Collision avoidance in multi-agent formation keeping cooperative control systems

This paper proposes a collision avoidance method for multi-agent formation keeping cooperative control systems. Each agent is assumed to evolve in discrete time, based on locally computed control laws, which depend on the local information gathered from the cooperating neighboring agents. The collision avoidance problem in multi-agent formation keeping cooperative control is formulated in a distributed model predictive control framework with an added safety guarantee condition. Furthermore, the relations between communication among agents and system performance is also shown in this paper. Finally, a simulation example is given to illustrate the method proposed in this paper.