neodymium magnets elevated by 20mm from the work surface for adhesion which is able to cope with 20mm height obstacles on the surface, such as weld caps etc. There are two driving wheels differentially controlled to provide smooth and continuous movement and two omniwheels, one in the front and the other one in the back to support the robot. The design of two sections jointed by a hinge joint mechanism increases the robot capability of moving on both concave and convex curvatures as well as transferring across angled surfaces. The robot is able to follow the welding robot by looking at the welding robot and the hot weld point using a number of sensors. The wireless control of the robot expands the working range of the robot without concerning the umbilical problem experiencing on many climbing robots. The safety factor of the robot climbing on vertical surface with the NDT payload is 2.3.

Power consumption affects a lot on the humanoid robot's performance. Therefore, the estimation of a humanoid robot's power consumption in the routine activities, such as walking, is a very important issue at the initial stage of the mechanical design of the humanoid robot. In this paper, dynamic walking is considered to be continuous oscillation, and a dual pendulum model is established in order to analyze a humanoid robot's power consumption when it walks at the ground level. Based on this model, we will show that the power consumption by a robot could be minimized by selecting an optimal value of the walking frequency. Interestingly, such a value turns out to be quite close to that of a human being's walking. In addition, the model is proven to be useful to determine the best placement of the hip joint for the purpose of minimizing the humanoid robot's power consumption.

The process of clustering and classification of erroneous patterns that may appear in the behaviour of a system is an inevitable constituent of any intelligent system. The paper is focused on the analysis of the sensed signals for purposes of the early detection and classification of imminent faults. In particular, the strings of the foot placing-lifting signals are grouped into clusters based on their similarity. The same is done with samples of the joint torques. The clustering is carried out by the adaptive resonance theory (ART) - based neural network. The results obtained with the experimentation with real robot demonstrate feasibility of the suggested approach.

A novel mobile self-reconfigurable robot is presented. This robot consists of several independent units. Each unit is composed of modular components including ultrasonic sensor, camera, communication, computation, and mobility parts, and is capable of simple self-reconfiguring to enhance its mobility by expanding itself. Several units can link into a train or other shapes autonomously via camera and other sensors to be a united whole robot for obstacle clearing, and disjoin to be separate units under control after missions. To achieve small overall size, compact mechanical structures are adopted in modular components designing, and a miniature ARM-based embedded controller is developed for minimal power consumption and efficient global control. The docking experiment between two units has also been implemented.

This paper explains how to singularize a robot to a unique hypothesis state from a state of multiple hypotheses. This it does by computing a path whose expected distance is minimum from a tree where each path to the leaf from the root results in a singular hypothesis. At various nodes of the tree the number of hypotheses is reduced as it proceeds down from the root. The nodes of the tree are computed as the best locations to move from an earlier higher hypotheses state. They are either frontier regions or a direction of traversal that result in reduction of hypotheses from an earlier multi hypotheses state. The method has been tested robustly in various simulation situations and its efficacy confirmed.

The walking robots are built to displace the loads on the not-aligned terrain. The modular walking system protect much better the environment when its contact with the soil is discrete, a fact that limits appreciately the area that is crushed. At the Polytechnic University of Bucharest has been new developed a walking modular robot to handle farming tools. This walking robot has three two-legged modules. Every leg has three freedom degrees (RRT) and a tactile sensor to measure the contact, which consists of lower and upper levels. The body of the MERO (MEchanism RObot) walking robot carries a gyroscopic bearing sensor to measure the pitch and roll angles of the platform. The legs are powered by hydraulic or electric drives and are equipped with transducers and sensors. They are used to control the walking robot in the adaptability to a natural ground. A vehicle like that is Romanian Walking Robot MERO (Fig. 1)

A main challenge in the field of service robotics is the navigation of robots in human everyday environments. Supermarkets, which are exemplary chosen here, pose a challenging scenario because they often have a cluttered and nested character. They are full of dynamic objects. Especially the presence of large numbers of people is a special challenge to cope with. It can often be difficult to map the locality because of a frequently changing environment. Therefore a broad approach is needed to cover three main tasks: the reactive local navigation, the interaction with dynamic objects and the flexible global navigation and task planning. A three layered navigation concept was developed where each of these fields is dealt with in a dedicated layer. This paper presents the bottom layer of this three-layered approach. The focus lies on providing an reactive local navigation system that is able to accomplish movement tasks in a complex scenario. The control is based on behavior networks. To enhance the manoeuvrability of the robot (Fig. 1) it uses a holonomic drive system with mecanum wheels. This paper is associated with the CommRob project.

A main challenge in the field of service robotics is the navigation of robots in human everyday environments. Supermarkets, which are exemplary chosen here, pose a challenging scenario because they often have a cluttered character. They are full of dynamic objects. Especially the presence of large numbers of people is a special challenge to cope with. It can often be difficult to map the locality because of a frequently changing environment. Therefore a broad approach is needed to cover three main tasks: the reactive local navigation, the interaction with dynamic objects and a the global navigation and task planning. A three layered navigation concept was developed where each of these fields is dealt with in a dedicated layer. This paper describes the top layer with a semantic navigation using RFID tags as landmarks. It is based on a topological map with semantic attributes. Barriers of RFID tags are used to discriminate the environment into topological areas. Combining a topological navigation with a behavior-based control makes the navigation independent of a global metrical map. This paper is associated with the CommRob project.

This paper presents the investigation of two climbing robot attachment mechanisms. The two mechanisms are tested to find their maximum holding forces under a range of conditions, and models to predict these forces are developed. We describe the experimental methodology and the results of testing three types of vacuum pads and an electromagnet are presented, which are then compared to the predictive models. The outcome of this work is to be used to develop a fully autonomous and efficient method of adaptive control for climbing robots.

This article presents a method for the creation of a topological map without having to previously create a geometric map. Independently obtained topological paths are compared pairwise to search for possible overlap in the view sequence. Each individual topological path is created from a sequence of raw data views that are sampled by leading the robot along a path in the environment. The multiple topological paths are ‘stitched together’ into a generic Topological map by identifying overlapping segments in the individual sequences. A general topological map can be created by considering all the multiple sequences or separate runs through the environment. Results on the merger of upto 8 separate paths indicate that this method of mapping large environments, by selective touring through the environment, can generate robust topological maps.

This paper presents an overview of the latest robot standardization activities being carried out under ISO TC184/SC2 to address changes in the robotics sector. Several new activities have been initiated recently to address the emerging emphasis of service robots and the shift away from manufacturing robots in industrial sectors. The emphasis of increasing human-robot collaborations appears to be a key requirement that needs to be satisfied to allow safe service robots to be developed and supplied to the community; the growing human-robot interactions is present even within manufacturing robots in industrial environments where collaborative robots are being proposed. The paper reports on the generic issues that are emerging and need to be addressed to allow the robotic community to move forward collectively and the need for revising the robot vocabulary for the new applications. Formulating a safety standard for the new service robots is essential so that close human-robot interactions may be permitted. In fact the issues that need to be specified for safe human-robot contact are being formulated for personal care robots covering both non-invasive and invasive applications.

As the complexity of developed and investigated clawar systems grows it is more and more relevant to be able to replicate results and to compare different implementations and different approaches, in order to enable the cumulative advancement of our knowledge and even to be able to assess disruptive changes. In this short discussion it is first surveyed the state of the art as regards replication and benchmarking in Clawar systems, in particular, and robotics, in general, showing both the opportunities and the issues related to such a program, then a possible roadmap for the future is proposed.

Motion control is a key aspect for the performance of wheeled mobile robots (WMR). There are several well known motion control methods for WMR, but usually those methods are very dependent from the robot physical constraints. One of the most commonly employed platforms to propose and demonstrate motion control algorithms is the differential drive platform and its variants. Every time a new motion control algorithm is proposed, some comparisons with traditional algorithms are usually made, but there are no set of benchmarks globally accepted to assess the performance of motion control algorithms. This work tries to make some contributions in this area, analyzing the problem and proposing a set of benchmarks to be used in the evaluation of mobile robots' motion control algorithms.

This paper presents the results of benchmarking the robot design process with an emphasis on robot component modularity that has been carried out within the CLAWAR community. A generic design methodology has been developed based on the CLAWAR “Interaction space”. The CLAWAR work on robot design benchmarking has adopted a Delphi study approach that involved robot experts who performed a number of conceptual robot designs for two application sectors (urban search and rescue and surface cleaning) to meet given specifications and key criteria. The individual designs produced have been assessed by the robot experts using various metrics via a specially developed benchmarking questionnaire based on a live point scale (disagree, partially agree, agree, strongly agree, 100% satisfied). Brief details of the Delphi study methodology adopted, the conceptual designs produced, evaluation questionnaire and the benchmarking results obtained are presented to bring out the generic results of the benchmarking study.

In this paper we present a neural network architecture for learning of robotic grasping tasks. The neural network model allows acquisition of different neural representations of the grasping task through a successive learning over two stages in a strategy that uses already learned representations for the acquisition of the subsequent knowledge. Systematic computer simulations have been carried out in order to test learning and generalization capabilities of the system. The proposed model can be used as a high level controller for a robotic dextrous hand during learning and execution of grasping tasks.

In this paper, we present a model of emotions that we proposed in EmotiRob project. We make a comparison of recent models of emotions and show that our model is generic in basing on the theories of emotions of Ortony et al., Lazarus, Scherer and then the personality theory of Meyers-Brigg and Meyers.

In this paper we study the locomotion behavior in the terrestrial tortoise, Geochelone graeca to present a biomimetic approach for modeling and controlling a virtual model of tortoise like robot. While the work performed focuses on a particular animal, the method used is not specific and can be applied for others. Many Experiments in vivo and vitro were performed on adult specimens, including X-ray filming of animal in motion, 3D scan of bones and Schell of naturally dead tortoises and measures of the contact forces between the animal and the ground. The collected data provided us with the inertial proprieties, the lengths and masses of body limbs, the nature of joints, their corresponding axis of revolution, the angular trajectories between joints and finally the ground reaction forces on each leg. The physical proprieties were useful to build the virtual robot while the real recording of the animal in motion were used to control it.

