Obstacle avoidance and formation control of multiple unmanned vehicles in complex environments based on artificial potential field method

Addressing the increasing complexity of tasks, single unmanned vehicles have become unable to meet actual operational requirements, prompting a shift toward multivehicle formation systems. However, in complex environments, issues such as high collision rates and unstable formations in multivehicle obstacle avoidance and formation control persist. A review of existing literature reveals that most research focuses on static obstacle environments, which do not accurately reflect real-world conditions. To tackle the issues of collision with obstacles and formation instability in dynamic and dense environments, a multivehicle obstacle avoidance and formation control method based on the potential field method was proposed. The attraction potential field function was modified to stabilize the attraction force at certain distances, addressing problems like vehicle–obstacle collisions and target point inaccessibility owing to excessive gravity in the early stage. A smoother repulsive was implemented to optimize the repulsive potential field function, preventing unmanned vehicles from lingering near obstacles caused by excessive repulsive force when too close to the obstacles. The A stability force was defined to maintain stable formations during movement, allowing vehicles to break free from local minima under its influence. The method also incorporated the velocity repulsive potential field for dynamic obstacles and an attraction potential field for sparse obstacles, enhancing the success rate of obstacle avoidance and path planning in complex environments. Compared to traditional artificial potential field methods and the improved algorithms, the simulation results show that the proposed method effectively maintains formation stability and exhibits high anti-interference capabilities in complex environments. Specifically, the success rate of obstacle avoidance in dynamic environments increased by 35% compared to traditional algorithms and by 10% compared to improved algorithms. In dense, dynamic obstacle environments, the success rate increased by 55% and 10%, respectively. The proposed method provides a reference method for multivehicle formation and obstacle avoidance in complex environments.