Research On Automatic Parking Control Algorithm For Vertical Parking Space Of Automatic Driving Vehicle

The development goal of automatic parking system is the last kilometer of driverless.Parking lot provides a landing scene for automatic driving,and it is also a good entry point to realize automatic driving.Therefore,automatic parking technology has become a research hotspot.Most of the vehicle rubbing occurs in parking,and the parking process is a more complex process in the process of vehicle driving.The automatic parking system does not need the intervention of the driver,and the vehicle automatically and safely parks in the warehouse,which effectively reduces the driver’s fatigue and the probability of rubbing with the surrounding environment.With the increasing degree of vehicle automation,automatic reverse charging is a typical application scenario.It should be considered that the distance between the charging socket of the vehicle and the charging port of the charging pile should be as small as possible when the vehicle is parking.This application scenario requires higher accuracy of vertical parking.This paper takes the automatic driving vehicle as the research object,and takes the automatic reverse charging as the application scenario.Firstly,based on Ackerman’s steering principle,the vehicle kinematics model is established,and the vehicle pose is estimated based on the vehicle speed and steering wheel angle and the rear two wheel speed.The real vehicle verification scene is established,and the real vehicle integrated inertial navigation system is used to verify the accuracy of the two pose estimation algorithms.This paper analyzes the rationality of vertical parking path planning with quartic polynomial curve.According to the constraints in the process of vertical parking,it determines the minimum parking space size of single step vertical parking and limits the starting position of vertical parking,According to the constraints of equality and inequality,the coefficients of polynomial terms are obtained.In order to ensure that the vehicle runs according to the expected trajectory,two kinds of vertical parking trajectory tracking controllers based on model predictive control and pure tracking control are designed.In order to control the parking speed,a speed controller based on PI control is designed and simulated.Then,the model of Simulink and Car Sim co simulation is established to verify the two kinds of trajectory tracking controllers based on pure tracking control and model predictive control.The feasibility,stability and error value of the two kinds of trajectory tracking controllers are analyzed by using simulation data and curves.Finally,the hardware in the loop test bench is built,the hardware in the loop simulation environment is established,and the hardware in the loop test of the vertical parking system based on pure tracking trajectory tracking control is carried out.The feasibility and stability of the vertical parking algorithm are further verified by the hardware in the loop test data.