Applications of biological methods and systems found in nature to the study and design of engineering systems and modern technology are defined as BIONICS. The present paper describes a bionics application of shape memory alloy in construction of orthopedic implant. The main idea of this paper is related to design modular adaptive implants for fractured bones. In order to target the efficiency of medical treatment, the implant has to protect the fractured bone, for the healing period, undertaking much as is possible from the daily usual load of the healthy bones. After a particular stage of healing period is passed, using implant modularity, the load is gradually transferred to bone, assuring in this manner a gradually recover of bone function. The adaptability of this design is related to medical possibility of the doctor to made the implant to correspond to patient specifically anatomy. Using a CT realistic numerical bone models, the mechanical simulation of different types of loading of the fractured bones treated with conventional method are presented. The results are commented and conclusions are formulated.

The crawling motions of snakes and other limbless animals have always been of great interest for specialists in mechanics and biomechanics. In this paper we intend to present a new locomotion mode which is inspired of snakes, as rectilinear locomotion gait, for perform in a mobile robot. To this end, first we show a simple model of motion mechanism, based on a manipulator, and develop its kinematics formulation. Then, the dynamic problem of the gait design is investigated and torque values are needed to achieve the desired joint motion will be computed.

The aim of the project InspiRat is the design of a biologically inspired climbing robot, capable to move in complex und semi-structured environment. High motional ability should not only be achieved by the control system. An intelligent mechanical design reduces the mass of the robot, increases the ability of adaptation to the different substrates and minimizes power consumption.

This paper proposes a method to control a multi-track type robot developed for reconnaissance missions. A centipede decides body motion using antennae on its head instead of using eyes. Inspired from that, we imitated the antennae using 3 IR sensors attached in front of the robot. Based on the information got from the sensors the robot decides next behavior automatically. By the proposed control method, the robot was controlled effectively in various environments. The control method and experiment results are mentioned in this paper.

In this paper, we proposed an analog neural controller (ANC) that is similar to physical modeling of the nervous system. ANC is composed of sensory neurons, motor neurons, synapses and dendrites. The functions of sensory neurons generate CPG signals and motor neurons change the sensory neuron signal for driving actuators. Synapses connect between neurons and the weigh to be given. Dendrites can accept external signal from the sensor. Because of parallel structure which is one of main characteristic of ANC, it is possible to walk by generating new pattern even though one of the neuron is out of working.

Biarticulate muscles – muscles that span more than one joint – are ubiquitous in nature and are believed to play an important role in locomotion. In particular, biarticulate muscles may be important during running and jumping tasks; tasks at which most current bipedal robots are deficient. In human, it is known that biarticulate muscles greatly assist in transferring energy from muscles near the body (i.e. thigh muscles) to muscles far away (i.e. ankle muscles). Here we reproduce this effect in a robot leg design based on the muscle architecture of the human leg, including biarticulate muscles. We demonstrate that biarticulate muscles can transfer a significant amount of work and power from upper limb segments. Further, we show that peak power transfer is dependent on the exact timing between biarticulate muscles and the monoarticulate muscles that they assist. We show analytically that the pull-but-not-push property of muscles is critical to ensure only positive work transfer from upper limbs to lower limbs.

The first part of this paper presents the unique behavior of Ionic Polymer-Metal Composites (IPMC) that respond quickly with a following back relaxation when used as actuators. This behavior is frequency dependent and appears within the lower spectrum frequencies. A model is presented that predicts the behavior of the IPMC actuator within its whole frequency spectrum separated into slow and fast modes. The frequency dependent output characteristics of the IPMC sensors are also shown. The second part of the paper presents the design of an Integrated Sensor-Actuator Device (ISAD) using IPMC as both the sensing and actuating elements. The need and potential of the two-element type of arrangement is addressed, which can be seen as a starting point to the development of more complex devices incorporating many sensing and actuating elements. The ISAD performance is compared with the individual performances of the same strips used prior to integration.

In this paper we present the design, analysis, simulation and implementation of a novel design for a naturally stable climbing robot that has been inspired from human pole/tree climbers. The other features of this robot are its simple design, ease of control, light weight, simple mechanism and fast climbing speed. The robot consists of three wheels, two free and one active wheel which enable the robot to climb or descend poles. The free wheels are almost frictionless while the active wheel has enough friction to be able to apply force on the pole for stable climbing or descending. The wheels are designed such that the robot can compensate for misplacements eliminating possible detachment from poles. The static analysis of the robot is formulated and the robot is simulated and actually tested. The results show the unique characteristics of this robot that make it more stable if more weight is carried.

3DCLIMBER is a pole climbing robot designed at ISR-UC which is able to climb from poles with bents and branches. The first prototype of the robot uses electrical motors as actuators. A research started to study possibility of using other types of actuators to reduce the weight of the robot and thus increase the efficiency. This paper presents a comparison study between the pneumatic muscles and electrical motors. The 3DCLIMBER robot was used as a case study for this comparison.

This paper describes the effect of voluntary upper body movement in assisting the indoor rowing exercise for paraplegics. The indoor rowing exercise is introduced as a total body exercise for rehabilitation of function of lower extremities through the application of functional electrical stimulation (FES). The model is developed using Visual Nastran Software. Three different damping levels of the flywheel are set up to provide different pulling forces during the pull phase of rowing manoeuvre. Fuzzy logic control is implemented to control the knee and elbow trajectory for each of the damper level settings. The generated level of electrical stimulations for activation of quadriceps and hamstrings muscles are recorded and analysed. In view of good results obtained, it is concluded that the voluntary upper body movement with proper rowing manoeuvre significantly contributes to assisting paraplegics’ indoor rowing exercise.

The indoor rowing exercise is introduced as a total body exercise for rehabilitation of function of lower extremities of paraplegics through the application of functional electrical stimulation (PES). Fuzzy logic control (FLC) is implemented to control the knee and elbow trajectory for the purpose of smooth rowing manoeuvre. Conventional FLC methods rely on simple experiences and trial and error for parameters identification. In this work, the FLC is optimized using genetic algorithms (GAs), a stochastic global search method. GA implementation incorporates dynamic crossover and mutation probabilistic rates for faster convergence. The objective function of GA is to maintain smooth rowing manoeuvre. The performance of the proposed GA-tuned FLC is compared to a conventional manually tuned FLC. It is demonstrated that the developed controller offers encouragingly better performance.

A novel control technique for FES assisted paraplegic rowing is introduced using computer simulation. The paralyzed extensor musculature of the lower limbs is electrically activated during the drive phase. An energy storing mechanism assists the recovery phase. In this way the number of stimulation channels required is minimized whilst rowing performance is maintained. A finite-state controller (FSC) is used for FES control to ensure coordination between voluntary upper limb motion and simulated leg movements. A four-channel, surface electrodes FES system is used to activate the quadriceps and hip extensor musculature during the drive phase based on command signals from the FSC. The FSC takes input from an instrumented rowing machine. In this approach, FES is applied in short bursts thereby preventing the early onset of muscle fatigue and prolonging the exercise period.

This paper reports on the development and preliminary evaluation of an electrode array. The array consists of 64 mini electrodes, which may be grouped together to form a software-controlled ‘virtual’ electrode. The array design is presented and an experiment is described that shows it is possible to locate appropriate virtual electrodes within the array.

A new form of body-weight supported treadmill walking technique is presented in this paper which may serve as a suitable form of therapeutic intervention with minimal therapists support. While the technique relies on spring brake orthosis (SBO) for the swing phase generation, hip extension is produced by the foot being literally dragged backwards by the treadmill belt, appropriate trunk position and orientation being simultaneously maintained by voluntary hand support. Voluntary hand support is simulated with fuzzy logic controllers and the whole gait cycle is controlled as a finite state machine with event detection from kinematic data. The simulation resulted in a complete gait cycle whose parameters were largely dependent on the chosen treadmill speed. The torque requirement from the hand joints (shoulder and elbow) for the successful implementation of the gait cycle is also analysed.

In this work a vortex-based suction cup is firstly described. This new kind of active suction cup is built upon an idea of Duke University and composed by a rigid plastic cup with a rotating centrifugal fan, powered by means of a small brushless DC motor. The fan generates a vortex in the centre of the cup that finally creates a strong adhesion force. The cup could be of non-contacting type, reducing friction between cup and wall when it is used as sliding suction cup, like in the Alicia VTX climbing robot, increasing movement speed, allowing passing over small obstacles on the wall and avoiding the construction / maintenance of sealing between cup and wall. In order to better understand and describe the cup behaviour, a set of measurements of the cup internal pressure and associated payload have been done. Based on these measurements, a dynamical model between motor speed and adhesion force has been computed. The model structure will be described and some result will be showed.

The non-destructive inspection of large concrete walls via robotic systems is no longer an unsolved problem. This paper will present first results with the climbing prototype CROMSCI which uses a vacuum system of seven controllable vacuum chambers and an omnidirectional drive to move and cling to vertical concrete surfaces. This platform is able to move and inspect vertical surfaces safely, fast and cost-efficient. The technician can check the building more safe without any telescopic crane or other complex access devices via remote control or semi-autonomously.

In this paper, the mechanical design and calculation of a magnetic wheeled climbing robot is presented. It is able to pass obstacles that previously had only been accessed by robots with more complex mechanisms, but only needs 2 actively controlled DOF. A comparison to other design alternatives and the influence of some core parameters are shown in calculations with simplified 2D models. According to these calculations, a prototype was realized to prove its functionality in real tests. The paper concludes with an outlook on further design improvements and shows possible scenarios for its industrial application in the field of power plant inspection.

Rescue robotics for disaster response is of relevant practical importance to support rescuers. Main tasks are inspection of collapsed areas and localization of victims. Rescue robots need to be agile in order to penetrate even the smallest openings in the rubble. Often paths in the rubble are steep and climbing ability is required to the rescue robot to move along. The efficacy of locomotion intended as amount and distribution of thrust force transmitted to ground contributes remarkably to determine agility and usability of the robot. A particularly efficient locomotion technique called ‘sliding sock’ has been proposed. The paper presents the application of sliding sock locomotion to a modular serpentine robot with climbing ability.

