Distributed Adaptive Coordinated Control of Multi-Manipulator Systems Using Neural Networks

On many occasions, all the manipulators in the multi-manipulator system need to achieve the same joint configuration to fulfill certain coordination tasks. In this chapter, a distributed adaptive approach is proposed for solving this coordination problem based on the leader-follower strategy. The proposed algorithm is distributed because the controller for each follower manipulator is solely based on the information of connected neighbor manipulators, and the joint value of leader manipulator is only accessible to partial follower manipulators. The uncertain term in the manipulator’s dynamics is considered in the controller design, and it is approximated by the adaptive neural network scheme. The neural network weight matrix is adjusted on-line by the projection method, and the pre-training phase is no longer required. Effects of approximation error and external disturbances are counteracted by employing the robustness signal. According to the theoretical analysis, all the joints of follower manipulators can be regulated into an arbitrary small neighborhood of the value of leader’s joint. Finally, simulation results are given to demonstrate the satisfactory performance of the proposed method.