Adaptive neural network consensus control of multi-robot systems with output constraints

Abstract

Addressing the leader-follower consensus control problem with time-varying output constraints for a class of second-order nonlinear multi-robot systems, we apply a unified barrier Lyapunov function to transform the constrained output state into the unconstrained one, while removing the feasibility condition existing in the traditional barrier Lyapunov function. Using the radial basis function-based neural network to approximate the unknown nonlinear function, we derive an adaptive neural network consensus controller to ensure that each robot follows the predefined trajectory of a leader. We then verify the effectiveness of the consensus control via simulation studies on a team of robotic manipulators.