Search

This paper studies the simultaneous trajectory and force control problem of mobile manipulators subject to holonomic and nonholonomic constraints with uncertainties and external disturbances. The main feature of this paper is that 1) adaptive robust control strategies, which avoid complex regressor matrix, are presented based on a suitable reduced dynamic model, the defined references signals and the mixed tracking errors; 2) the proposed control schemes guarantee the states of the system asymptotically converge to the desired trajectory, and ensure the boundedness of the constrained force with controllable bounds at the same time; and 3) asymptotic stability in the Lyapunov sense is acquired and rigorously proven. A numerical example is presented to illustrate the developed method.

In this paper, we develop a robot with the ability to secure human safety in humanrobot collisions arising in our living and working environments. The humansymbiotic service robot using compliant hybrid joints realizes human safety, absorbs impact force, and fulfills task. In unexpected or expected collisions with human, the arising impulse force is attenuated effectively by the proposed physical model. Owing to the displacement of the links, several recovery controls have been developed for the endeffector to maintain its desired task position after the collision. The force attenuation property has been verified through collision simulations and experiments in that the capability of the proposed passive arm in overcoming the limitations of active compliance control has been demonstrated.

We are aiming to improve the performance of a passive vibration isolator for ship-mounted antennas by designing the parameters of the isolator including nonlinear springs and dampers. It is known that a passive isolator has an inevitable problem, trade-off between minimizing the magnification of resonance and suppressing the magnifications in the region of high-frequency into lower levels. To solve this problem, we introduce a nonlinear model with more design parameters and adopt an optimization method, that is, minimizing the evaluation function consisting of two terms which indicate effects on resonant magnification and vibration isolation. By this approach, several available solutions, which can improve vibration isolation performance while with reduced magnification of resonance, are obtained.

A new design method based on the utilization of dynamically-coupled driving and joint stops is proposed for high dynamic performance robots, such as the robot performing hi high-speed dynamic motions. In the method, the desired dynamic perfo.mance and the boundary conditions of the dynamic motion are combined to be a design index (DI) first. Then the robot is initially designed by selecting light actuators and links under an assumption of utilizing dynamically-coupled driving and joint stops. By increasing the actuator capabilities step by step, an iterative process of minimizing DI is implemented to yield a robot that is lighter than conventional robots and capable of performing dynamic motions more efficiently by utilizing dynamically-coupled driving and joint stops. Based on the method, a two-link golf swing robot performing high-speed swings is designed. Simulation results indicate the method can reduce the needs for the torque and power as compared with conventional design methods. The experiment illustrates the dynamically-coupled driving and the joint stops can be used.

A new golf swing robot to simulate human motion has been developed. This paper deals with motion control of the robot for hitting a ball. For the case, it becomes important to control the motion of the robot after the impact, to avoid breaking a club. In this paper, a method for controlling the robot adaptable to various impact conditions is proposed. The motion state of the robot is estimated first by an extended Kalman filter, and a new reference trajectory is generated on-line according to the estimated state. The method has been implemented to the robot successfully

Macro Fiber Composites (MFC) is a new type of piezoelectric material with softness and big displacement, developed by NASA recently. It can be used for actuating, sensing and power generating. The purpose of our work is to develop a new and intelligent flight machine by utilizing the features of the MFC. In this paper, the concept of such flight machine and the driving method of a MFC-combined structure are proposed. The feasibility of flight by using the MFC-combined structure is discussed by simulations. Based on the simulation results, an active wing capable of flapping and feathering motions has been developed. The elementary performance of the wing is investigated by an aerodynamics experiment.

In this paper, robust force/motion control strategies are presented for mobile manipulators under both holonomic and nonholonomic constraints in the presence of uncertainties and disturbances. Robust controls are proposed based on a simplified dynamic model, the defined reference signals and the mixed tracking errors. The proposed control strategies guarantee that the system motion converges to the desired manifold with prescribed performance. Constraint force control is developed, which does not utilize force feedback control but is based on the passivity of hybrid joint. Experiment results validate not only the states of the system asymptotically converge to the desired trajectory but also the constraint force asymptotically converges to the desired force.

To achieve high-speed navigation in narrow, cluttered, dynamic, unstructured environments, a sensing system using a sonar ring for a mobile robot capable of rapid detection and avoidance of obstacles has been developed and a prototype of the sensing system with 24 sonar sensors is fabricated. A composite method of firing all ultrasonic sensors simultaneously with the scanning rate up to 66 Hz of panoramic detection and Dynamic Window Approach (DWA) is presented. The experimental results show that a rapid navigation of the mobile robot in the unstructured environment becomes possible.

Usually the dynamic capability of conventional manipulators is limited by heavy structure. Different from those manipulators, human can do many hyper dynamic manipulations while in a smart structure by utilizing dynamically-coupled driving between body and arms. Aiming at improving the dynamic capability of a manipulator in a large extent, a basic concept and general method on the mechanism design and control of the manipulator is proposed, just to simulate the action of human utilizing the dynamically-coupled driving. Furthermore, using joint stop in a manipulator is enhancing the effect of dynamically-coupled driving. To show the feasibility of the proposal, a hyper dynamic manipulator has been fabricated and fundamental experiments have been carried out. Experimental results have shown the important role of the dynamically-coupled driving, to realize an efficient hyper dynamic manipulation while in a smart structure. This paper describes the developed prototype and experimental results.

A golf swing robot consisting of an actuated joint and a passive joint with stopper has been developed to simulate the motion control skill in human swing. This paper describes a topic of control by learning, that is, how to achieve an optimal trajectory for the real system though learning. To realize the purpose, a method by repeating swing, model improvement by learning and optimal motion generation is proposed. Method, implementation and experimental results of the control are described in this paper.