The Industrial Automation Institute (IAI-CSIC) has been very active since its foundation (1971) in providing innovative and advanced solutions to a relevant number of industrial problems. Applied research in the field of Automatic Control paved the way to investigate robot design and control, where contributions can be traced back for more than thirty years. Both industrial and special purpose robots were the subjects of research and technical development. Since twenty years ago, the Automatic Control Department of the IAI-CSIC is researching in climbing and walking robots. In this paper, the evolution and perspectives of research in the field of climbing robots at the IAI-CSIC are revisited, and, on this foundation, lessons learned are revealed and new directions are proposed.

In robotic demining, the robot relies on a path-planner capable of generating trajectories to search for all the mines while avoiding obstacles whose locations are unknown. Several families of coverage algorithms exist but there is only one that guarantees complete coverage, the exact cellular decomposition family. This paper details the modifications performed to a cellular decomposition method for unstructured environments for its application to walking robots. Experiments show preliminary results and improvements to the method are proposed.

This paper addresses the searching of indoor fires with mobile robots using olfactory information. An experimental environment and devices necessary to study maze searching algorithms using olfaction are described. A maze solving algorithm, taking into account the smoke direction is proposed and validated in a simulated scenario. A simulated Khepera III mobile robot equipped with an olfaction system and an infrared array is used in a set of trials searching small simulated fires inside a weakly ventilated maze with 4 × 3 meters.

Locomotion over rough terrain has been achieved mainly by rigid body systems, including crawlers and leg mechanisms. We have proposed an alternative method, which uses deformation of a robot body, and developed a prototype of this deformable robot. But, electric power was externally supplied, such that power supply cables hindered the locomotion of the robot. We describe here the rolling locomotion of a deformable soft robot with an internally supplied power source. We applied dynamic simulation with particle-based modeling to analyze the rolling motion of this robot and found that increased weight had little influence on the kinematic performance of this robot on flat surfaces. Increased weight, however, was effective in providing greater stability on slopes.

As jumping is an effective method of moving over rough terrain, there is much interest in building robots that can jump. Deformation of a soft robot's body is an effective method to induce jumping. Our aim was to develop a jumping robot by deformation of a circular shell made of spring steel to result in the highest jump. Higher jumping requires enlargement of the contact area between the robot body and the floor. We developed a jumping mechanism that utilized a dish shape to maximize contact area.

This paper presents an investigation into the development of an augmented control scheme for vibration suppression and rigid body motion control of a twin rotor multi-input multi-output system (TRMS) in hovering mode. The augmented control scheme comprises feedforward and feedback control methods. A 4-impulse input shaper is designed and employed as a feedforward control method to pre-process the command signal applied to the system, based on the identified modes of vibration. It is then combined a hybrid feedback controller to form an augmented control scheme referred to as hybrid PD-type FLC and PID with 4-impulse input shaper (FPDPID+IS). The developed control strategy is designed and implemented within the real-time environment of the TRMS rig. Results of the response of the TRMS with the developed controller are presented in time and frequency domains. The performance of the proposed control scheme is assessed in terms of input tracking and level of vibration reduction. This is accomplished by comparing the system response with the developed controller to the one without control action in both open and closed loop configurations.

In this paper, we propose a new robotic catapult with a high stiffness endpoint. The conventional robotic catapults based on the closed elastica are the robotic elements for generating impulsive motions by utilizing the snap-through buckling. In a typical closed elastic catapult, the two ends of an elastic strip are fixed to a free joint and an active joint, respectively. Here we found that by adding only the high stiffness at the free joint, compared to the conventional type, more elastic energy can be stored and release surely without loss of a characteristic of generating impulsive motion repeatedly. By utilizing the high stiffness endpoint, we develop the faster impulsive swimming robot than the conventional one and perform experiments including start and changing direction motion.

In this paper, we propose a new robotic catapult based on the closed elastica with an anisotropic stiffness point which can generate impulsive motions for specific direction with higher repeated frequency than conventional one. Using proposed robotic catapult, we propose a compact jumping robot and try for realizing continuous jump of the robot.

The authors have proposed the robotic catapult based on the closed elastica as a robotic element for generating impulsive motions by using the snap-through buckling of an elastic material. In this paper, we show a quasi-static energy analysis of the robotic catapult based on the closed elastica. The quasi-static energy analysis with respect to the control variable allows us to see the relationship between mechanistic values and geometric ones of the closed elastica.

This paper presents the design and control of a new style overhead crane. The design approach is based on a modified traditional model of the overhead crane to allow the crane to move in three dimensions. Two overhead cranes acting in the same work space are designed to show the benefit of the proposed strategy. The modeling is done in the Visual Nastran (vn4d) software environment; and this is integrated with Matlab-Simulink for analysis and control purposes. The simulation results show that the proposed approach is effective with multiple cranes acting in the same work space.

A methodology is proposed to control the positioning of an overhead crane and the sway and residual oscillation of the associated payload. The MSC. Visual Nastran (Vn4D) is used to model the system where joints, forces and physical parameters such as dimensions and material type can be easily assigned. The model is integrated with Matlab-Simulink for analysis and control purposes. Investigations are carried out on mechanisms to attach the hook to the trolley for reduced the vibration in the payload. The simulation results show that good performance in the reduction of vibration and accurate positioning of the payload is achieved with a four ropes attachment mechanism.

In this paper genetic algorithm (GA) optimization of controller for a flexible manipulator is proposed. The dynamic model of the system is derived using the Lagrange equation and discretised using the finite difference (FD) method. GA optimization is used to tune parameters of proportional-integral-derivative (PID) controllers for the system. PID controllers are employed in the feedforward and feedback paths for control of rigid-body and flexible motion dynamics of the system. Simulation results of the response of the manipulator system with the developed controllers are presented and discussed.

This paper deploys the core methodologies of soft computing for the identification of 1 DOF hovering motion model of a twin rotor multi input-multi output system (TRMS). A TRMS is a scaled and simplified version of practical helicopter often used as a laboratory platform for control experiments. Although the dynamics of the TRMS are simpler than those of a real helicopter, they retain the most important helicopter features such as couplings and strong nonlinearities as well as it can be perceived as an unconventional and complex ‘air vehicle’. The fusion of artificial neural network and fuzzy logic through a adaptive neuro-fuzzy inference system (ANFIS) technique has embarked to a remarkable result in identification of the TRMS model. However, evolutionary computing using genetic algorithm with its inherent adaptation capabilities brings powerful optimization mechanisms for dynamic system modeling. The results and evidence from these meta-heuristics evolutionary algorithms are justified, presented and described.

In this article, A feedback control law for controlling the tip position of a flexible arm is proposed and evaluated. The objective of this research is to design a sensing antenna, a robot based on a single - link flexible arm which will enable us to locate a contact position with an object in order to detect the precise shape of that object. In a previous work, a nonlinear dynamic model was derived through the Lagrangian formulation where elastic characterics of the arm were modeled using the Euler - Bernoulli beam theory. Based on this model, an effective nonlinear controller was developed. Simulation results are given to show the controller's effectiveness.

This paper presents an investigation into the development of an augmented control scheme for vibration suppression and rigid body motion control of a twin rotor multi-input multi-output system (TRMS) in hovering mode. The augmented control scheme comprises feedforward and feedback control methods. A 4-impulse input shaper is designed and employed as a feedforward control method to pre-process the command signal applied to the system, based on the identified modes of vibration. It is then combined a hybrid feedback controller to form an augmented control scheme referred to as hybrid PD-type FLC and PID with 4-impulse input shaper (FPDPID+IS). The developed control strategy is designed and implemented within the simulation environment of the TRMS rig. Results of the response of the TRMS with the developed controller are presented in time and frequency domains. The performance of the proposed control scheme is assessed in terms of input tracking and level of vibration reduction. This is accomplished by comparing the system response with the developed controller to the one without control action in both open and closed loop configurations.

An output feedback model predictive control approach for constrained nonlinear systems is presented in this paper. The state variables are observed using an unscented Kalman filter that has some advantages over an extended Kalman filter. The nonlinear dynamic model of the system has been developed considering all possible effective elements. The nonlinear model is adaptively linearized during the prediction procedure. In order to improve the performance of the control system as many linearized models as prediction horizon are found at each real sample time. The optimum results of previous sample time are utilised for linearization of current sample time. Developing the equations to form a linear quadratic objective function with constraints is then carried out. A constrained highly nonlinear aerodynamic test rig, twin rotor MIMO system (TRMS) is selected to validate the performance and effectiveness of the proposed control approach.

This paper presents new methods for an intuitive human-robot interaction. On the one hand, a new explicit command interface is introduced that uses the robot's artificial skin as input modality. On the other hand, a novel implicit command interface for proactive execution of robot tasks is described, which is based on the estimation of the human intention. In order to accommodate these new human-robot interfaces, the robot architecture was augmented by a task planner and an execution supervisor.

This paper investigates the influence that the human factors of situation awareness, telepresence and workload have on task performance, as they have been identified as important factors for effective collaborations between humans and robots. The relations between the variables are studied in detail through a multiple linear regression model. The proposed novel measurement methods for each human factor are presented in general. All these issues were studied in a developed virtual urban search and rescue scenario with a teleoperated robot system.

This paper presents and compares two approaches for brain computer interface to steer a wheelchair, namely a new visual based P300 paradigm consisting of 8 arrows randomly intensified used for direction selection and a motor imagery paradigm for discrimination of three commands. Classification follows Bayesian and Fisher Linear Discriminant approaches both based on prior statistical knowledge.

Results in P300 paradigm reached false positive and false negative classification accuracies above 90%. Motor imagery experiments presented about 70% accuracy for left vs. right imagery and imagery vs. non-imagery.

