| At present,many achievements have been made in the research of unmanned intelligent vehicles,but there are still some problems in the research of speed planning and trajectory tracking control,mainly including the following two points: first,in the field of vehicle speed planning,the existing speed planning of intelligent vehicles does not consider the amount of vehicle state,resulting in unreasonable planning speed,which is not conducive to improving the accuracy of subsequent trajectory tracking control.Second,in the field of vehicle trajectory Second,in the field of vehicle trajectory tracking control,the traditional Model Predictive Control(MPC)does not consider the road and planned speed information further outside the predicted time domain,resulting in poor tracking effect and ride comfort to be improved.In order to solve the above problems,this paper designs a vehicle speed planning algorithm based on variable domain fuzzy rules and a trajectory tracking controller based on Piecewise Preview Model Predictive Control(PPMPC).The details are as follows.(1)A vehicle-road model coupled with a vehicle dynamics model and a road error model is established.Firstly,the mathematical model of vehicle dynamics in the transverse and longitudinal directions is established.Secondly,the road error model considering the deviation of heading angle and vehicle lateral deviation is established.Finally,the vehicle-road model for subsequent tracking control is obtained by coupling the two models.(2)A vehicle longitudinal speed planning algorithm based on variable domain fuzzy rules is designed.Using the fuzzy rule,the variable domain speed planning algorithm is designed with the road friction coefficient and road curvature as the input,the vehicle longitudinal speed as the output,and the vehicle lateral deviation as the input of the variable domain speed scaling factor.The simulation results show that,compared with the traditional fuzzy rule speed planning algorithm without considering the speed variational domain,the improved speed planning method can plan a lower vehicle longitudinal speed for the case of larger vehicle lateral deviation to better match the human driver experience,and can quickly reduce the tracking lateral deviation in the subsequent tracking control,which is helpful to improve the tracking control accuracy.(3)The PPMPC is designed for intelligent vehicle trajectory tracking control.Unlike the traditional MPC,the target state of the PPMPC cost function is calculated by dividing the target state into two parts: the first part is based on the traditional MPC target state selection method,while the second part of the target state is innovatively combined with further road information and planning speed information.The simulation results show that PPMPC can improve the tracking accuracy and control stability of vehicle speed compared with traditional MPC.(4)Speed replanning triggered by obstacle avoidance safety distance is designed.Reasonable speed planning results are the basis to ensure the tracking control to achieve safe obstacle avoidance.When the front and rear vehicle distance is less than the obstacle avoidance safety distance,the rear vehicle needs to carry out speed replanning to reduce the rear vehicle speed to prevent the rear vehicle from colliding with the cruising vehicle,and on this basis,the designed PPMPC algorithm is used to complete the trajectory tracking control so as to realize the obstacle avoidance function.The simulation results show that the vehicle can reasonably plan the speed and achieve tracking control,so as to complete obstacle avoidance.In this paper,the speed planning of intelligent vehicles and the trajectory tracking control are studied,and the speed planning and tracking control based on segmental pre-scanning considering the variable theory domain,which improves the tracking accuracy of vehicle speed and the stability of control,and is of great significance to the realization of safe automatic driving. |
PDF Full Text Request