This paper discusses a planar 2-DOF (degrees of freedom) parallel kinematic machine with actuation redundancy. Its inverse dynamic model is constructed by utilizing the Newton–Euler method based on the kinematic analysis. However, the dynamic model cannot be solved directly because the number of equations is less than the number of unknowns, which is due to the redundant force. In order to solve this problem, the relationship between the deformations of the links and the position errors of the moving platform are further explored. Then a novel method, which aims at minimizing the position errors of the machine, is proposed to optimize the redundant force. It also enables to solve the dynamic model. Finally, the dynamic performance analyses of this machine and its non-redundant counterpart are provided by numerical examples. Besides, another optimization method proposed for minimizing the constraint forces is also applied for comparison. The results show the effectiveness of the novel methods in improving the position precision of the machine.

A disturbance rejection controller is proposed based on the general dynamic model of 3D biped robots. For the first time, with this proposed approach, not only the Zero Moment Point (ZMP) location remains unchanged in presence of disturbances but also the longitudinal and lateral ground reaction forces and the vertical twist moment remain unchanged. This way, slipping as well as tipping is prevented by the controller. The swing phase of the robot's walking gait is considered. An integral sliding mode architecture is chosen for the disturbance rejection. The support forces and moments of the stance foot are the control outputs. The acceleration of the arm/body joints are chosen as the inputs. During the disturbance rejection, the leg joints remain at their desired trajectory. Since the leg joint trajectories are unaffected, the robot is still able to complete its step as planned, even when bounded disturbances are experienced. For simulations, the general method is applied to an 18-degree of freedom biped humanoid robot. Simulations show that the controller successfully mitigates bounded disturbances and maintains all of the support reactions extremely close to their desired values. Consequently, the shift in the position of the ZMP is negligible, and the robot foot does not slip.

In this paper, the robust, optimal, output control problem is dealt with for a 3-degree-of-freedom laboratory helicopter. The control goal is to achieve the practical tracking of the desired elevation and pitch angles without the angular velocity feedback. A nominal linear time-invariant system is introduced and the real system is considered as the nominal one with uncertainties, including parameter perturbations, nonlinear time-varying uncertainties, and external disturbances. An observer is first used to estimate angular velocity. Then a nominal controller based on the optimal control method is designed for the nominal system to achieve the desired tracking properties. Lastly, a robust output compensator is added to restrain the effects of uncertainties in the real system. It is shown that asymptotic tracking properties and robust stability can be achieved. Experimental results on the laboratory helicopter are shown to verify the effectiveness of the proposed control method.

We propose a fuzzy weighted subtask controller for a redundant robot manipulator. To expand the feasibility of the inverse kinematic solution, we introduce a weighted pseudo-inverse that changes the null-space of the Jacobian. The weights of elements in the pseudo-inverse are obtained using fuzzy rules that are related to the null-space velocity tracking error. With the pseudo-inverse, we develop a task space controller to track a desired task space trajectory and subtask control input. We propose a weighted subtask controller for multiple subtasks. The results of a simulation and experiment using a seven-degree-of-freedom whole arm manipulator robot show the effectiveness of the proposed controller with multiple subtasks.

This paper presents a novel approach for the description of physical human-robot interaction (pHRI) tasks that involve two-arm coordination, and where tasks are described by the relative pose between the human hand and the robot hand. We develop a unified kinematic model that takes into account the human-robot system from a holistic point of view, and we also propose a kinematic control strategy for pHRI that comprises different levels of shared autonomy. Since the kinematic model takes into account the complete human-robot interaction system and the kinematic control law is closed loop at the interaction level, the kinematic constraints of the task are enforced during its execution. Experiments are performed in order to validate the proposed approach, including a particular case where the robot controls the human arm by means of functional electrical stimulation (FES), which may potentially provide useful solutions for the interaction between assistant robots and impaired individuals (e.g., quadriplegics and hemiplegics).

In this paper, a consensus-based control strategy is presented to gather formation for a group of differential-wheeled robots. The formation shape and the avoidance of collisions between robots are obtained by exploiting the properties of weighted graphs. Since mobile robots are supposed to move in unknown environments, the presented approach to multi-robot coordination has been extended in order to include obstacle avoidance. The effectiveness of the proposed control strategy has been demonstrated by means of analytical proofs. Moreover, results of simulations and experiments on real robots are provided for validation purposes.

