Cooperative formation control of fixed-wing UAVs based on deep reinforcement learning

In recent years, artificial intelligence has attracted research interest and developed rapidly. As one of its representative technologies, deep reinforcement learning methods have gradually combined with various fields to develop numerous research results. Aiming at the motion constraints of fixed-wing aircraft and the existence of disturbances, model uncertainties, etc., this paper designs a fixed-wing aircraft formation controller based on the PID control structure and using the Proximal Policy Optimization algorithm (PPO). The simulation results show that the PID parameters change adaptively with the state of the system, and the system can quickly form the desired formation.