Abstract: A piecewise Gauss pseudospectral method is proposed as a path planning scheme for autonomous parking in different parking spaces in this paper. Firstly, the path planning problem for autonomous parking is transformed into an optimal control problem by establishing a vehicle kinematics model with dynamics constraints, end-point constraints and path obstacle avoidance constraints applied, and the optimal control problem is discretized into a nonlinear problem by Gauss piecewise pseudospectral method in different parking spaces. Then with minimizing total parking time as the objective, an optimization is conducted on nonlinear problem by using interior point method to obtain the optimal parking path. Next, a simulation model is built with Matlab/Simulink, and five autonomous parallel parking conditions with different lengths of parking slot defined, in which the length of parking slot is 1.1-1.8 times of vehicle length. The results of simulation show that the proposed method can uniformly and effectively solve the problem of path planning for autonomous parking, with a shorter parking time and a significantly higher convergence speed in simulation compared with the traditional pseudospectral method. Finally a real vehicle test is also performed to further verify the effectiveness of the method proposed.

Key words: autonomous parking, path planning, piecewise Gauss pseudospectral method, optimal control