Research On Man-machine Torque Cooperative Steering Control Based On Driving State Prediction

In recent years,traditional vehicles are facing the transition to intelligent vehicles.As a key technology of vehicle intelligence,man-machine cooperation technology has been widely studied,which is also a necessary stage for vehicles to realize unmanned driving.Man-machine collaboration requires the comprehensive judgment of the driver and the road condition,and the cooperation between the automatic control system and the driver to ensure the safe operation of the vehicle and the driver’s comfort.How to reduce the driver’s driving burden on the basis of ensuring safety in the steering process and how to improve the man-machine consistency rate to reduce the interference of the control system to the driver is the main research problem of this paper.In order to make the driver have a more real driving feedback feeling,the vehicle control is carried out by means of torque interaction,that is,the torque is used as the control input.At the same time,the driver state is considered in the design of the cooperative controller to enhance the controller’s perception of the driver.The two-point preview driver model is used to establish the state space equation of human-vehicle-road system for controller design.The validity of the model is verified.The driving characteristics of the two-point preview driver model in different parameters,different preview distances and different speeds are analyzed.Aiming at the problem of improving driving safety and reducing driving burden,a human-machine torque assisted steering control scheme based on driving state prediction is proposed.The model predictive control method is used to predict the driving state for a period of time in the future,so as to increase the information interaction between human and machine.With the minimum lateral position deviation as the control target,the assisted torque to the driver is determined according to the prediction information.Due to the different types of drivers,the controller can adapt to the driving styles of different drivers by changing the parameters of the driver model,so as to achieve personalized driving assistance.At the same time,it has better control effect on drivers with different speed and driving ability.To make the vehicle safe and reduce the driver’s driving burden.The effectiveness of the control scheme is verified by simulation experiments.Aiming at the problem of how to improve human-machine consistency,the fuzzy method is proposed to take human-machine state(consistency degree,inhibition degree,conflict degree)and road danger degree as the input of fuzzy control,output assisted weight coefficient,and improve human-machine consistency rate by changing assisted weight coefficient in real time.Aiming at the problem of how to further reduce the driver’s driving burden,the method of combining the proportion of driving task and the driving burden function is adopted to guide the driver,which further reduces the driving burden,but the consistency rate is reduced.Finally,combining the above two methods,the human-machine torque collaborative control scheme based on driving state prediction is proposed.Through the cooperation between the driver and the controller,that is,the controller can improve the human-machine consistency rate by changing the assisted weight to adapt to the driver’s operation,and the driver can operate according to the guidance of the controller to further reduce the driving burden.Compared with the human-machine torque assisted steering control scheme based on driving state prediction,the human-machine consistency rate is improved and the driver’s driving burden is further reduced on the basis of ensuring safety.The effectiveness of the control scheme is verified by simulation experiments.