Study On Analysis Of Vehicles’ Anti-rollover Stability And Early Warning Of The Vehicles’ Rollover

The main content of this paper is to design an early warning system for preventing vehicles from rolling over based on wheel force.Due to the vehicle is a system of high unlinearity,quantities of factors will have an effect on the rollover of the vehicle,including the lateral load transfer rate,longitudinal speed,steering angle,and yawing moment,which increases the complexity of the rollover procedure.Several indicators have been formulated on the basis of wheel force in this paper,and the appropriate prewarning strategies in different operating conditions have been set by analysing the changing patterns of indicators.The vehicle rollover prewarning system is designed in terms of three aspects in this paper,which includes real-time preprocessing of wheel force data,formulation of indicators of vehicle’s stability,high relability prewarning model in complex operating conditions.The following are the main contents of this paper:1.Real-time preprocessing method for wheel force data.As noise exits in the measurement of wheel force,real-time adaptive filtering is required to be implemented in order to improve the accuracy of wheel force in different operating conditions.In this paper,filters of different types have been evaluated and and appropriate one has been selected to design a real-time adaptive filter,which can use different schemes adpatively for filtering under situations with different wheel dynamics.2.The anti-rollover evaluation and early warning system platform is designed.Based on the WFT six-dimensional wheel force sensor,inertial measurement sensor,and steering wheel angle sensor,this chapter designs the anti-rollover evaluation and early warning system platform,constructs the platform sensing layer,data processing layer,and application layer,and describes the functions and connections between each layer.3.Formulation of the indicators for vehicle’s stability based on wheel force data.Two indicators based on wheel force data were formulated,including lateral load transfer rate and yawing moment.At the same time,longitudinal speed,steering angle,yawing velocity were selected for assisting the early warning of vehicles and effectiveness of each indicator has been evaluated.4.Construction of the model for prewarning system in complex operating conditions.In this paper,the rollover prewarning system is designed based on the thorem of support vector machine,and the model can predict the vehicle’s stability in the future based on the history data of vehicles.The model can predict the time to rollover of a vehicle precisely in J-Turn steering condition,Fishhook steering condition,90 degree’s angle steering condition,accident avoidance in high-speed condition,and synthetical operating condition.