通信受限条件下基于切换协议的编队避障算法

Formation Obstacle Avoidance Algorithm Based on Handoff Protocol under Communication Constraints

针对通信距离受限条件下多智能体编队队形易受障碍物影响而打散队形进而无法继续维持编队的问题,提出了一种基于分布式协议与集中式协议相互切换的多智能体编队避障算法。首先,分别使用基于一致性理论的分布式控制和集中式控制算法使智能体系统形成编队;其次,引入人工势场函数,保证智能体的避障性能;最后,设计协议切换控制器,根据智能体系统是否处于障碍物斥力场内进行协议切换,保证智能体系统既能实现避障,又能维持较低的通信量。仿真结果表明,该算法与仅使用集中式协议的算法相比,系统的总通信量降低了32%,领导者智能体的通信量降低了53.5%,表明该算法有更强的适用性和鲁棒性。

Aiming at the problem that the formation of multi-agent formation is easily affected by obstacles and can not continue to maintain the formation under the condition of limited communication distance,a multi-agent formation obstacle avoidance algorithm based on mutual switching between distributed protocol and centralized protocol is proposed.Firstly,distributed control and centralized control algorithms based on consistency theory are used to form agent system formation.Then,the artificial potential field function is introduced to ensure the obstacle avoidance performance of the agent.Finally,the protocol switching controller is designed to switch the protocol according to whether the agent system is in the obstacle repulsion field,so as to ensure that the agent system can not only avoid obstacles,but also maintain low traffic.Simulation results show that compared with the algorithm using only centralized protocol,the total traffic of the system is reduced by 32%,and the traffic of the leader agent is reduced by 53.5%,indicating that the algorithm has stronger applicability and robustness.