Cooperative Control Method of Multi-Agent Formation and Obstacle Avoidance

With the development of intelligent agent technology, single-intelligent agent can no longer meet the needs of complex tasks such as search and rescue. Multi-agent collaborative work has become the mainstream of development, and the formation control technology is the basis of multi-agent to carry out the task smoothly. In search and rescue and other complex tasks, multi-agent formation needs to quickly form stable formations, avoid dynamic obstacles and accurately arrive at the target area. However, the current formation control suffers from poor adaptability to the dynamic environment, poor formation transformation and stability, and slow convergence velocity, etc. Therefore, this paper proposes a multi-agent formation control strategy combining the Particle Swarm Optimization (PSO) algorithm, consensus algorithm and artificial potential field (APF) method. Specifically, the PSO algorithm is employed for global optimization of target formation configurations, the artificial potential field method is utilized to enable real-time evasion of dynamic obstacles, and the consensus algorithm is leveraged to maintain the structural stability of the formation. This framework achieves a positioning precision of 0.2 meters across diverse formation transitions. It effectively navigates obstacle avoidance within a 2-meter detection radius and regulates system convergence time to within 30 seconds. These findings establish a robust solution characterized by high precision and rapid responsiveness, offering a reliable technical approach for multi-agent cooperative control in dynamic operational environments.