This paper presents a new system to estimate the head pose of human in interactive indoor environment that has dynamic illumination change and large working space. The main idea of this system is to suggest a new morphological feature for estimating head angle from stereo disparity map. When a disparity map is obtained from stereo camera, the matching confidence value can be derived by measurements of correlation of the stereo images. Applying a threshold to the confidence value, we also obtain the specific morphology of the disparity map. Therefore, we can obtain the morphological shape of disparity map. Through the analysis of this morphological property, the head pose can be estimated. It is simple and fast algorithm in comparison with other algorithm which apply facial template, 2D, 3D models and optical flow method. Our system can automatically segment and estimate head pose in a wide range of head motion without manual initialization like other optical flow system. As the result of experiments, we obtained the reliable head orientation data under the real-time performance.

Teleoperation indicates operation of a machine at a distance. The goal of our presented project was to implement a face-to-face semi-contact robot fighting game that can be played by two remote human players over the Internet. Our paper focuses on the system design of the robot game, where the behaviour of each physical mobile robot is synchronized to a central virtual world.

EXOSTATION is a project aiming at developing a complete haptic control chain. The demonstrator is composed of a 7-DOF portable arm exoskeleton master, controlling an anthropomorphic slave robot interacting with its working environment. This system is meant to be used in a telemanipulation and telepresence modes (master/slave control), making it usable in many applications including space exploration missions (planetary surface exploration and surface habitat construction), medicine (medical acts training or guidance, CBRNE crisis management (interventions requiring precise remote manipulation or performed in unstructured environment), industrial environments (remote maintenance).

Over the last years researchers have put forth climbing robots for applications like cleaning and inspection of buildings, many composed of large glass facades or plain painted surfaces. Besides adhesion to the surface, dexterity and energy supply have turned out to be large challenges. This paper proposes an energy-autarkic and dexterous robot concept, able to cross frames and climb arbitrarily inclined smooth surfaces. To demonstrate this concept, the four-legged prototype CLAUS with kinematics of a simple four-legged walking robot is introduced. Passive suction cups on each foot provide a simple and robust method of adherence for climbing and avoid the constant energy consumption of active adherence systems. A simple release mechanism controls these suction cups. By reducing weight, cost and energy consumption of active adherence devices, the presented concept of the CLAUS robot comes one step closer to the ambition of low-cost, dexterous and energy-autarkic climbing robots. To illustrate this concept basic investigations are made using the CLAUS platform and considerations of control are discussed.

The paper proposes a wireless wall climbing robotic system for reconnaissance purpose. Firstly, we introduce the mechanism of the wall climbing robotic system. The wall-climbing robot of one single suction cup with four wheels locomotion system enables fast motion and can adapt nearly any kind of vertical wall surface in urban environment. Secondly, embedded control systems of wall climbing robotic system based on 485 BUS are designed. Thirdly, This paper analyzed the noise producing mechanism of the climbing robot using vacuum sucker, deduced the noise occurrence source and noise-power spectral density and a stable pressure generating radical fan of low noise was designed with the Computational Fluid Dynamics (CFD) tool. Experiments indicated that the noise of the robot is very low and reached 62 dB by testing in I meter distance while the robot was moving stably on tile wall.

New design of climbing cleaning robot for vertical surfaces is considered, Special crane is used to compensate force of gravity, It is shown, that the robot can suffer a vibroimpact regime. Powerful fun is used to prevent this regime. Experimental results with industrial prototype have confirmed that the new robot can work with reliability and high productivity.

New design of wheeled robot for climbing stairs is considered. It is shown that a combination of movable wheel axis and special wheel design enable the robot to use three modes of movements: the traditional mode for wheeled robot continuous movement, the regime of discontinuous movement upon the road with low coefficient of friction and the regime of movement up and down the stairs. Experimental results with prototype confirmed that the new robot can climb up and down the stairs.

A rapid expansion of wind turbine farms for sustainable electric power production is planned in Europe by 2020. At least in the UK, these will largely be located offshore to meet growing concerns about the visual intrusiveness and noise generation produced by onshore based farms. The necessary structural integrity inspection of offshore wind turbine blades poses tremendous problems of access, danger to human operatives and costs in the event of blades having to be taken out of service and transported on shore for schedules inspections. For these reasons robotic in-situ blade inspection of offshore wind turbines has been proposed and micro/nano focus computed axial X ray tomography (MNCAT) has been identified as the optimal if not the only solution for identification of safety critical defects in the thickest blade sections. The weight of such an inspection system is very high, typically 200kg and typical cross sectional scanner dimensions of 1 m × 2 m to encircle as blade, clearly involve very high destabilizing moments to be countered by the deployment robot. The solution is a climbing ring robot completely encircling a turbine tower, typically 3 meter in diameter, to provide the necessary adhesion forces and anti-destabilizing force moments. Because of the size and thus development costs of such a huge robot the optimal design path is to prototype a small scale model. First results on such a model are described and from its performance the load carrying capabilities of a full scale version can be computed and the scale model can then be refined by ‘reverse engineering’ to guarantee that a full scale construction is able to meet requirements. The key design innovation is that the adhesive forces between the robot and climbing surface a provided entirely by mechanical means rather than by using the usual methods of vacuum suction or magnetic force, making the system much cheaper and easier to manipulate. Furthermore the design is entirely modular.

In this paper the development of a robotic multi-axis crawler to accurately deploy inspection equipment on glass fiber plastic pipe and pipe joints is presented. The developed crawler is able to carry a range of sensors to automatically inspect complex geometry components, making traditional scanners redundant. First results show that the dexterity of the mechanized system is such that component coverage and positional accuracy are maximized using only a minimum number of degrees of freedom.

This paper describes the development of an automated NDT system deployed by a climbing robot for the inspection of vertical off ground weld structures of ship hull. The aim of this automated inspection system is to monitor quality of both cold and melt weld and detect melt weld pool flaws at early stage of the welding process and cooperate with a welding robot through a common intelligent system called Central Task Manager (CTM) so that it can be repaired before the weld solidifies. An on-board couplant supply system which normally applies in medical application and the design of a phased array probe holder with a linear actuator to avoid obstacles during the transition motion climbing from the ground on to the vertical surface is discussed. The central novel feature of the system is that complete and continuous 100% volume coverage of long weld line is achieved by combining electronic ultrasonic beam steerage with just a single axis mechanical scan devices from the linear motion of the NDT robot parallel to the weld. Experimental results obtained from a vertical section of the ship hull weld with the well visible defect are presented. The results focus on the NDT signal stability achieved during a continuous scan of a long weld line.

The inspection of complex environments is a challenging task for autonomous service robots. An autonomous inspection robot should actively examine entities of interest (EOIs), e.g. defects, and should take additional inspection actions if the data analysis results are uncertain. The selection of these actions should be driven by the assessment of the individual circumstances. In this paper a semantic approach for inspection planning, assessment of the data analysis results, decision making and replanning is investigated. For the experimental evaluation of this approach the detection and classification of waste on irregular terrains with the hexapod walking machine LAURON is chosen. First preliminary simulation results are presented.

The aim of this work is to develop a portable Non-Destructive Testing (NDT) robotic system that can be carried by CLAWAR to evaluate defects in geometrically complex surfaces and industrial products. In this paper, the difference in quality of defect data during manual inspection and automated inspection will be compared. Tests have been performed amongst others on complex shaped turbine blades using eddy currents. The challenge is to be able to follow the surface by keeping the NDT probe normal to the surface while maintaining a constant contact force with it. The approach to maintaining a constant contact force and angle to the object was to use a secure contact using permanent magnets and a force adapting control of the manipulator which resulted in improvements in the quality of inspection compared to manual inspection. Furthermore, a novel application of the position-force-moment (PFM) control was developed. Here, the robotic arm scans unknown contoured surfaces by keeping the sensor probe normal to the test surface, maintaining at the same time constant contact force, thereby ensuring good data acquisition.

This paper describes a solution for a mobile climbing robot which uses magnetic wheels for adhesion. The design has been created for the automatic nondestructive inspection in the surface of oil tanks, where direct access for a human operators is extremely expensive owing to the need for scaffolding and is highly dangerous due to the presence of a hostile environment.

Our main optimization criteria in this design were that it should not have the limitations of the umbilical power connection, that it should be as light as possible to reduce the actuators torque and to increase the battery life and capacity load, and that it should have a turning radius close to zero to improve maneuverability, adaptability to non-flat surfaces and a high surface upon which to install the ultrasonic sensor.

The proposed wall climbing robot with permanent magnetic adhesion mechanism is briefly described, including the details of mechanical system architecture in reversed tricycle along with the results of some test, on the magnetic wheels with twelve cylindrical permanent magnets.

The paper describes the development of an underwater teleoperated wall climbing robot that can carry another pipe crawler robot to nozzle openings inside a nuclear pressure vessel (PV). The wall climbing robot is positioned over each nozzle in the pressure vessel via teleoperation using visual feedback. The pipe crawler robot then transfers into the nozzle where it non-destructively tests circumferential welds located at a distance of 700mm from the inside walls of a pressure vessel. The results of trials with the robot in water tanks and a mock-up of the PV are presented. The paper describes the performance of the two underwater wall climbing robot prototypes, with emphasis on the design of a highly manoeuvrable robot and the performance of a special design of suction cups that have a low coefficient of friction and permit motion in cylindrical pressure vessels.

The paper presents the test results of a swimming and floor moving robot inspection system to test welds located inside a floating production storage and offloading oil tank (FPSO tank). Currently these welds are inspected manually by first emptying and cleaning the tank. This is a time consuming and expensive operation that requires operators to enter a hazardous environment. Significant cost reductions could be made by automating the inspection with robots that provide access to the welds. The simplest way to do this is to empty the tank so that only two to three centimeters of oil remain on the tank. A floor moving robot would then operate autonomously in the tank to follow and inspect the welds. A better solution is to perform the inspection in a full tank. In the first case the robot would operate in air and an explosive environment but would eliminate the need to swim the robot through a very complicated maze of partitioning walls and rows of strengthening plates that occur every 700-900 mm. In the latter case the robot would swim to a strengthening plate and operate under oil thereby eliminating the need to empty the tank. An amphibious mobile robot called FPSO is described which is capable of performing NDT in air and when submerged in liquids.

