A fuzzy decision-making system for multi-criteria robot motion planning and control is described for a constant speed omni-directional mobile robot with active stereo cameras capable of panoramic viewing. The robot can be commanded to move to a target point through an unknown environment while satisfying multiple criteria. Advantages of the proposed method are confirmed by simulation. A multi-agent approach is also described for path planning with multiple robots.

The authors propose robot force control strategies that are robust against disturbance and parameter variation. These strategies are expansions of a disturbance observer which estimates disturbance and parameter variations of motors with a simple computation. Observers are proposed to estimate parameter variations of the environment on which the force is imposed. Since the control system is fixed to a nominal system by the observer, a second derivative of force can be controlled. Force control is introduced using a controller of the second derivative of the force. The force response similar to the force command is realized in spite of the disturbance and the parameter variations of the control object. The effectiveness was confirmed experimentally.>

The authors propose a simple high-performance control strategy based on disturbance compensation by the disturbance observer in task space. The disturbance observer makes robot manipulators linear in task space and realizes acceleration control. The inverse kinematics routine can be simplified, because the disturbance observer compensates not only the disturbance but also the error due to the simplification of the inverse kinematics routine. The strategy was applied to a three-degree-of-freedom direct drive robot to show its effectiveness experimentally.>

A simple high-performance hybrid position/force control that is based on the acceleration tracing orientation method (ATOM) is proposed. From the analysis of ATOM, it is pointed out that precise acceleration control is a key to improve the performance of the motion of the robot. The realization of a strict acceleration controller, however, is almost impossible by the conventional disturbance compensation methods. Recent research on a disturbance observed shows the realization of the acceleration controller. Compared with the inverse dynamics, the observer is simple and robust against parameter variation. Moreover, the disturbance observer can perform parallel computation in each joint by the same algorithm. ATOM was applied to a three-degrees-of-freedom robot to show the effectiveness of the ATOM experimentally.>