Obstacle avoidance path planning method for intelligent vehicle based on obstacle projection

Abstract Facing obstacles that can be crossed by the vehicle, the existing path planning can only make the vehicle detour as a whole. This method leads to excessive lateral displacement of obstacle avoidance. To solve this problem, this paper proposes and verifies a new path planning method based on obstacle projection. Firstly, the wheel collision region projection is determined according to the geometric size of the wheel and the range of the steering angle. And the projection vectors from wheel collision region projections to the center of gravity are obtained. The obstacle is projected to obtain four projections of the obstacle, so that the problem of wheels avoiding the obstacle is transformed into the problem of the wheel collision region projection at the center of gravity avoiding four projections of the obstacle. Then, in order to further improve the applicability of this method, it uses the position and orientation change of the vehicle in the curve for obstacle projection. Then, the hierarchical path planning realize the path planning with the wheel collision region projection at the center of gravity as the obstacle avoidance main body and avoid four projections of the obstacle. Finally, through the simulation of multiple groups of obstacles in different scenes, the effectiveness of the method is verified. The results demonstrate that the method proposed in this paper can plan paths crossing over obstacles, which is unattainable with traditional methods. Paths crossing over obstacles exhibit smaller lateral displacement of obstacle avoidance compared to detour paths, leading to enhancements in stability and comfort.