CLAWAR 2008 Special Session “Fast Biped Locomotion”:

One of the major issues in humanoid walking and running is to keep the trunk upright while the system is basically an unstable inverted pendulum. Here, we investigate trunk stability based on the bipedal spring-loaded inverted pendulum (SLIP) model. The proposed control strategy is to redirect the ground reaction force (GRF) to a point on the trunk located above the center of mass.

For keeping the trunk upright, no external sensors are required. In a perturbed situation, the proposed strategy leads to pendulum-like pitching motions. The model predicts a hip torque similar in shape and magnitude to that observed in human walking.

Robots of any kind, highly integrated mechatronic systems, are smart combinations of mechanics, electronics and information technology. The development of bipedal robots in particular, which perform human-like locomotion, challenges scientists on even higher levels.

Facing this challenge, this article presents a biomimetic bottom-up approach to use knowledge of biomechanical experiments on human walking and running, computer simulation and neuronal control concepts to sequentially design highly adaptable and compliant walking machines.

Fast dynamic biped walking also includes a vertical “hopping” component which demands provisions for take-off as well as for touchdown. To explore the conditions for stable hopping with minimal risk of damage, we designed – inspired by the previously described Marco robot - a hopper model with a leg consisting of two cascaded compliant elements. Optimization runs resulted in positive damping coefficients (i.e. negative velocity feedback) to be applied from touch down until midstance, but negative damping (i.e. positive velocity feedback) from mid stance to take off, while stiffness barely differed in both stance phases. Thereby, the energy management by positive and negative velocity feedback turned the hopper model functionally into a mass-spring assemble capable of both stable hopping and avoidance of hard impacts.

Two-legged locomotion is a much reseached topic in the robotics community since many decades. Nevertheless human walking and running is still unequaled. This paper introduces a biologically motivated approach of controlling bipeds that is based on recent results from neurological research on human walking. It features a hierarchical network of skills, motor patterns and reflexes that works locally and distributed and tries to exploit the natural dynamics of the system. The control concept is illustrated by the process of walking initiation.

This paper develops the results of works Refs. 1–4. Methods for designing of insectomorphic six-legged robot motion to overcome complicated obstacles by means of Coulomb friction are presented. Those obstacles are two identical lofty horizontal shelves connected by a narrow horizontal beam, a ladder leaned against a vertical wall of the shelf and a ball which can roll along a horizontal plane. The 3D computer simulation was fulfilled to demonstrate effectiveness and robustness of elaborated methods for obstacles overcoming.

In this paper the influence of knee joint on the motion of the robot and the effect of leg segmental proportions and angle between leg segments on the performance of the robot is explored via parametric analysis. The spring loaded inverted pendulum (SLIP) running in the sagittal plane, which is a simple model commonly used to analyze the basic qualitative properties of running, is used. The performance index is the cost of locomotion, which is given as the actuator power to sustain a certain motion, very close to a passive one, normalized to forward velocity per body length and robot weight. Simulation results show that the effect of leg configuration is great and that considerable variations exist on the value of the energetic cost of locomotion that makes some leg configurations more desirable than others.

The purpose of this study is to design a robust attitude control method of a six-legged robot. The walking robot is a system including many uncertainties like the modeling error, the perturbation of a parameter, and so on. These uncertain elements effected to the performance of the attitude control method by optimal servo system. Sliding mode control is a very robust control method for above uncertain elements. As a robust attitude control method, sliding mode control is applied. The validity of the proposed control method is verified by 3D simulations and experiments of the six-legged robot, in which each leg link is constructed by the worm gear.

This paper dealt with a walking control method of a six-legged robot consisting of one link without the signal of the angular sensor. The leg of the six-legged robot is semicircle arc shape. In details, the transfer function of the leg is derived from the frequency response, and the feedforward control input of the leg is obtained from the inverse function of the transfer function. The attitude of the robot was classified by eight types, and the generating method of FE control signal was proposed in four types where the deflection between the reference signal and the angle of the leg was small. The validity of the proposed control method was verified by experiments of the walking.

This work deals with a intuitive method for a planar biped to walk. A key feature of the proposed intuitive method is that feet of the robot are controlled to track a given trajectory, which is specially designed relative to the base body of the robot. The trajectory of feet is presumed from analysis of the walking motion of a human being. The author names the proposed scheme Relative Trajectory Control (RTC) method. A simple method to maintain a stable posture while the robot is walking is also introduced in RTC method. In this work, the biped is modeled as a free-floating robot, of which dynamic model is obtained in the Cartesian space. Using the obtained dynamic model, the robot is controlled by a model-based feedback control scheme. The author shows a preliminary experimental result to verify that the biped robot with the proposed RTC method can walk on surface, which may be irregular.

In this paper, we present the first results about dynamic obstacle avoidances using stepping over strategy for biped robots. The proposed approach is based on the Fuzzy Q-learning (FQL) concept. The originality of our strategy is that we consider it should be possible to design separately the high-level (path planning) control and the low-level control (gait pattern). The first results show the effectiveness of the proposed approach in the case of a flat dynamic obstacle.

This article describes a probabilistic control model based on Markov Decision Processes for the locomotion of a hexagonal hexapod robot. Uncertainty is naturally taken into account in probabilistic models, resulting in flexible control models that enable a robot to react to both, expected and unexpected situations. The model was tested using a simulated robot under various experimental conditions.

This paper describes the robot control for switching locomotion mode between leg-type and wheel-type adaptively toward forward and backward. Knee ahead and knee behind gaits are typical features discriminated with respect to the knee joint location from the hip joint toward the robot's traveling direction. In the mode transition simulation, we show results satisfying the direction of the hip joint rotation as a favorable phase to make continuous and smooth movement in the period of changing the mode from leg-type to wheel-type and vice versa. Also, by using the robot, PEOPLER, we demonstrate the control to verify that the selection of an appropriate gait is useful in practical application.

This paper presents the control and gait generation for the DLR-Crawler, an actively compliant hexapod based on the fingers of DLR-Hand II. The first gait generation method combines a fixed pattern with a leg extension reflex and an obstacle avoidance reflex. The second method employs leg coordination based on Cruse's rules and some reflexes to master uneven terrain. Both methods show smooth walking on even ground as well as on gravel.

This paper extends and modifies fast bipedal locomotion method for inclined floors. The fast bipedal locomotion method is a real-time method for generating the joint trajectories of a humanoid robot. It had been developed only for ideal flat floors so if the floor has a little inclination (2 or 3 degree) the robot motion will not be stable. Modification basically is done in the controlling part of the motion on inclined floors. The modified method is analyzed and verified by numerical simulation.

The biologically inspired hexapod walking machine LAURON has been developed to implement statically stable walking in rough terrain. The mechanics and the motion control have been modelled on the stick insect. LAURON autonomously perceives the environment and plans a path to a given target. Using its sensors LAURON can detect obstacles and avoid them while walking. As a walking machine, the robot can also climb over obstacles without damaging them, as it is the case with tracked or wheeled systems. This paper describes the new flexible behaviour based control of the six legged walking machine LAURON IVc using behaviour networks.

In robotics, gaits generation is a first step of locomotion control strategies. A particle swarm is used to generate the joints trajectories, such a proposal is an alternative to classical kinematics' modeling. Our proposal consists in a hybrid swarms placed in the skeleton-joints. The skeleton structure is directly inspired from a human-like locomotion system. A collaboration strategy is applied to extract efficient locomotion gaits ensuring the walker stability. The stability control uses a classical method based on the projection of the center of mass, COM, of the robot on the sustention polygon.

Urban Search and Rescue (USAR) involves locating, rescuing (extricating), and medically stabilizing victims trapped in confined spaces. In this paper we state the current approach to USAR, address the limitations and discuss the way for moving in rugged topography. To achieve objectives such as surveillance, reconnaissance, and rescue, it is necessary to develop a driving mechanism that can handle rugged geographical features. We propose a new type of driving mechanism for a rescue robot that has a variable shape single-track. By using a variety shapes, it can get the gain of steering and rotating and the ability to overcome stairs. In this paper, we analyzed the design parameters for making variable transform shapes and determined the specifications of the robot to enhance adaptability to stairs.

In this paper we develop a model of a Linear Force Actuator based actively articulated suspension vehicle with toroidal wheels and propose a strategy to control its contact forces to improve traction on uneven terrain in 3-D while maintaining a desired posture. We develop the quasi-static analysis for our vehicle and compare the maneuverability of our vehicle with that of a passive suspension system. Extensive uneven terrain simulations depict the efficacy of our proposed system.

Inspired by quadruped animals we developed the hybrid legged-wheeled robot ASGUARD. We showed already that this robot is able to cope with a variety of stairs, very rough terrain, and is able to move very fast on flat ground. We will describe a versatile adaptive control approach for such a system which is based only on proprioceptive data. An additional inclination and roll feedback is used to make the same controller more robust in terms of stair-climbing capabilities. At the same time, high velocities can be reached on flat ground without changing the configuration of the system. By using twenty compliant legs, which are mounted around four individually rotating hip-shafts, we abstract from the biological system. For the locomotion control we use an abstract model of bio-inspired Central Pattern Generators (CPG) which can be found in biological systems from humans to insects. In contrast to existing work, ASGUARD uses the sensed feedback of the environment to adapt the walking pattern in real time.

This paper concerns the control of an autonomous high mobility wheel-legged rover crossing uneven terrains. A new control strategy, using active redundancies of the robot, leads to elaborate a posture control based on the potential field approach of the stability measurement. Then a decoupled posture and trajectory control algorithm based on the velocity model of the robot is proposed. Last, simulation results showing performance of the control algorithm are presented.

A robot that has powered wheels suspended by active actuators or legs can utilize the wheels and legs independently or combined, in different locomotion modes. This paper describes criteria and corresponding logic that can be used to automatically determine the best locomotion mode in order to achieve best mobility according to the local terrain conditions.

Visual Servoing is one of the techniques that focus on more research projects in the robotics field due to the great number of unsolved problems. One of these unsolved issues is the reduced velocity of the visual servoing task, therefore this paper shows a high speed acquisition and processing image system that, combined with a powerful cartesian robot can be used to catch flying objects. To estimate the trajectory of the moving/flying object (a ball in this research), two Kalman filters with different cinematic models (constant velocity model for horizontal movement and constant acceleration model for vertical movement) are used. In this work, a great number of experiments are done to check the reliability and robustness of the configuration setup and algorithms presented.