For reproducing the manipulation of Traditional Chinese Medicine (TCM) remedial massage and meanwhile guaranteeing safety, a 4-degree-of-freedom anthropomorphic robotic arm with integrated elastic joints is developed, and a passivity-based impedance control is used. Due to the series elasticity, integrated joints may minimize large forces that occur during accidental impacts, and, further, may offer more accurate and stable force control and a capacity for energy storage. Human expert's fingertip force curve in the process of massage therapy is acquired in vivo by a dedicated measurement device. Then three massage techniques, pressing, kneading, and plucking, are implemented by the soft arm, respectively, on torso model in vitro and on human body in vivo. Experimental results show that the developed robotic arm can effectively replicate the TCM remedial massage techniques.

In this paper, we present the design of two serial spherical mechanisms to substitute for a single spherical joint that is usually used to connect the platform with the base in three degrees of freedom parallel mechanisms. According to the principle derived from the conceptual design, through using the two serial spherical mechanisms as the constraint limb, several redundantly actuated parallel mechanisms are proposed for ankle rehabilitation. The proposed parallel mechanisms all can perform the rotational movements of the ankle in three directions while at the same time the mechanism center of rotations can match the ankle axes of rotations compared with other multi-degree-of-freedom devices, due to the structural characteristics of the special constraint limb and platform. Two special parallel mechanisms are selected to analyze their kinematical performances, such as workspace, dexterity, singularity, and stiffness, based on the computed Jacobian. The results show that the proposed scheme of actuator redundancy can guarantee that the redundantly actuated parallel mechanisms have no singularity, better dexterity, and stiffness within the prescribed workspace in comparison with the corresponding non-redundant parallel mechanisms. In addition, the proposed mechanisms possess certain reconfigurable capacity based on control strategies or rehabilitation modes to obtain sound performance for completing ankle rehabilitation exercise.

We consider a team of autonomous kinematically controlled non-holonomic planar Dubins car-like vehicles. The team objective is to encircle a given target so that all vehicles achieve a common and pre-specified distance from it and are uniformly distributed over the respective circle, and the entire formation rotates around the target with a prescribed angular velocity. The robots do not communicate with each other and any central decision-maker. The sensing capacity of any vehicle is heavily restricted: It has access only to the distance to the target and to the distances to the companion vehicles that are in a given disc sector centered at the vehicle at hand; no robot can distinguish between its companions, and does not know their parameters. A distributed control law is proposed, and mathematically rigorous proofs of its non-local convergence as well as collision avoidance property are presented. The performance of the control law is illustrated by computer simulations and experiments with real robots.

This paper presents optimization-based dynamic three-dimensional (3D) human running prediction. A predictive dynamics method is used to formulate the running problem, and normal running is formulated as a symmetric and cyclic motion. In addition, a slow jog along curved paths has been formulated. It is a non-symmetric running motion, so a stride formulation has been used. The dynamic effort and impulse are used as the performance measure, and the upper body yawing moment is also included in the performance measure. The joint angle profiles and joint torque profiles are calculated for the full-body human model, and the ground reaction force is determined. The effects of foot location and orientation on the running motion prediction are simulated and studied. Simulation results from this methodology show good correlation with experimental data obtained from human subjects.

This paper presents a vision-based obstacle avoidance design using a monocular camera onboard a mobile robot. A novel image processing procedure is developed to estimate the distance between the robot and obstacles based-on inverse perspective transformation (IPT) in an image plane. A robust image processing solution is proposed to detect and segment a drivable ground area within the camera view. The proposed method integrates robust feature matching with adaptive color segmentation for plane estimation and tracking to cope with variations in illumination and camera view. After IPT and ground region segmentation, distance measurement results are obtained similar to those of a laser range finder for mobile robot obstacle avoidance and navigation. The merit of this algorithm is that the mobile robot can have the capacity of path finding and obstacle avoidance by using a single monocular camera. Practical experimental results on a wheeled mobile robot show that the proposed imaging system successfully obtains distances of surrounding objects for reactive navigation in an indoor environment.

A continuous finite-time control scheme for networked bilateral teleoperation is proposed in this brief. The terminal sliding mode technology is used and new master–slave torques are designed. With the new controller, the coordination error of the master manipulator and the slave manipulator converges to zero in finite time. Moreover, the reaching time and the sliding time can be derived. Finally, the comparisons are performed and simulations show the effectiveness of the proposed approach.