Formation Control of a Multi Agent System Based on Robust Model Predictive Control

In this paper, a method for formation control of a discrete-time nonlinear multi agent system has been presented using a robust model predictive control(RMPC) approach. Formation control of robots is one of the requirements for expanding use of grouped cooperation to accomplish complex activities. In thismethod, the problem has been solved by assuming decentralized architecture and leader-follower communication structure. The goal of control is toachieve the desired formation of agents while tracking and following the leader in the presence of structural uncertainty. Using the considered formation control method based on robust model predictive control makes the closed-loop system robust to the structural uncertainty previously considered in the literature and handled by a state feedback approach. Simulation results demonstrate the effectiveness of the robust model predictive control method and its better performance in a formation control problem consisting a group of three robots following a leader, compared to previously proposed state feedback method.

Keywords:

Multi Agent Systems , Formation control , Robust model predictive control , Structural uncertainty