International Journal of Humanoid Robotics

3D visual tracking is useful for many applications. In this paper, we propose two different ways for different system configurations to optimize particle filter for enhancing 3D tracking performances. On the one hand, a new data fusion method is proposed to obtain the optimal importance density function for active vision systems. With this method, the importance density function in particle filter can be modified to represent posterior states by particle crowds in a better way. Thus, it makes the tracking system more robust to noise and outliers. On the other hand, we develop a method for reconfigurable vision systems to maximize the effective sampling size in particle filter, which consequentially helps to solve the degeneracy problem and minimize the tracking error. Simulation and experimental results verified the effectiveness of the proposed method.

This paper presents impedance controllers with adaptive friction compensation for the five-finger dexterous robot hand DLR-HIT II in both joint and Cartesian space. An FPGA-based control hardware and software architecture with real-time communication is designed to fulfill the requirements of the impedance controller. Modeling of the robot finger with flexible joints and mechanical couplings in the differential gear-box are described in this paper. In order to address the friction due to the complex transmission system and joint coupling, an adaptive model-based friction estimation method is carried out with an extended Kalman filter. The performance of the impedance controller with both adaptive and parameter-fixed friction compensations for the robot hand DLR-HIT II are analyzed and compared in this paper. Furthermore, gravity estimation is implemented with Least Squares technique to address uncertainties in gravity compensation due to the close proximity and complexity of robot hand components. Experimental results prove that accurate position tracking and stable torque/force response can be achieved with the proposed impedance controller with friction compensation on five-finger dexterous robot hand DLR-HIT II.

This paper introduces a relay-based approach to control the motion of robot joints with friction and gravity load. This approach uses relay feedback tests to estimate the disturbances including Coulomb friction, viscous friction, and the gravity load. The relay feedback tests take the robot joint angular as the feedback signal and identify the friction value and the gravity load at the same time. A control scheme is then presented including a feedforward friction compensator and a feedforward gravity compensator based on the estimated results. With the disturbances properly compensated, the proposed approach improves the tracking performance of a robotic system. Simulation and experimental results are presented to verify the effectiveness of the proposed method.

Some of noncontact biometric ways have been used for user authentication system of partner robots, such as visual-based recognition methods and speech recognition. However, the methods of visual-based recognition are sensitive to the light noise and speech recognition systems are perturbed to the acoustic environment and sound noise. Inspiring from the human's capability of compensating visual information (looking) with audio information (hearing), a visual-audio integrating method is proposed to deal with the disturbance of light noise and to improve the recognition accuracy. Combining with the PCA-based and 2DPCA-based face recognition, a two-stage speaker recognition algorithm is used to extract useful personal identity information from speech signals. With the statistic properties of visual background noise, the visual-audio integrating method is performed to draw the final decision. The proposed method is evaluated on a public visual-audio dataset VidTIMIT and a partner robot authentication system. The results verified the visual-audio integrating method can obtain satisfied recognition results with strong robustness.

This paper presents a novel hands-free human machine interface (HMI) for elderly and disabled people by fusing multi-modality bioinformation abstracted from forehead electromyography (EMG) signals and facial images of a user. The interface allows users to drive an electric-powered wheelchair using face movements such as jaw clenching and eye blinking. An indoor environment is set up for evaluating the application of this interface. Five intact subjects participated in the experiment to drive the intelligent wheelchair following designated routes and avoiding obstacles. Comparisons are made between this new interface and the traditional joystick control in terms of the easiness of control, travel time, wheelchair trajectory, and error command. The experimental results show that the proposed new control method is comparable to the joystick control method and can be used as a hands-free controller for the intelligent wheelchair.

Current tendency of electromyography (EMG)-based prosthetic hand is to enable the user to perform complex grasps or manipulations with natural muscle movements. In this paper, empirical copula-based templates; including the unified motion template and the state-based motion template, are introduced to identify the naturally contracted surface EMG (sEMG) patterns for hand motion recognition. The unified motion template utilizes a dependence structure as a motion template, which includes one-to-one correlations of the SEMG feature channels with all the sampling points, while the state-based motion template divides the sampling points into different states and takes the union of the dependence structures of the different states. Comparison results have demonstrated that the proposed Empirical Copula-based methods can successfully classify different hand motions from different subjects with better recognition rates than Gaussian mixture models (GMMs). In addition, the state-based motion template has a better performance than the unified motion template especially for the complex hand motions.

This paper presents a wearable exoskeleton upper limb rehabilitation robot that consists of eight active joints and one passive joint. Each active joint is driven by two pneumatic muscles (PMs) in opposing pair configuration and each PM drives the joint through a steel wire with a flexible sleeve and a tension device. These factors incorporate with characteristics of PM make the rehabilitation robot system very complex (nonlinear, time-varying, and time-delay). In this paper, the friction force and characteristics of open-loop control of one of the robotic joints has been studied. The quasi-static model of the robotic joint have been established based on experimental methods. In addition, the performances of three control modes have also been researched. These works would be helpful to the future work such as the control of the rehabilitation robot.

Grasping can be seen as two steps: placing the hand at a grasping pose and closing the fingers. In this paper, we introduce an efficient algorithm for grasping pose generation. By the use of preshaping and eigen-grasping actions, the dimension of the space of possible hand configurations is reduced. The object to be grasped is decomposed into boxes of a discrete set of different sizes. By performing finger reachability analysis on the boxes, the kinematic feasibility of a grasp can be determined. If a reachable grasp is force-closure and can be performed by the robotic arm, its grasping forces are optimized and can be executed. The novelty of our algorithm is that it takes into account both the object geometrical information and the kinematic information of the hand to determine the grasping pose, so that a reachable grasping pose can be found very quickly. Real experiments with two different robotic hands show the efficiency and feasibility of our method.

Since tactile sensing feedback can provide very useful and crucial information to the problems of contact, it can be widely applied in the control of the mobile manipulator in case of contacting with the complex or unknown environments. This paper presents a control method for a mobile modular manipulator (MMM) interacting with the environment, in which the feedback does not only include the force-sensing feedback but also the tactile sensing feedback. In order to detect the feedback information from the environment, a kind of tactile sensor is mounted at the tip of the end-effector of the MMM. The dynamic equations of the whole system are formulated and the hybrid control method is designed after several tactile sensing-assisted strategies are considered. A mobile manipulator is built up to make real experiments and the results validate the proposed methods.