Abstract: Four-wheel independent steering(4WIS) vehicles possess outstanding maneuverability, enabling more flexible motion control in complex environments. However, existing trajectory planning methods have yet to fully exploit the motion characteristics of 4WIS vehicles. Most approaches approximate the vehicle contour using a fixed number of discs to impose obstacle avoidance constraints, which limits the balance between avoidance accuracy and computational efficiency and reduces adaptability to dynamic environments. To address these issues, this paper proposes a dynamic trajectory optimization method for 4WIS vehicles operating in unstructured and narrow environments. First, the hybrid A^* algorithm is enhanced by incorporating the multi-modal motion characteristics of 4WIS vehicles to improve node expansion and cost evaluation. A dynamic heading adjustment strategy is introduced, along with an adaptive multi-disc collision detection method that adjusts the collision model according to environmental complexity, thereby improving search efficiency. Next, a trajectory optimization model for 4WIS vehicles is constructed. An adaptive multi-disc obstacle avoidance constraint is proposed, and driving corridors are generated based on the collision detection model corresponding to each path point, enabling linearization of obstacle avoidance constraints and improving the solution success rate and computational efficiency in complex environments. Furthermore, to address trajectory planning in dynamic environments, a local trajectory replanning method based on a fuzzy dynamic window approach is proposed, enabling effective avoidance of dynamic obstacles and newly introduced static obstacles. Simulation results demonstrate that the proposed method can generate smooth, collision-free trajectories. Compared with the conventional hybrid A^* algorithm, the proposed method increases the planning success rate by 11.25% and reduces computation time by 27.64 s, significantly enhancing the trajectory planning efficiency and safety of 4WIS vehicles in complex scenarios.

Key words: four-wheel independent steering, trajectory optimization, hybrid A^*, adaptive multi-disc, trajectory replanning