A New Distributed Model Predictive Control for Unconstrained Double-Integrator Multiagent Systems

In this paper, a distributed model predictive control (DMPC) is proposed for static formation of unconstrained double-integrator multiagent systems. The formation problem is formulated in the leader-follower framework, where the leaders have access to both their own and relative neighboring information, and the followers only have access to relative neighboring information. In the process of optimization, only current relative information is communicated between neighboring agents. For each agent, its predicted control vector and neighboring control vectors are regarded as decision variables for optimization, but only the own optimized current control of the agent is implemented. An analytical solution to the DMPC is obtained, and new sufficient conditions are given for achieving the static formation.