International Journal of Humanoid Robotics

This article reviews R&D activities on automated highway systems (AHSs) since 1950s with emphasis on recent ones. AHSs in the 1950s and 1960s were featured by a cooperative system between roads and vehicles; those in the 1970s and 1980s were by autonomous systems. Intelligent transport system (ITS) projects since 1980s regarded AHS as an important system, not only a single automated vehicle but also an automated platoon were developed. In recent years, the R&D focuses on automated transit buses aiming at precision docking and driving along a narrow lane and automated truck platoons aiming at energy saving. The automated driving can be common ground of automobiles and robots.

A walking robot is considered. For leg lightness, it is suggested in some cases to miniaturize its cross section. At that buckling of the leg rectilinear form is possible under dead weight. To keep the rectilinear form, it is proposed to place a vibroexciter for vertical oscillations in the bottom of the leg. It results in parametrical oscillations that make onset for the effect of vibrational stabilization of the rectilinear form. This effect can be treated as an increase of flexural stiffness of the flexible rod. Scheme of the leg's model in the form of flexible rod with distributed mass and concentrated mass at its free end is developed. Taking into account the finite rotations of its sections, we derive the equation of oscillations for the rod. It is shown that the effect of rectilinear form stabilization occurs under certain magnitudes of vibration frequency and amplitude. Stability conditions for obtained solutions are analyzed.

A new model-free approach to precisely control humanoid robot joints is presented in this article. An input–output online identification procedure will permit to compensate neglected or uncertain dynamics, such as, on the one hand, transmission and compliance nonlinear effects, and, on the other hand, network transmission delays. Robustness to parameter variations will be analyzed and compared to other advanced PID-based controllers. Simulations will show that not only good tracking quality can be obtained with this novel technique, but also that it provides a very robust behavior to the closed-loop system. Furthermore, a locomotion task will be tested in a complete humanoid simulator to highlight the suitability of this control approach for such complex systems.

Due to the advantages of more intensiveness, long duration, repeatability, and task-orientation, robot-assistant training has become a promising technology in stroke rehabilitation. Regarding the upper extremity, the natural coordination called shoulder rhythm is the most challenge to the design of ergonomic shoulder exoskeleton. Based on kinematic analysis of movement of shoulder complex, a 10-degree-of-freedom (DoF) exoskeleton rehabilitation robot with six-DoF shoulder actuation mechanism driven by pneumatic muscle through Bowden cable transmitting force is proposed. The kinematic relationship between shoulder girdle motion and the humerus flexion/retroflexion and abduction/adduction was described. The compact mechanisms for cable tension and cable disconnect/connect respectively were proposed to realize the cable automatic tension and drive-unit/manipulating-unit detachment. In order to verify the manipulability of the proposed robot during assisting patient with performing activities of daily living (ADLs), the performance criteria, i.e., normalized dexterity measure and manipulability ellipsoid, are used to evaluate and compare with human upper extremity. The evaluated results confirm the ergonomic design of shoulder mechanism of the rehabilitation robot that can provide approximate dexterity of human upper extremity in ADLs.

Gesture-changeable under-actuated (GCUA) function is put forward to make traditional under-actuated hands feel easy to grasp different objects and do simple operations dexterously, simultaneously, this function can lower control difficulties of robotic hands. Based on GCUA function, a GCUA hand based on pulley-belt mechanism is designed in detail and manufactured. The Hand can grasp different objects self-adaptively and change its initial gesture dexterously before touching objects. The hand has 5 fingers and 15 DOFs, each finger utilizes screw-nut transmission, flexible drawstring constraint and belt-pulley under-actuated mechanism to realize GCUA function. The analyses on grasping static forces and grasping stabilities are given. The analyses and experimental results show that GCUA function is very nice and valid. The hands with GCUA function can meet the requirements of grasping and operating with lower control and cost, which is the middle road between traditional under-actuated hands and dexterous hands.

This article discusses the use of a brain–computer interface (BCI) to obtain emotional feedback from a human in response to the motion of humanoid robots in collaborative environments. The purpose of this study is to detect the human satisfaction level and use it as a feedback for correcting and improving the behavior of the robot to maximize human satisfaction. This article describes experiments and algorithms that use human brains activity collected through BCI in order to estimate the level of satisfaction. Users wear an electroencephalogram (EEG) headset and control the movement of the robot by mental imagination. The robots responds to the mental imagination may not be the same as human mental command and this will affect the emotional satisfaction level. The headset records brain activity from 14 locations on the scalp. Power spectral density of each EEG frequency band and four largest Lyapunov exponents of each EEG signal form the feature vector. The Mann–Whitney–Wilcoxon test is then used to rank all the features. The highest rank features are then selected to train a linear discriminant classifier (LDC) to determine the satisfaction level. Our experimental results show an accuracy of 79.2% in detecting the human satisfaction level.

Recently, there has been a growing body of research that supports the effectiveness of using non-pharmacological cognitive and social training interventions to reduce the decline of or improve brain functioning in individuals suffering from cognitive impairments. However, implementing and sustaining such interventions on a long-term basis is difficult as they require considerable resources and people, and can be very time-consuming for healthcare staff. Our research focuses on making these interventions more accessible to healthcare professionals through the aid of robotic assistants. The objective of our work is to develop an intelligent socially assistive robot with abilities to recognize and identify human affective intent to determine its own appropriate emotion-based behavior while engaging in assistive interactions with people. In this paper, we present the design of a novel human-robot interaction (HRI) control architecture that allows the robot to provide social and cognitive stimulation in person-centered cognitive interventions. Namely, the novel control architecture is designed to allow a robot to act as a social motivator by encouraging, congratulating and assisting a person during the course of a cognitively stimulating activity. Preliminary experiments validate the effectiveness of the control architecture in providing assistive interactions during a HRI-based person-directed activity.

