Policy iteration based near-optimal control scheme for robotic manipulator with model uncertainties

摘要

This paper addresses a single network adaptive critic(SNAC) based continuous time near-optimal control strategy for robotic manipulator with partially known dynamics. The optimal control of the robot manipulator is generalized to the control problem that of a continuous time nonlinear input affine system and the solution is obtained through adaptive critic based approach. Such generalization facilitates to achieve near-optimal solution with SNAC, which results in a computationally efficient control scheme. The discussed policy iteration scheme reaches optimality through learning even in the presence of unmodelled dynamics. The validation of the proposed algorithm is done through simulation on a robotic manipulator model. The results show that the near optimal performance is achieved while controlling the manipulator with the proposed SNAC based strategy.