Distributed Predictive Control of Multi-Agent Systems Based on Error Upper Bound Approach

In this paper, a novel distributed predictive control (DMPC) method for multi-agent systems based on error upper bounds is proposed. To reduce the communication burden, the error upper bound condition between the subsystem and the neighbor subsystems is calculated by introducing the min-max function from the local state error of neighbouring subsystems. Additionally, an improved coupling constraints to describe the relationship between the neighbouring subsystems is introduced. Then, the proposed DMPC algorithm with kinds of the constraints is given, including the terminal cost, the terminal set and the terminal controller. Furthermore, the feasibility of the proposed DMPC algorithm is analyzed and the stability conditions of multi-agent systems are derived. Finally, a numerical example is given to verify the effectiveness of the method.