A framework of neural networks based consensus control for multiple robotic manipulators

A framework for neural networks (NN) based consensus control is proposed for multiple robotic manipulators systems (MRMS) under leader–follower communication topology. Two situations, that is, fixed and switching communication topologies, are studied by using adaptive and robust control principles, respectively. Radial basis function (RBF) NN enhances estimator and observer are developed to estimate system uncertainty and obtain the leader manipulator׳s control torque online. By using the Lyapunov stability theory, an adaptive consensus control algorithm is designed to tune the weight of the RBF NN online, which can stabilize the consensus error to a small residual set. On this basis, a novel robust control algorithm is presented to eliminate the estimating errors caused by RBF NN, which can achieve asymptotical stability. The stability of the proposed approaches is analyzed by using Lyapunov methods. Finally numerical bench tests are conducted to validate the effectiveness of the proposed approach.