Distributed model predictive control for multi-agent systems with coupling constraints

This paper proposes an improved distributed model predictive control (DMPC) scheme for multi-agent systems with coupling constraints by applying compatibility constraint and deviation penalisation. Firstly, a sufficient condition based on the compatibility constraint to satisfying cooperative coupling constraints is given, which enables control performance and coupling constraints dependent on the compatibility constraint. Then, in order to optimise the control trajectory and to improve the consistency of the actions between agents, the deviation between what an agent is actually doing and what its neighbours believe it is doing is penalised in the cost function. At each sampling instant, the compatibility constraint of each agent is set tighter than the previous sampling instant. The optimal control problem is formulated as quadratic programming with quadratic constraints. Finally, two simulation examples are given to illustrate the effectiveness of the proposed scheme.