Study On Vehicle Driving State Estimation Algorithm Based On Information Fusion Technology

Vehicle active safety control system can effectively improve the vehicle handing stability,and avoid the happening of traffic accident. the real-time and accurate to obtain the vehiclestate is a prerequisite for automotive active safety control. the cost is high for speed sensorand gyroscope at present. and directly used to test the speed and sideslip angle and otherparameters will increase the cost of the car. it is not conducive to the popularization of activesafety control system. therefore how to use low cost sensors based on information fusiontechnology to estimate vehicle driving state has become a hotspot for vehicle active safetycontrol. this paper study is based on the project supported by the National Natural ScienceFoundation (51305190) launches estimation for vehicle driving state. at frist, the three degreeof freedom nonlinear vehicle with Dugoff tire model are established in the paper. and throughthe information fusion of longitudinal acceleration lateral acceleration, yaw rate, steeringangle and wheel speed low cost sensor. the state variables is estimated based on CubatureKalman Filter theory, such as the longitudinal velocity, lateral velocity and side slip angle.andthen Considering the effect of road adhesion coefficient to vehicle driving state. The vehicledriving state and road adhesion coefficient estimator are established, and realize accuratelyestimate the vehicle driving state and road adhesion coefficient. then considering the effect ofvehicles mass parameters, the vehicle centroid position parameters and the inertia parametersin the vehicle driving. therefore the vehicle driving state and parameters estimator is designedbased on Triple Cubature Kalman Filter. through the Mutual feedback and correction ofvehicle speed estimator, road adhesion coefficient and vehicle parameter, and realize theaccurately estimate the vehicle driving state and parameters. at last the driving simulatorexperiments in the loop and CarSim and Simulink co-simulation are applied to verify thevehicle driving state and parameter estimation algorithm based on Triple Cubature KalmanFilter which is put forward in this paper. the results show that: the estimation algorithm canaccurately estimate vehicle driving state.