RRT based Path Planning for Autonomous Parking of Vehicle

Path planning is one of the most issues in the automatic parking system for vehicle. This paper presents a path planning method based on rapidly-exploring random tree (RRT) with non-holonomic constraint and kinematics model of vehicle. First, the kinematics model of car parking according to the vehicle kinematics equation is set up, and the non-holonomic constraints are put forward. Based on this model, the RRT algorithm is used to search parking path with the constraints. And then, to optimize the search efficiency, two strategies--target preference and bi-RRT are used and also the cost function is added for optimization. Besides, because of the new detected obstacles, a replanning method is used to replan the path using the feature of the RRT algorithm. Finally, the performance of the proposed method is verified on a simulation model based on matlab.