This article presents an algorithm for 3D upper body tracking. This algorithm is a combination of two well-known methods: annealing particle filter and belief propagation. It is worth to underline that the 3D body tracking presents a challenging problem because of the high dimensionality of state space and so because of the huge computational time. In this work, we show that with our algorithm, it is possible to tackle this problem. Experiments both on real and synthetic human gesture sequences demonstrate that this combined approach leads to reliable results, as it reduces computational time without loosing robustness.

Network robot system (NRS) is a new concept that integrates physical autonomous robots, environmental sensors, and human–robot interactions through network-based cooperation. The aim of this paper is to provide a ubiquitous and cooperative service framework for NRS. We first present foundational concepts of semantic map and service definition for the framework. Then, in order to generate feasible service configurations to fulfill tasks, we propose service configuration and reconfiguration algorithms, which dynamically search the appropriate service configurations for different tasks. Additionally, we put forward a service reasoning and enabling process to tackle the service unavailable problems. A cost evaluation function for service configuration is also proposed to facilitate the selection of suitable configurations. We tested and evaluated the framework in both simulation system and physical environment. Specifically, by separately varying the parameter settings, system performance was measured in three aspects: the success rate of tasks, the average waiting time per task, and the average cost per task. The experiment results indicate that the versatile service framework provides self-adaptive capability and utilizes available resources efficiently under a range of different scenarios.

This work presents a stereovision-based obstacle avoidance method for autonomous mobile robots. The decision about the direction on each movement step is based on a fuzzy inference system. The proposed method provides an efficient solution that uses a minimum of sensors and avoids computationally complex processes. The only sensor required is a stereo camera. First, a custom stereo algorithm provides reliable depth maps of the environment in frame rates suitable for a robot to move autonomously. Then, a fuzzy decision making algorithm analyzes the depth maps and deduces the most appropriate direction for the robot to avoid any existing obstacles. The proposed methodology has been tested on a variety of self-captured outdoor images and the results are presented and discussed.

Swimming performance is one of the primary concerns and applications to the underwater robots, such as thrust force relating to swimming velocity. As fish's swimming involves the kinematics of its own body and the hydrodynamic interaction with the surrounding fluid, it is difficult to formulate a precise mathematical model by purely analytical approaches. In order to avoid tedious parameter studies in evaluating its performance, this paper proposes a semi-empirical method to model the steady-state swimming performance of a BCF (body and/or caudal fin) biomimetic robotic robot. By using a dimensional analysis method, the semi-empirical model for predicting the thrust force generated by a BCF-oscillating swimming mode is derived. Thereafter, the swimming velocity prediction model is established based on the predictive thrust model together with the use of fundamental theory on drag force and the regression analysis on the experimental data. The model shows a reasonable prediction capability as the resultant predicted results are in good agreement with experiment data. Therefore, the proposed modeling method can be used for a quick prediction of the swimming performance in terms of thrust and velocity. The proposed methodology can be extended to other types of fish robots in real environment, by including changes to relevant parameters.

In recent times, the chances of robot–human contact have increased; hence, safety is necessitated with regard to such contact. Thus, manipulators using a pneumatic rubber artificial muscle, which is lightweight and flexible, are studied. However, this artificial muscle manipulator has faults such as slow response and limited instantaneous power due to operation by air pressure. Because of these faults, uncontrollable vibrations can occur, leading to instability in the arm when an object is held and lifted. In this study, an artificial muscle manipulator with one DOF and a variable rheological joint mechanism using MR fluid is developed. Vibration control of the arm using MR fluid is realized when an object is held and lifted, confirming the reduction in vibration due to the MR effect.

Extended Kalman filter (EKF) has been used as a popular choice to solve simultaneous localization and mapping (SLAM) problem. However, SLAM algorithm based on EKF-SLAM has two serious drawbacks, namely the linear approximation of nonlinear functions and the calculation of Jacobin matrices. For solving these problems, SLAM algorithm based on unscented Kalman filter (UKF-SLAM) has been recently proposed. However, the performance of the UKF-SLAM and thus the quality of the estimation depends on the correct a priori knowledge of process and measurement noise covariance matrices respectively denoted by Q_k and R_k. Imprecise knowledge of these statistics can cause significant degradation in performance. This article proposes the development of an adaptive neuro-fuzzy UKF (ANFUKF) for SLAM. The Adaptive neuro-fuzzy attempts to estimate the elements of R_k matrix in the UKF-SLAM algorithm at each sampling instant when measurement updating step is carried out. The adaptive neuro-fuzzy inference system (ANFIS) supervises the performance of the UKF-SLAM with the aim of reducing the mismatch between the theoretical and actual covariance of the innovation sequences. The free parameters of ANFIS are trained using the steepest gradient descent (GD) to minimize the differences of the actual value of the covariance of the residual with its theoretical value as much as possible. The simulation results show the effectiveness of the proposed algorithm.