This paper presents the teleoperation method of manipulators which have different kinematics with respect of the master robots using bilateral control by state convergence. This method makes a relation between the kinematics of the master and slave robot using a virtual robot. This method allows controlling manipulators which are a part of different kinds of robot as: climber robots, underwater robots, human robots, etc.

The grasp space for 2D and 3D discretized objects describes the possible grasps on the object that fulfill a desired property, namely the resistance to external disturbances or force-closure. The grasp space may be generated by a brute force search, however, a more efficient approach takes advantage of the information provided by a low number of samples of the space that allow the construction of the Independent Contact Regions (ICRs) and Non-Graspable Regions (NGRs). This paper presents the algorithm for the structured exploration of the grasp space, and compares the performance of two sampling strategies in terms of coverage and time.

A Neural Network force control of a robot manipulator collaborating with a human to handle a fabric is presented. A robotic gripper holds the one end of a piece of fabric while a human hand holds the opposite end. As the human moves the fabric arbitrarily, the robot tries to manipulate the fabric according to the required handling task. The task could be defined in a higher-level decision making process which is based on the artificial constrains of the handling task and influenced by the human needs. A force sensor mounted on the wrist of the robot manipulator measures the 3-components (F_x, F_y, F_z) of the actual force applied by the human hand to the fabric, and the formulated force errors are used in the backpropagation algorithm, which trains the Neural Network force controller. The controller is tested in a simple case of a desired handling task and the results are discussed and compared with the reversed case, i.e. the robot moves the fabric arbitrarily and the human tries to manipulate the fabric according to the handling task. The response of the controller show that the robot is capable to handle the fabric according to the desired constrains.

A subset of grasps useful for service robots were identified and a low-cost robotic hand was designed and assembled to execute those. Dynamic simulation allows control algorithms to be efficiently developed and enables testing also in simulation. The performance of the actual hand is evaluated. The bill of materials, assembly instructions, drawings, and CAD-files are all available on the Internet (www.md.kth.se/kthand) to facilitate for those wishing to build their own sub € 1000 robot hand for service robotics research and development.

(Invited special session: Manipulation and Gripping) The present work shows recent results in the robotic manipulation context, after suitable integration between computer vision and radio frequency identification (RFID). The experimental validation has been carried out in a household environment, by using current manufactured objects, labeled previously, and processed through a RFID reader situated in the robot hand. A camera-in-hand also was used for segmentation and localization of the corresponding objects. The main problems solved in this work are those related with hardware integration between the robot and the RFID system, including the RFID reader, antenna and necessary interfaces. Furthermore, a graphical user interface was developed for displaying the results from visual processing, including both, segmentation and localization of the objects in the scene, visualizing at the same time the corresponding RFID labels associated which each previously identified object. These preliminary results have demonstrated the feasibility and reliability of this system to succeed with localization, identification and handling of daily objects, without predefined models at all, in household environments.

The paper investigates the prospects of extending flexible automation and intelligent manufacturing options in the field of textile and clothing industries. In particular the paper focuses on the automatic gripping and handling of limp and soft materials that is a needed step to extend automation beyond the fabric parts cutting by exploiting options of robotics, already in use for most of the manufacturing of automotive electric, electronic and mechanical goods. The manufacturing section here considered in clothes manufacturing is the cutting room and the task is picking up a piece of fabric from the cutting table and forward it single out, flat in a well defied pose to the following manufacturing sections. The robotic system purposely developed is presented: it includes: - a metamorphic robotic gripper to carry out automatically the cut part picking tasks and giving back the part hung; - a reconfigurable passive hanger. The system architecture and main components are presented.

This paper presents a multimodal approach to the problem of planning and performing a reliable grasping action on unmodeled objects. The proposed system consists of three main components. The first, fundamental component is a grasping architecture based on primitives rather than on contact point selection. The second component is a visual processing module that uses stereo images and biologically-inspired algorithms to accurately estimate pose, size and shape of a target object. A grasp action is planned and executed by the third component of the system, a reactive controller that uses tactile feedback to compensate possible inaccuracies and thus complete the grasp even in difficult or unexpected conditions.

The superiority of deformable human fingertips as compared to hard robot gripper fingers for grasping and manipulation has led to a number of investigations with robot hands employing elastomers or materials such as fluids or powders beneath a membrane at the fingertips. It is interesting to study the phenomenon of contact interaction with an object and its manipulation. In this paper bond graph modeling is used to model the contact between a grasped object and two soft fingertips. Detailed bond graph modeling (BGM) of the contact phenomenon with two soft-finger contacts which are placed against each other on the opposite sides of the grasped object as is generally the case in a manufacturing environment is presented. The stability of the object is determined which includes friction between the soft finger contact surfaces and the object, the stiffness of the springs is exploited while achieving the stability in the soft-grasping. The weight of the object coming downward is controlled by the friction between the fingers and the object during the application of contact forces by varying the damping and the stiffness in the soft finger.

This paper discusses an approach to determine the size of a soft-finger contact required to hold and manipulate an object with two soft fingers. The grasp is based on the equivalent stiffness resorting to the soft finger contact to balance the external forces. In this study, the soft-finger is modeled as elastic elements and placed at the boundary of equivalent contact patch circle. The size of the soft contact is determined based on the variables that include the dimensions of the grasped object, externally applied load and couples. The stiffnesses of the elastic elements are then used to model the soft-finger and the fine displacement resulting from the line springs under the application of preloads.

Manipulation of flesh, lies at the core of all surgical procedures, both as the objective of the procedure, but also as a means of gathering necessary knowledge to support the procedure. This is also true of the relatively simple procedure of drawing a blood-sample. In this paper, we present our progress to locate, develop and validate tactile sensing methods, that could enable fully automated blood-sampling, in order to support the rapidly growing demand in modern health-care systems.

In this article dynamics of vibration driven mobile micro-robots are developed. During locomotion these robots use ideas of periodical motion of internal masses. In previous papers the motion of robots without separation from supporting surface was developed. But in this paper hopping robot is presented. Such kind of robots is especially useful for applications where small devices with high speed are needed. It could be monitoring of relatively big terrain with no smooth surface. Considering robot moves due to rotation of internal mass inside of robot body. Simplest mathematical model of the robot takes into consideration only two rigid bodies moving relatively each other. Robot body contacts periodically with rigid supporting surface. The flight regimes and landing of robot on the supporting surface are described. Such approach allows defining parameters of robot providing different regimes of motion of considering system.

An interesting novel application of the Gough-Stewart platform as a climbing robot and its kinematics control has been proposed to climb autonomously through long structures describing unknown spatial trajectories, such as palm trunks, tubes, etc. For planning the motion of the parallel robot, inverse and direct kinematics problems have to be solved continuously in the path planning algorithm in a minimum time. Computation efficiency of the model is very important. This paper presents a comparison between two models of the 6-UPS parallel mechanism. Inverse and direct kinematic problems have been numerically solved with classic methods and compare for four different configurations for the two models. The analysis and simulation of the kinematics problems show the computational efficiency of the proposed model for the path planning of the climbing parallel robot.

The paper gives a description of the kinematics and the mechatronics design of RobuDOG, a four-legged robot developed for supporting research and education in robotics. Several aspects related to control and programming of this platform are considered as its inverse kinematic model, its velocity-based control and its dynamic behaviour analysis.

This work presents a new improved six legged mobile robot “ANTON”. Its mechanical structure, sensor system and control system are discussed in details. The hierarchically and modular build control algorithms are presented. The Software-in-the-Loop and Rapid Control Prototyping frameworks, which are used by robot development, are presented. An industrial Ethernet-based real-time communication protocol is introduced and the communication ability between the robot-side hardware and PC-side control system is investigated.

Collision and contact are complex interactions vital to the analysis of several research fields. In many models, collision and contact are unable to be described by a single equation because the change in energy loss is related to the relative collision velocity of the two objects. In this paper, an improvement is made to an existing nonlinear model to increase the realism of simulations of collision and contact. The new formulation removes the inconsistencies and numerical instability of previous treatments while maintaining the computational simplicity of the original model. Given a set of coefficient of restitution data, the new model simulates realistic collision behavior across a wide range of velocities. The new model is compared to two previous formulations in their ability to simulate experimental data from a croquet ball colliding with a croquet mallet in the normal direction.

The bipedal locomotion is performed by means of rhythmic and synchronized motions. These movements have a pattern produced by nervous networks in the spinal marrow. These specialized systems are known as central pattern generators (CPGs). Oscillators can be used to create similar systems to the human CPG, providing approximate patterns of locomotion. The objective of this work is to present a nonlinear oscillators system used to generate patterns of bipedal locomotion, taking into consideration a 2D model, with three determinants of human gait, which performs parallel movements to the sagittal plane. Using this system, the behavior of the hip and knee angles was determined. Modification of the step length and gait frequency can be obtained from change of few parameters in the oscillators. An analysis of the system provides excellent results when compared to experimental analyses. Based on these results, we conclude that the use of nonlinear oscillators can represent an excellent method for creation of pattern generators, allowing their application to simulate a CPG of bipedal locomotion.

The 3.0 or 3D internet, exemplified by environments such as LindenLab's SecondLife, supplies a massively multiagent simulation environment, whose capability of simulating, for example the multi body dynamics, is so far quite limited. Nethertheless it is possible to do some preliminary simulations of the collective behaviours of multirobot systems, in a mixed environment where some of the agents are directly controlled by a human being.

In this paper we discuss the limits and future opportunities of Massive Multiplayer Game technologies applied to Internet 3.0 for the simulation of massively multi robot systems. In particular in the cooperative behaviour analysis of bioinspired locomotion systems. Some reasoned guesses on the time frame of this developments are shared and a few examples are given.

One application of the sensor head designed for terrain scanning in humanitarian demining tasks in the DYLEMA project is presented, where the lateral distance to obstacles measured by a network of lateral range sensors is converted into a virtual contact force, which in turn is feed as input for a contact force control loop. The sensor head sweep movement is modified when an obstacle is detected (or “touched”) also helping to detect the position of the obstacle's contour.

