Road Type Identification And Peak Adhesion Coefficient Estimation Under Complex Vehicle Excitation Conditions

The estimation of peak road adhesion coefficient is of great importance to active vehicle safety control.Most of the existing adhesion coefficient identification algorithms are based on the assumptions of standard typical pavement,but when the road surface characteristics change randomly,the theoretical values of curves of actual pavement and typical pavement are not identical.At this time,the existing road surface identification methods and estimation algorithms are difficult to get the true peak adhesion coefficient directly.In this paper,based on the analysis of similarity characteristics of typical road surface standard curve,road surface identification algorithm and adhesion coefficient estimation algorithm are improved.And a recognition algorithm based on characteristic value of typical road surface attribution domain and dimensionless estimation algorithm are proposed.The combination of the two methods is extended to estimate the comprehensive peak adhesion coefficient of three-dimensional pavement.The accuracy and feasibility of the estimation algorithm are verified by simulation and experiment.Firstly,according to the equal ratio characteristics between the six typical road surface standard curves,the typical road surface attribution degree area is set.And two road surface identification algorithms are proposed based on the attribution degree area road surface characteristic value identification and based on the peak road surface adhesion coefficient identification.The simulation results show that under the same conditions,the road surface identification algorithm based on the attribution domain road surface characteristic value has fast recognition speed and small hysteresis.Secondly,a dimensionless peak adhesion coefficient estimation strategy is designed to estimate the peak adhesion coefficient of any road surface in a typical road attribution domain.The strategy achieves dimensionless tire model by establishing the mapping relationship between tire force and road adhesion coefficient.The dimensionless Lu Gre tire model is combined with extended Kalman filter to estimate the peak road adhesion coefficient.The strategy of non-quantitative estimation has the following characteristics:(1)It is inclusive to tire models and can realize dimensionless to most tire models(Dugoff,MF,Lu Gre,etc.).(2)Highly sensitive to road excitation.Tyre slip rate is difficult to reach the optimal slip range under general braking conditions.However,due to the high excitation sensitivity,the estimation results are still accurate and timely.Simulation and test verify the feasibility of the strategy without quantification.Thirdly,a method of estimating peak adhesion coefficient by fuzzy adaptive synthesis is designed.Firstly,the multi-point contact vertical model is established as the tire road contact model under three-dimensional road surface excitation.Then the dimensionless estimation strategy is extended longitudinally and laterally to obtain the longitudinal and lateral peak road adhesion coefficient.Four kinds of fuzzy rules are designed according to the magnitude of longitudinal/lateral excitation.And the limit reference values of road surface adhesion coefficient for longitudinal/lateral utilization under actual conditions are given,which provides quantitative parameter basis for comprehensive active control of vehicle longitudinal,lateral and longitudinal/side.Longitudinal/lateral road adhesion coefficient limit value fusion is used to estimate the peak road adhesion coefficient under complex excitation of three-dimensional road surface.The feasibility and accuracy of the fuzzy adaptive comprehensive peak adhesion coefficient estimation method are verified by simulation and experiment.Finally,an ABS sliding mode control strategy based on the estimation of the comprehensive peak adhesion coefficient of the whole road surface is proposed by combining the fuzzy adaptive comprehensive peak adhesion coefficient estimation algorithm.And the optimal slip rate controller is designed to make the wheel slip rate control more close to the optimal slip rate.The simulation results show that the ABS sliding mode control strategy based on the estimation of the comprehensive peak adhesion coefficient of the whole road surface has better effect than ABS logic threshold control and ABS sliding mode variable structure braking control.Under the same conditions,the braking distance and braking time are improved by 20.84% and22.5% respectively compared with traditional ABS control.