Redundant manipulator infinity-norm joint torque optimization with actuator constraints using a recurrent neural network

In this paper, a neural network based on the projection and contraction method is employed to compute the minimum infinity-norm joint torques of redundant manipulators, which explicitly takes into account the joint torque limits. While the desired accelerations of the end-effector for a specified task are fed into the network, a driving joint torque vector which has the maximum component in magnitude being minimized and is never exceeding the joint torque limits is generated as the neural network output. The proposed neural torque control scheme is shown to be capable of effectively generating the bounded minimum infinity-norm driving joint torques of redundant manipulators.