Optical sensors supply by far the most information and as greater and greater processing capabilities become readily available their use becomes more widespread. Many researchers and companies have made more or less successful attempts at creating optical sensors for speed measurement, however there are still many questions open for research. The aim of this article is to introduce a novel method for optical speed measurement and put it into perspective by summarizing and reviewing recent related work. Practical considerations on texture analysis and sensor parameters are discussed backed up with simulation results.

Simple optoelectronic exteroceptive sensor for controlling and learning the dynamical equilibrium of a walking robot is presented. Sensor consists of the digital camera and structured light source, for example laser diode module with Diffractive Optical Element. Digital camera captures structured configuration of light spots projected on a surface in front of a robot. There are two variants, the light source position is fixed to the digital camera and there is no reference object at scene. In these case only medial and lateral tilt to plane in front of the robot is estimated and second variant when only light source of the robot is fixed to any part of robot and camera is placed somewhere else. In these case multi-degrees-of-freedom information can be estimated. The image information from the digital camera is input for the control of a dynamical equilibrium of a walking robot.

Service Robots are intended to support people in daily activities (complex environments), and, for this reason, human-robot communication is an important subject. This work is based on the fact that humans noticeably exploit motion information from other humans' actions. This paper presents a new approach of analyzing Human Robot Interaction based on robust optical flow techniques to calculate motion information using a catadioptric system implemented to enhance the Field of Vision. This allows capturing much more information for the interpretation procedure that employs a hierarchical fuzzy making decision technique designed for the learning model.

This paper focuses on current state of the art in sensory equipment for walking robotic systems and walking humanoids oriented on assistive technologies, military, security and rescue robotic systems. Described sensory systems are based on modular design that measures axial shiftings and angular displacements is a part of many sensory systems in robotics and human-machine interface applications. This is done by means of a square or annular CCD, or a set of four linear PSD elements, and four laser diode rays or planes, creating the shape of a pyramid. The positions of four light spots on the CCD or eight light spots on the PSDs are processed to produce three axial shiftings and three angular displacement values.

This paper describes a Smart Transducer, inspired by the IEEE 1451 Standards, for olfactory sensing. The device is composed by five different types of gas sensors, three anemometers and one temperature and humidity sensor. This system senses the environmental conditions, process the acquired information and send it to a Khepera III mobile robot through an I²C bus. An overview on the implemented real time software and data structures associated with each sensing node will be made.

Using quadriceps only to obtain smooth FES-cycling reduces the number of stimulated electrodes applied to a paraplegic. Therefore, the preparation process for FES-cycling become more comfortable and less time consuming. This paper discusses how elastic cables can be used to assist quadriceps FES-cycling to eliminate the dead points of the pedal cycle. Implementation of an energy storage device (elastic cable) has many desirable influences on FES-cycling, such as decreasing the number of stimulated muscles. This in turn reduces the number of FES surface electrodes, thereby saving excess energy that is released during the cycling. This subsequently reduces stimulation intensity and duration and minimizes muscle fatigue.

The goal of this study is to investigate torque profiles of each active joint in sit-to- stand (SiSt) mode arm free and with seesaw exercise machine. This paper considers the development of a dynamic humanoid model for SiSt demonstration. It consists of two main parts, the development of human model using Visual Nastran and the development of a controller for controlled movement of the system. A closed loop feedback finite-state PID control strategy is used. In order to reduce torque amplitude profile on each joint, a seesaw mechanism is realised to reduce the upper body load on lower extremity. In this method only knee joint torque in the range of stimulation muscle torque (50Nm) was adequate to perform sit to stand without relying on hip and ankle joints.

This paper proposed a new gravity compensation system which is suitable for the lower extremity rehabilitation therapy. The proposed system can compensate the gravitational moments exerting on the leg joints perfectly using the passive mechanical elements. In contrast with the previous systems, the proposed gravity compensation mechanisms are wholly embedded in the link body, so that the system is safer and well-suited to the wearable equipment. The feasibility of the gravity proposed mechanisms is confirmed through the experiments using an actual equipment. The effectiveness of the system as a lower extremity rehabilitation equipment is examined through computer simulations for the bending and stretching exercises.

This paper proposes a robotic walker system with standing, walking and seating assistance function. Our system focuses on domestic use for aged person who needs nursing in their daily life. Our key ideas are two topics. The first topic is combination of standing assistance function and walking assistance function. In previous works, many assistance devices are specialized in only “standing-up operation” or “walking operation”. However, in their daily life, elderly person needs standing, walking and seating assistance continuously by a same device. Therefore, our developing assistance system can support both operations by a small sized mechanism which is easy to use in the home. The second topic is a seating position adjustment assistance. From questionnaires of nursing specialists, a seating position adjustment requires the elderly to walk backward and it is difficult operation for them. Furthermore, in many cases, a failure of this operation causes a fracture which has high risk to fall into bedridden life. Thus, our developing system can assist the elderly to adjust the seating position safety. The performance of our proposed system is verified by experiments using our prototype.

This paper presents a development of a four-wheel drive (4WD) omnidirectional wheelchair with an advanced step climbing capability. For enhancing the mobility of standard wheelchairs, a new type of omnidirectional mobile platform with a 4WD mechanism is proposed. The 4WD system provides enhanced step climb capability however, chair inclination becomes large when either front or rear wheels are on a step because a wheelbase of the developed wheelchair is almost same as a standard one. This large inclination not only gives a high risk for falling but also brakes a load distribution condition between front and rear wheels which is required for climbing the step. To solve these problems relating to the chair inclination, we develop a chair tilting system for the 4WD omnidirectional wheelchair including a linear drive and a variable center of rotation mechanism and an inclination sensor.

This paper discusses the steering control for a wheelchair on two wheels. A novel fuzzy logic control (FLC) system is designed and implemented for this highly nonlinear system. The wheelchair system is modeled in Visual Nastran software environment as a plant and controlled with the developed FLC in Matlab/Simulink environment. The steering motion takes place after lifting and stabilizing has been achieved. There are noticeable challenges in the modeling of the wheelchair where limited actuators are used for different functions, thus suitable controllers need to be developed. Results show that the FLC strategy works very well and gives good system performance.

Although shield tunnel construction and tunnel boring machines have developed greatly, these machines are still large in size and consume large amounts of energy. A robot small enough to be able to explore beneath the ground on its own would extend the range of underground investigation both under the Earth's surface as well as on the moon in the future. This study focuses on peristaltic crawling of earthworms as a locomotion mechanism for an underground explorer robot. In peristaltic crawling, extension and contraction waves are propagated in the anteroposterior direction by varying the thickness and length of the earthworm's segments, and a large surface area is brought into contact during motion. Furthermore, it requires no more space than that of an excavation part on the anterior of the robot. The proposed robot consists of several parallel link mechanisms. The robot could move on a plane surface and in vertical perforated dirt, and could climb a tube. The robot showed good performance in the experiments.

As a recent trend in robotic research, legged robots gained importance since they are capable of performing vast maneuver abilities on rough terrain. Especially in case of planetary exploration, legged robots seem to offer the most suitable mechanism in order to accomplish any given mission. Therefore, this paper is aimed at introducing the development of a two segmental, eight-legged mobile robot for planetary exploration, which has a novel mechanical structure and a hierarchical control system. Its mechanical and electrical hardware design is explained in detail. In conclusion, it is anticipated that the designed robot can be employed as an outdoor explorer robot.

This paper presents a new visual odometry approach for mobile robot self-localization utilizing natural circular invariant features during motion. It is proposed that the on-board camera acquires sequences of overlapping mages and senses the distance and orientation of the vehicle with respect to identified markers. The paper uses an effective image filtering technique based on convolution that can be used to localize the natural markers in the images. The proposed approach simplifies the problem of feature localization and allows a robust estimation of the vehicle's trajectory. Initial experiments are carried out on a mobile robot and results are presented.

In earlier articles we had developed a formation control method based on single view depth estimation. In this paper, we implement that strategy on a robotic swarm composed of non-holonomic agents using the physics based Webots robot simulator. First, we review the single view depth estimation based scheme and the distributed control laws for the agents. Then, we discuss the related modifications due to the non-holonomic constraints of the agents and some related implementation issues and present some simulation results obtained using the proposed control scheme. The scheme is based completely on local information implying that neither global position information nor communication between robots are needed for the implementation of the algorithm.

This paper studies kinematics and dynamics of a reconfigurable modular robot consisting of ten modules. Motion planning for 3 different configurations; snake-like, inch-worm like and loop like robot are studied in detail. Actuating motors selection and mechanical design are done based on the simulation results. Designed trajectories are implemented in real open-loop control experiments for all configurations. The experiments show very satisfying results.

This paper presents an analysis of a differential mobile robot with an arbitrary position and orientation (x,y,θ) for best approach to a pose vector goal (0,0,0°). Although the control law implemented is a feedback motion controller, which is relatively simple and does not satisfy Brockett's theorem, the results show that this control algorithm can be tuned to develop solutions to a differential drive mobile robot's motion control problem. Various sets of control parameters are tested on the mobile robot and a set of control parameters are obtained for the system that requires fewer turns while approaching the origin of the inertia frame coordinates.

An understanding of track-terrain interaction dynamics and vehicle slip is crucial for skid-steered tracked vehicles traversing over soft terrain. There is a lack of experimental data for validating dynamic, kinematic and control models developed for tracked vehicles on soft terrains. The objective of this paper is to develop a test rig that will generate experimental data for autonomous tracked vehicles following a steady state circular trajectory on soft terrains. The data will be used in the future to validate a traversability model for predicting track thrusts, a visual odometry technique for predicting vehicle slip and in controlling autonomous tracked vehicle following a steady state circular trajectory on soft terrains that were developed at King's College London.

3DCLIMBER is a pole climbing robot which was developed at ISR-UC. It can climb from 3D structures with bents and branches and scan whole area of the structure. Upon the completion of the mechanical structure some tests were conducted for proof of concept. During this test the errors which might happen were revealed and then a sensorial architecture were designed in order to measure the errors and a control architecture was proposed in order to compensate them. This paper discusses the source of errors and the approach for compensation of the errors and also the Mechatronics and the control architecture of the 3DCLIMBER.

