Hierarchical Control of Trajectory Planning and Trajectory Tracking for Autonomous Parallel Parking

For the parallel parking problem in narrow space, this paper proposes a trajectory tracking control method with novel planning layer for autonomous based on numerical optimization algorithm and model predictive control. In layer, vehicle kinematics suitable low-velocity scene is established. Considering physical constraints, boundary condition obstacle avoidance constraints during process, task described as an optimal problem, further transformed into nonlinear programming by Gauss pseudo-spectral method. Taking shortest completion time objective function, trajectories of large, medium small spaces are obtained, respectively. A controller designed layer. The linear error used prediction model, quadratic adopted rolling controller. velocity front-wheel swing angle obtained signals tracking. Through CarSim Simulink's co-simulation, feasibility effectiveness proposed verified. co-simulation results show that maximum errors horizontal longitudinal positions less than 0.15m. heading 2° under three different spaces. Real tests carried out to verify hierarchical test can park space safely, quickly accurately when actual detected. plan complete operation along planned trajectory.