Research On Steering Right Switching Control For Lane Centering Assistance

With the continuous development of autonomous driving,advanced driving assistance and intelligent chassis concepts are gradually recognized.As a part of advanced driving assistance,lane centering systems can effectively improve the safety of vehicle driving,attracting attention.This article takes the lane centering system as the research object,and conducts research on the realization of lane centering function,steering wheel position following control,driving right switching process,driving intention recognition,man-machine game,and driving experience.Firstly,the control of three operating modes of the steering by wire road feel simulation system is implemented,which can provide a steering by wire road feel simulation control strategy,a steering wheel angle interface control strategy,and a virtual steering wheel torque interface control strategy.The steering by wire simulation control strategy is based on the traditional EPS strategy,using the reconfiguration method to design the steering load,applying a compensation strategy to the steering system,and designing simulation conditions to conduct joint simulation and test bench validation of the steering by wire road feel simulation strategy.The steering wheel angle interface control strategy is used for Level L3 and above ADAS systems,using feedforward plus motor three closed-loop feedback control,and has been validated on a test bench for emergency steering conditions and normal driving conditions.The virtual steering wheel torque interface control strategy is used for L1 and L2 level ADAS systems.The problem that the steering wheel torque gradient may increase significantly under emergency takeover conditions by the driver during L2 level ADAS system application is considered.The motor command torque is converted into a virtual steering wheel torque using EPS nonlinear assist characteristics,and its amplitude is limited to 3 Nm,achieving the superposition between the driver steering torque and the ADAS system control torque,Thereby transforming the position interface into a torque interface,making it easier for the driver to take over emergency and providing a comfortable driving experience.Secondly,a feedforward and feedback control strategy based on optimal preview is designed to achieve the lane centering function.The least square method is used to fit the output coordinate points of Carsim to obtain the lane line equation,and the preview point information in front of the vehicle is calculated based on the lane line equation.Considering the limitations of single point preview,a multi-point preview method is used to adjust the weight parameters of different preview points based on information such as vehicle speed and curvature.A fuzzy controller is designed for the course angle feedback control to achieve the lane centering function.Based on the lane keeping assistant standard design conditions,simulation verification was conducted,and the preview time was adaptively modified from two aspects: vehicle speed and road curvature.Thirdly,a vehicle driving right switching strategy is designed for lane centering conditions,which can achieve automatic intervention of the lane centering system compared to other switching strategies.Analyze the state decision-making of the lane centering system,complete the architecture design of the system state decision-making and top-level state condition determination,and use finite state machines to conduct state management for the lane centering system.Aiming at the intervention process of the lane centering system,a pilot coefficient is used to identify the driver’s intention to turn,realizing man-machine game and man-machine co driving.For the intervention of the large offset lane centering system,a traction method is designed to avoid vehicle instability due to large control effects by reconstructing the virtual lane centerline.For the exit process of the lane centering system,the steering wheel torque is used to determine the driver’s intention to take over,and to control the torque disappearance time and torque change form of the lane centering system.The steering driving right switch is changed from the high and low jump mode to the time domain variable weight,realizing the flexible takeover of the driver,and the integral separation method is used to eliminate the integral link in the control strategy after taking over.Finally,in order to verify the effectiveness of the designed control strategy,a vehicle chassis domain control HIL test rig was established to verify the designed control strategy.Design straight road conditions and curve road conditions for lane centering algorithm;For the driving right switching strategy,a straight line overtaking and lane changing scenario,a lane line distortion scenario,and a large offset intervention scenario are designed.Experimental results verify the effectiveness of the developed control strategy.