基于分布式模型预测控制的欠驱动 AUV 编队控制

Compared with centralized model predictive control, distributed model predictive control(DMPC) is characterized by lower computational complexity and stronger fault tolerance and robustness, and it is widely used in multiagent formation control. In this study, an underactuated autonomous undersea vehicle(AUV) formation control method based on DMPC is proposed. Based on local neighbor information, the cost function and constraints of predictive control are constructed for each AUV controller, and the optimal control input in a certain time domain is solved by using an optimization algorithm. To solve the obstacle avoidance problem and communication delay problem that may exist in the formation system, obstacle avoidance methods based on distance and relative line of sight, as well as a waiting mechanism for problem solving after receiving all neighbor information, are designed. The simulation results demonstrate that, by using the method proposed in this study, the multi-AUV formation can remain stable under the conditions of obstacles and communication delays.