Automatic obstacle avoidance control for autonomous vehicles based on potential field-ant colony algorithm

This paper puts forward an obstacle avoidance control scheme for autonomous vehicles integrating artificial potential field (APF) and ant colony optimization (ACO). In path planning, a relative distance factor is used to optimize the repulsive function of APF. To overcome premature convergence and blind searching of ACO, the resultant force of the modified APF provides heuristic guidance. Hybrid pheromone update and cubic B-spline smoothing are incorporated to accelerate convergence and enhance path performance. Additionally, a decoupled coordinated tracking framework is constructed: lateral motion is managed using LQR feedforward-feedback control, and longitudinal motion is governed by RBF neural network self-tuning PID control. Real-time speed feedback and lateral stability constraints are leveraged to build a lateral-longitudinal coordination rule. Simulation experiments confirm that the proposed method yields shorter collision-free paths, fast convergence, and low lateral errors at different speeds, realizing high-precision trajectory tracking of autonomous vehicles.