Obstacle Avoidance Formation Strategy for Unmanned Vehicles via Improved Grey Wolf Optimizer and Artificial Potential Field Method

This paper proposes an obstacle avoidance formation control strategy for multi-unmanned vehicles, integrating an improved grey wolf optimizer (IGWO) with a kinematics-integrated artificial potential field (APF) method. By embedding steering constraints and velocity limits into a kinematic-coupled APF model, path feasibility is ensured by adaptive attraction-repulsion forces. Furthermore, the IGWO enhances parameter optimization efficiency by employing a nonlinear convergence factor and a multi-objective fitness function that simultaneously optimizes path length, formation stability, and safety constraints. To improve adaptability in complex environments, a dynamic formation-switched strategy is introduced, which autonomously adjusts formation patterns and APF gains based on real-time obstacle density and channel width.