Research On Path Planning And Control System For Automatic Parallel Parking

With the continuous increasing of car ownership,parking spaces become more precious and parking becomes more difficult.At the same time,parking is one of the main cause of traffic accidents.The automatic parking system can effectively solve these problems,so now the automatic parking system has become the more popular research topic,and this thesis focuses on parallel parking in the automatic parking system.Firstly,the kinematic model of vehicle is established by applying the basic knowledge of parking.The structure of this thesis is determined by investigating the research status and application status of parking.The model of vehicle and parking space is established and related parameters are introduced.Based on the ackerman steering geometry,the kinematic model of vehicle is established,and it is concluded that the vehicle trajectory is not related to the parking speed.Meanwhile,the speed requirements of parking are briefly explained.Secondly,parallel parking path planning is studied.According to different parking spaces,the arc-tangent parallel parking path planning method and the double-step adjustment parallel parking path planning method are studied and simulated.The simulation results show that the two parking methods can make the vehicles park smoothly,among which the double-step adjustment parallel parking method can realize parking in smaller parking spaces.Due to the curvature mutation at the tangent point of the arc-tangent path planning trajectory considering the vehicle's own constraints,the curve of the fifth order polynomial is used to optimize the trajectory,and the problem of curvature mutation at the tangent point is solved.Thirdly,the designed model predictive controller is used to track the parallel parking planning path.Combining path tracking and model predictive control algorithms,a vehicle parking path tracking controller is established based on the kinematic model of vehicle.Tracking simulation research is carried out with parallel parking path planning trajectories under different parking spaces,and is carried out with the PID algorithm under the same conditions.The simulation results show that the controller can track the trajectory in the arc-tangent planned path of the standard parking space and the two-step adjustment parking planning path in the smaller parking space.Compared with PID control,the controller's path tracking is more stability,smaller error.Parking can be achieved.Finally,an automatic parallel parking control system is designed and developed for parallel parking path planning and tracking.The experimental cart with STM32F103C8T6 as the controller and equipped with hardware such as motor drive,steering gear,sensors,communication modules,and display modules was determined.The software part of the experimental car is mainly divided into two parts: a remote control module and a parking control module.The car must not only implement the parallel parking function,but alsomonitor the heading angle,speed,and displacement of the car in real time during parking.The parking environment was built in the laboratory and parallel parking experiments were carried out.The experimental results show that the control system can realize the parallel parking function of arc-tangent path planning in the laboratory.Meanwhile,the experimental results are analyzed and the improvement measures are proposed.