The Study Of Driver Steering Characteristics Based On The Driving Simulator

With the continuous development of automotive technology, automotive intelligent andhuman nature design has been becoming an important concept for research. The four-wheelin-wheel motors electric vehicle provides the ideal carrier for this concept. In-wheel motorselectric vehicle has become the research and development hot topics of major automobilecompanies and research institutes in the worldwide. The paper comes from the NationalNatural Science Foundation Project that study on adaptive steering for intelligent four-wheelindependently driven in-wheel motors electric vehicle (51305190). The paper studies on thedriver’s steering characteristics. The steering characteristics control for the four-wheelindependently driven in-wheel motors electric vehicle based on the driver’s characteristics ismainly studied in the project. Driver steering characteristics classification and the referencemodel of using to control the in-wheel motors electric vehicle are studied in the paper.Based on the driving simulator, the driver steering characteristics are studied in the paper.The main contents and conclusions are described as below:(1) Vehicle driving simulator is built. Funded by the central and local governments, theresearch-based vehicle driving simulator has been built in the Liaoning University ofTechnology. Vehicle driving simulator includes a large ring screen, real vehicle cockpit,projector and the real-time simulation control system of dSPACE. The research-based vehicledriving simulator has a good openness, the research of driver’s steering characteristics can beable to meet in this driving simulator.(2) The driver’s steering characteristics are classified. Experimental conditions aredesigned based on the vehicle driving simulator. Drivers are selected to do the experiment.Different driver’s steering characteristics gain factors are collected. K-means algorithm isapplied to divide drivers’ characteristics into cautious type, average type and radical type.(3) The driver’s steering characteristics are identified. According to the test data ofclassification, models of the driver steering characteristics identification are built based on theLVQ neural networks. According to the steering characteristics gain factor, the driver’ssteering characteristics are identified. The test results show that the model has high precision.(4) Reference models of the driver steering characteristics are built. According to theclassification results of the driver’s steering characteristics, the driving simulator test isdesigned and datum of speed, steering wheel angle, yaw rate and side slip angle are collected.The driver steering reference model is established, inputs of the model are speed and steeringwheel angle and outputs of the model are yaw rate and side slip angle. The test results showthat the model can reflect the different driver’s steering characteristics, the model has highprecision and good real-time performance, the reference model can meet the needs of thesteering control. Based on the driving simulator, the driver steering characteristics are studied in the paper,this research methods has certain innovation. The research conclusions have some scientificvalue and practical significance to the vehicle’s in-wheel motors and the vehicle’s controlalgorithm.