Stable Control of Robot Manipulator with Collision Phenomena (3rd Report, Force Control Experiment of the Robot Manipulator with Collision Phenomena by a Learning Control Using the Weighted Least-Squares Method)

Collision phenomena are very fast and nonlinear making it difficult to control the manipulator with collision phenomena. Therefore, in the past, manipulators moved slowly in order to avoid a collision. However, the necessity for high-speed tasks has been growing ; thus, it is indispensable to control the manipulator with collision phenomena. In such fast phenomena, it is effective to use learning control in the forward manner. In this paper, we propose a learning control method to optimize the weighted least-squares criterion of learning errors. This method is applied to obtain a unique control gain by the Riccati equation which has the state dimension equal to the sampling number. It is shown that the convergence of learning error can be readily assured because the present learning rule consists of a steady-state Kalman filter. Using this learning control method, we report the results of the experiment on force control with a collision phenomena.