Research On Road Feeling Simulation Of Steer-By-Wire System Based On Steering Load Observation

With the continuous development of automobile electrification and intelligence,new requirements are put forward for the automobile chassis system in terms of execution accuracy,response speed and safety.Due to its structural characteristics and performance advantages,the wire-controlled chassis can well meet the above needs,it is regarded as an important carrier of automobile electrification and intelligence.And it has become the development trend of automobile chassis systems and has broad development prospects.Steer-By-Wire system is the core component for controlling lateral movement in wire-controlled chassis,it adopts a modular design and realizes the decoupling of steering angle and force in steering control.Compared with traditional mechanical steering systems,Steer-By-Wire system has multiple advantages and is one of the development trends for automobile steering systems in the future.Due to the elimination of the mechanical connection between the steering wheel and the steering actuator,Steer-By-Wire system cannot transmit road feedback,i.e.,road feeling,to the driver.Therefore,when someone is driving,it is necessary to simulate and generate road feeling.Based on the summary and analysis of the current research status at home and abroad regarding the above content,relying on the national key research and development program“Multi-System Efficient Integration of In-wheel Motor Action Module and Vehicle Torque Vector Distribution Technology”(Project Number: 2021YFB2500703)and the science and technology project of Jilin Provincial Department of Education,"Research on decision-making and control strategy of automatic lane change of X-by-Wire vehicle in intelligent transportation environment"(Project Number: JJKH20231148KJ),this article comprehensively considered that the steering load is the main source of road feeling transmission and therefore,studied the road feeling simulation algorithm of Steer-By-Wire system based on steering load observation.The main work contents are as follows:(1)A Steer-By-Wire system and the vehicle dynamic model were established and validated.Firstly,the dynamic models of the steering road feeling assembly and steering actuator assembly of the Steer-By-Wire system were established using MATLAB/Simulink.Secondly,an overall vehicle dynamic model without the steering system was established using Car Sim.The Steer-By-Wire system model in Simulink was embedded into the Car Sim vehicle dynamic model using co-simulation to obtain the Steer-By-Wire vehicle dynamic model.Finally,multiple typical verification schemes were designed to validate the model,and the simulation results showed that the established model can meet the requirements of subsequent research.(2)Based on the study of the steering load observation principle of Steer-By-Wire system,the observation effect and computational power requirement of different steering rack force observers were compared and analyzed.Firstly,the steering load was analyzed,and the rack force was selected to represent the steering load.On the basis of analyzing the existing rack force observers,the reduced-order linear extended state observer and second-order nonlinear sliding mode observer were designed as the research objects of the rack force observer in this paper.Secondly,an improved tracking differentiation algorithm was designed to obtain rack speed information to meet the condition of using the reduced-order linear extended state observer.Meanwhile,Particle Swarm Optimization was applied to optimize the parameters of the second-order nonlinear sliding mode observer to improve its observation accuracy.Finally,the observation accuracy was used to represent the observation effect,and the calculation time of the observer was used to represent the computational power requirement.Rack force observation experiments were carried out on both flat and uneven roads,and the observation effect and computational power requirement of four rack force observers,including the linear extended state observer,the reduced-order linear extended state observer,the second-order nonlinear sliding mode observer,and Kalman Filter,were compared and analyzed.The simulation results showed that the same rack force observer had better observation effect on flat road than on uneven road,and the computational power requirement reflected by the calculation time was less affected by the road conditions due to the fixed structure of the observer.At the same time,it also showed that the reduced-order linear extended state observer had better observation effect and appropriate computational power requirement among the four observers mentioned above.(3)A road feeling simulation algorithm based on observed rack force was researched.Firstly,we designed a vehicle speed-adaptive low-pass filter to eliminate high-frequency noise in the observed rack force.Secondly,we designed a road feeling main torque based on the observed rack force,while considering the curve-type assistive characteristics of electric power steering systems,and ensuring a reasonable trend in road feeling intensity changes.We also designed a road feeling compensation torque based on the steering system’s inertia,damping,and friction.Thirdly,we proposed a return-to-center control strategy for the return-to-center condition.This strategy uses steering wheel angle,driver torque,and its rate of change for return-to-center judgment.We designed a return-to-center control algorithm based on backstepping control and analyzed its stability.The final road feeling is a combination of the main torque,compensation torque,and active return-to-center torque.Finally,we validated the designed Steer-By-Wire system road feeling simulation algorithm through simulation.The simulation results showed that there is little difference between the road feeling effect generated based on observed rack force and those generated based on actual rack force.The designed road feeling simulation algorithm can meet the expected design requirements,and the return-to-center control strategy can accurately judge return-to-center condition and achieve good control effect.(4)A hardware-in-the-loop(HIL)test bench with a Steer-By-Wire system was constructed and was used to validate the designed rack force observer and road feeling simulation algorithm.Using Car Sim RT for real-time simulation and vehicle dynamic model,and NI PXI as the hardware controller,we respectively used the Senso Wheel steering wheel to input steering control commands,the Logitech G29 throttle and brake pedals to input longitudinal velocity control commands,and NI Veri Stand for software-hardware signal interaction configuration to build the Steer-By-Wire system hardware-in-the-loop test bench.The hardware-in-the-loop validation work was completed through tests on the central zone steering,free driving,and return-to-center.The experimental results show that the designed reduced-order linear extended state observer can effectively observe rack force;the road feeling effect based on the observed rack force is only slightly different from that based on actual rack force;and the designed road feeling simulation algorithm can provide the expected road feeling and the alignment control algorithm has a good alignment effect.