Discrete Neural Dynamics Combined With Kalman Filter for Cooperative Control of Multirobot Systems With Constraints

Abstract

This article presents a distributed scheme for cooperative control of multirobot systems. Specifically, the scheme consists of a minimum velocity norm (MVN) strategy with temporal and physical constraints and a discrete neural dynamics solver incorporated with a Kalman filter, which ensures the least interference from lagging errors and internal/external noises during the cooperative control of multirobot systems. It is concluded that the proposed scheme is convergent and robust through theoretical analyses. Simulative results show that the multirobot system aided with the proposed scheme is able to complete the given task successfully in the presence of noise. Qualitative comparisons of the proposed scheme with other existing ones are carried out to highlight its superiority.