This paper describes a riding robot system like as a horse developed for healthcare and entertainment applications. The riding motion of this robot named by RideBot is similar to riding a horse for human. The developed RideBot can to follow the intention of horseman using by the rein and spur mechanism and to simulate the walking and running motion using by the saddle mechanism of 3 DOF. And also this RideBot have the bio-handle mechanism which is to check the horseman's bio-signals include blood pressure, pulse and heartbeat for the health care services of users. In order to evaluate the performance of RideBot, we carried out the experiments on the several riding motions as follows: the slow walking, the fast walking, and the turning of direction with horseman.

The interest in the development of climbing robots is growing rapidly. Motivations are typically to increase the operation efficiency by obviating the costly assembly of scaffolding or to protect human health and safety in hazardous tasks. Climbing robots are starting to be developed for applications ranging from cleaning to inspection of difficult to reach constructions. These robots should be capable of travelling on different types of surfaces, with varying inclinations, such as floors, walls, ceilings, and to walk between such surfaces. Furthermore, these machines should be capable of adapting and reconfiguring for various environment conditions and to be self-contained. Regarding the adhesion to the surface, they should be able to produce a secure gripping force using a light-weight mechanism. This paper presents a survey of different applications and technologies proposed for the implementation of climbing robots.

UNIFIER describes the concept of a robotic system to service solar power plants both on the ground and on inclined surfaces. They usually consist of several seperated generators and modules. While wall climbing robots are able to move on the solar modules they are not able to cross the large distances between the generators. An additional robot carries the robot from one generator to another and supplies it with electricity and cleaning agent. The wall climbing robot becomes a satellite robot of its carrier.

This paper proposes a novel search and rescue concept that aims to overcome the most basic obstacle in utilising a search and rescue teleoperated robot for a long distance - energy autonomy. The concept utilizes a number of small robots capable of creating an energy supply chain that extends, to a degree based on the requirements of the search area. In this present work, the collaborative group of robots' predominant task is to maintain a constant supply of energy to the leading robot. Feasibility of the energy transfer and ‘energy cost’ have been simulated which consequently produced a mathematical description of the cost function. The results presented are obtained using a set of identical robots capable of conveying energy from the last deployed all the way to the leading robot. A robot single-line formation is important as it maps out the most frugal energy supply line. The methodology used was to simulate both methods of energy transfer, robots charging each other and robots exchanging batteries. The first method was implemented using MARCO2 (an in house built robot) to test the validity of the computer simulation and to study the effect of inaccurate localisation on the system.

This paper deals with the tracking and following of a person with a camera mounted mobile robot. A modified energy based optical flow approach is used for motion segmentation from a pair of images. Further a spatial relative veolcity based filering is used to extract prominently moving objects. Depth and color information are also used to robustly identify and follow a person.

This paper presents an LQR controller approach for the simulation and controls of an affordable commercial humanoid robot doing a handstand on a high bar, by considering it as an underactuated 3-link inverted pendulum with off-centered masses. The developments presented include: i) the software development for interfacing the Matlab® Real Time Workshop Toolbox® with the humanoid robot servos; ii) the identification of the internal and external dynamic parameter of the humanoid servos and structure, respectively; iii) the dynamics modeling and simulation of the humanoid robot; iv) the deduction of the equations of motion for an underactuated n-link inverted pendulum. Simulation results proved the adequacy of LQR controller.

We will present different approaches to generating efficient multi-robot cooperative strategies. We will focus on learning policies for multiple robots, including humanoid robots, by demonstration from a human. The robots incrementally learn the policy by requesting further demonstration based on their confidence on successful execution. The humans can provide additional guidance as more demonstration in terms of examples, functional guidance or corrections. The work presented is in conjunction with several of my students, in particular Sonia Chernova and Brenna Argall.

No abstract received.

This paper presents the development and gait generation of biped robot Stepper-Senior. Parallel double crank mechanism and elastic materials are introduced in the 10 DOF lower limbs to mechanically restrict the sole to contact flat with the ground for stable fast walking. Virtual Slope Walking is used for biped gait generation, which is a simple method with strongly intuitive parameters for real-time utilization. In walking experiment, Stepper-Senior reaches the speed of 0.65 m/s and accomplishes omnidirectional walking.

Planning and control of humanoid biped walking have been an active research topic for many years. But, there is no definite answer to the question of how to practically achieve speedy and stable walking in real-time and in a changing environment. In this paper, we first re-examine the issue of planning and controlling humanoid biped walking. Then, we propose two new ideas. The first idea is to treat the supporting foot of a biped to be part of the ground. In this way, there is a foot reaction force acting at a fixed virtual joint, which can be at, or below, the ankle joint. And, we come out a new concept, which is named as in-foot ZMP in contrast to the existing concept of on-ground ZMP. The unique benefit with this new concept of in-foot ZMP is that the ZMP control is no more an issue because the in-foot ZMP can be controlled so as to to be at a fixed virtual joint during a stable walking. And, such a fixed virtual joint can be called a ZMP joint. The second idea is to focus on hip's trajectory (instead of on-ground ZMP's trajectory) and to split a hip's dynamic response into two independent parts: one is the steady-state response contributing to the stability of walking (or standing), and the other is the transient response contributing to the speed of walking. This idea allows us to explicitly postulate the necessary and sufficient condition for achieving leg stability as well as the necessary and sufficient condition for achieving foot stability. We will show that the implementation of these two new ideas help realize a unified framework for task-guided, intention-guided, and sensor-guided, planning and control of humanoid biped walking.

This work deals with the multibody dynamics modelling of the humanoid robots. An alternative geometric method will be presented based in the screw theory. The obtained torques are used for selecting the right motors. Otherwise, this computation allows to compute the joint torques at any humanoid robot motion, those are constrained for the physical robot limits. Furthermore, the joint torque references could be obtained for doing the torque control loop and develop any motion pattern. The simulation results in order to check the joint torques are shown and discussed.

This paper deals with collaboration of human with humanoid robot in common collaborative tasks. The papers is focusing in the motion planning for transportation tasks. The humanoid's motion planning is described from approaching to the objective until carrying an object with human cooperation. In order to obtained the overall control, ZMP and Generalized ZMP (GZMP) concepts had been taking into account. The simulation results of the system have been presented.

This paper proposes an online walking pattern generation approach using convex optimization. With an inverted pendulum model based zero moment point dynamics formulation in the framework of Lie group, the convex optimization is introduced to revise the reference trajectory of joints which is generated by the cart-table model and the inverse kinematics to achieve dynamically-stable gait. The optimal objectives, that minimize the desired ZMP tracking error and joint trajectories' tracking error, and the constraints for single support phase of dynamic walking are discussed and formulated into the form of second order cone programming program. The effectiveness of the proposed method is verified by simulation experiments which show the iterations for each convex optimization is no more than 15.

This paper reports on the importance of understanding human walking biomechanics for the design of new robotic and/or prosthetic feet. Based on human ankle behavior, the design specifications for new ankle-foot systems are determined. Three electrically powered ankle-foot design concepts are described.

This paper describes development of a humanoid robot which walks, crawls, and tolerate obstacles in daily environment. The robot has thick soft polyurethane covers and mechanical torque limiters so that it tolerate impulsive and static overload. At the latter part of the paper, tolerability of the robot was evaluated.

This paper presents a novel retargeting method, which is included in an interface that allows a social robot to imitate the gestures performed by a human demonstrator. The input for this interface are human pose data obtained using any motion capture system. A general human 3D model adopts the perceived pose, and then this pose is retargeted to the particular robotic platform being used. The retargeting module combines two different strategies to translate data from human to robot. Experimental results show that this combined approach is able to preserve the characteristics of both static and dynamic gestures in different scenarios. The system has been tested over two different robotic platforms: the Fujitsu HOAP-1 humanoid robot and the NOMADA, a new social robot that is currently being developed in our research group.

This paper describes a humanoid robot simulator supporting joint trajectory optimization, following accurately the real robot characteristics. The simulator, based on a rigid body simulator (Open Dynamics Engine) and an OpenGL based graphics library (GLScene), provides instant visual feedback and realistic dynamics. It allows to design and test behaviours and control methods without access to the real hardware, preventing damages in the real robot in the earlier stages of development. Having in mind the energy consumption minimization, the low level trajectory planning is discussed and experimental results are presented. The proposed methods are shown to minimize the total energy assuming two intervals of constant acceleration.

This paper presents a unified formulation of realtime Zero Moment Point(ZMP) detection using 6-axis force/torque sensor in the framework of Lie group aiming to achieve dynamically-stable gait for a humanoid robot. The proposed approach is able to detect ZMP in both single-support and double-support phases for humanoid walk. Experiment is conducted to verify the effectiveness of the proposed Lie group formulation for realtime ZMP detection for balance control and dynamic walking control of humanoid robot.

This paper describes the hardware and software setups that allow a humanoid robot developed from scratch to perform visual tracking based exclusively on onboard components. The robot which was started on earlier projects was finally given its full autonomy in what concerns perception and vision capabilities. An embedded PCI04-based controller running Linux is now able to interface a IEEE1394 color camera and, using the OpenCV library, can now perform visual tracking of some objects moving on its neighborhood. This embedded controller, besides being responsible for image acquisition and processing, serves as an interface between external monitoring and the distributed control architecture based on a master-multi-slave CAN bus of local controllers for joint actuation and sensor monitoring.

This paper describes an approach for real-time preparation of grasping tasks, based on the low-order moments of the target's shape on a stereo pair of images acquired by an active vision head. The objective is to estimate the 3D position and orientation of an object and of the robotic hand, by using computationally fast and independent software components. These measurements are then used for the two phases of a reaching task: (i) an initial phase whereby the robot positions its hand close to the target with an appropriate hand orientation, and (ii) a final phase where a precise hand-to-target positioning is performed using Position-Based Visual Servoing methods.

AUTHOR INDEX