UAV formation control with obstacle avoidance using improved artificial potential fields

Researches on unmanned aerial vehicle (UAV) formation control are attracting more and more researchers' attention. Obstacle avoidance is a key problem of UAV formation. In this paper, an improved artificial potential field (APF) method is proposed for solving the problem of UAV formation control with obstacle avoidance in a complex environment. Point-mass models with kinematic constraints have been employed for UAVs, and autopilots are modelled as first-order systems. Structural constraints of desired formation configuration are considered in the design of the attraction potential field among UAVs, which ensures the convergence of UAVs to a given formation configuration for any initial condition. Furthermore, for UAVs flying at high speeds, an improved APF method combined with formation division method is introduced to realize more flexible formation obstacle avoidance in congested settings where many static and dynamic obstacles exist. Finally, simulations of UAV formation keeping and obstacle avoidance are presented. The proposed methods enable UAVs with different initial states to build expected formation, track the desired trajectory and avoid collisions and obstacles during formation flight. The simulation results demonstrate the effectiveness of the improved APF method.