Control method of intelligent automobile under variable adhesion coefficient repetitive track

Abstract

The invention disclose a control method of tan intelligent automobile under a variable adhesion coefficient repetitive track, which changes the prediction range according to the adhesion coefficient change probability on the basis of a traditional MPC single prediction range so as to improve the emergency capacity of the intelligent automobile. According to the invention, an iterative learning control method is adopted to track the path, the tracking precision of the path is improved, and the driving stability of the intelligent automobile is improved in combination with a yaw stability controller. The MPC controller calls a corresponding control strategy which may occur in advance according to information provided by an environment sensing module, wherein the corresponding control strategy comprises a corresponding prediction time domain and a constraint set; afterwards, iterative learning control is used as a method for determining correct steering input of transient driving, the tracking performance is improved through multiple iterations, and the driving stability of the intelligent automobile is improved in combination with the yaw stability controller.

Description

technical field [0001] The invention belongs to the field of intelligent automobile control, and in particular relates to a control method for an intelligent automobile under the repetitive track of variable adhesion coefficient. Background technique [0002] With the rapid development of computer information processing technology, intelligent driving has become the development direction of the automobile industry. Traditional MPC focuses on improving the accuracy of path tracking, but the accuracy will often decrease when driving on roads with variable adhesion coefficients, and responds only when the adhesion coefficient changes, making it impossible for unmanned vehicles to accurately track expectations path, this is because the wheel force tends to be saturated from high adhesion coefficient to low adhesion coefficient, and when an emergency occurs, the steering but the tires cannot provide more lateral force, so the vehicle responds to the emergency prone to instabilit...

AI Technical Summary

Problems solved by technology

Traditional MPC focuses on improving the accuracy of path tracking, but the accuracy will often decrease when driving on roads with variable adhesion coefficients, and responds only when the adhesion coefficient changes, making it impossible for unmanned vehicles to accurately track expectations path, this is because the wheel force tends to be saturated from high adhesion coefficient to low adhesion coefficient, and when an emergency occurs, the steering but the tires cannot provide more lateral force, so the vehicle responds to the emergency prone to instability

Therefore, the traditional single prediction range algorithm cannot solve the control requirements and control functions of the vehicle under different emergencies, and the traditional controller is only based on the general physical model and uses the current error to control the driving trajectory, and does not make full use of historical information to improve Driving speed and trajectory tracking accuracy

Images