Research On Skid Resistance Of Asphalt Pavement And Measurement Requirements For Autonomous Vehicle During Braking Process

As a development trend of future transportation,autonomous vehicle has many advantages such as reducing traffic accidents and relieving traffic pressure compared with those traditional manned vehicles.However,there is large difference in braking model between autonomous vehicles and traditional ones.The existing traditional vehicles rely on driver judgement to take appropriate braking measures.While the braking measures of autonomous vehicles mainly depend on the empirical model considering the wheel-road coordination.The current braking model for autonomous vehicles ignores the influence of the road texture characteristics under different road conditions.In practice,the texture characteristics of road surface directly relates to its skid resistance.Good anti-sliding performance of road surface can provide sufficient friction for vehicles at high speed to ensure driving safety.In addition,the interaction between the tire and the road surface straightly affects braking performance of the vehicle during braking process,particularly on wet road surface.In order to achieve maximum braking efficiency,the traditional ABS braking principle mainly controls the tire slip rate within the optimal range(between 15% ? 20%)neglecting the strong nonlinearity of vehicle dynamics,also the diversity in vehicle braking system and uncertainty in road surface adhesion characteristics.As a result,it cannot adapt to the variation in skid resistance of road surface.In view of the above,combining the technology characteristics of autonomous vehicles and the anti-sliding mechanism of road surfac,the skid resistance of asphalt pavemenrt under the tire-pavement interaction was analysed by ABAQUS in this study.Then,the braking stability of autonomous vehicles in different braking scenarios and the perception requirements for skid resistance of asphalt pavement were both investigated based on the Car Sim/Simulink co-simulation.Furthermore,an evaluation system for skid resistance of asphalt pavement during braking process of autonomous vehicles with considering driving comfort was also established.According to the frictional mechanism between rubber tire and rough road surface,the typical prediction models for pavement friction coefficient were analysed.The Persson viscous friction model was determined to calculate the friction coefficient.Thereafter,the automatic close-range photogrammetry(ACRP)system,which was constructed based on the principle of close-range photogrammetry,was uesed to collect the real-time image of asphalt pavement texture automatically.Then,the three-dimensional(3D)model for the textured asphalt pavement was rebuilded to extract the 3D coordinate datas for road surface elevation points.Furthermore,the power spectral density curves under dry and wet road conditions were drawed by MATLAB program to calculate the dynamic friction coefficient of road surface with the Persson friction model.The results indicate that the friction coefficient curves for the three types of asphalt pavement under dry and wet conditions tend to be consistent,and the friction coefficient between the tire and the road surface decreases with increasing the driving speed.The friction coefficient of asphalt pavement linearly reduces between 0 km/h and 10 km/h.Moreover,when the speed exceeds 40 km/h,the friction coefficient curve of asphalt pavement tends to be flat,indicating that the friction force mostly comes from the hysteresis force at a high speed.In addition,the on-site DFT test was conducted to validate the the accuracy of the theoretical calucation results.Furthermore,the actual friction coefficient database of asphalt pavement was developped to provide skid resistance parameters for the accurate braking behavior of the autonomous vehicles.As the driving speed increases,the adhesion between the tire and the road surface drops significantly,it results in the tires slippage during braking.Moreover,there is a distinct difference in braking characteristics and sensing requirements between autonomous vehicles and conventional vehicles.Therefore,combined with the pavement texture information data recognized by the ACRP system in real-time,the ABAQUS finite element program was apllied to build a contact model of the textured pavement and the tire based on the coupled EulerLagrange(CEL)method.Then,the interaction mechanism between the tire and the asphalt pavement was analysed by considering the fractal characteristics of asphalt pavement.Moreover,the tire-pavement contact model was develpped to investigate the variation laws of pavement adhesion characteristics for different influence factors.The significant parameters of the real-time identified road surface texture required by autonomouis vehicles were determined based on the orthogonal experimental design theory.Finally,the curves of the peak adhesion coefficient of asphalt pavements were drawed under dry road condition.According to the established tire-pavemnt contact model,the Euler fluid model was builded firstly.Then,the finite element coupling model of tire-fluid-pavement three phases was established to simulate tire hydroplaning process using the CEL method.Based on the simulation of the tire hydroplaning speed,the accuracy of the built coupling model was determined by comparing with the American NASA empirical formula.Finally,a mathematical model for the critical hydroplaning speed of the tire was developped by the regression analysis.Within considering the influencing factors(such as travel speed,macrotexture parameters and rainfall,et al.)of water film thickness for different types of asphalt pavement and the theory of elastic hydrodynamic lubrication,the longitudinal adhesion of asphalt pavement under wet road condition was obtained.The change rule of the influence of water film on asphalt pavement adhesion were analysed by the orthogonal experimental design theory.The curves of the pavement peak adhesion coefficient varied with vehicle speed under different pavement conditions were drawed according to the simulation results.With obtaining the peak adhesion coefficient of the road surface,the perception requirements of asphalt pavement information for autonomous vehicles during braking process were analysed according to the vehicle braking characteristics.Then,the autonomous vehicles braking system was estabished considering the influence of the road surface adhesion characteristics on the braking behavior.Considering the strong nonlinearity of vehicle dynamics during braking process,the formula for calculating the expected brake pressure reached by the vehicle was developped.Then,the reverse braking model for autonomous vehicles was estabished in the MATLAB/Simulink module.In addition,using the vehicle model parameters provided in Car Sim and according to the Car Sim/Simulink co-simulation,the braking performance of autonomous vehicles was analysed.Furthermore,comparied with the braking performance of traditional ABS system,the feasibility of the built braking system for autonomous vehicles was verified.Based on the above built braking control model of the autonomous vehicles,the dynamic simulation model was established by the co-simulation of Car Sim and Simulink.The driving characteristics of autonomous vehicles was considered under typical braking scenarios,such as normal and emergency braking,steering braking on curved section(including superrelation)and braking on slope road.Thereafter,the braking deceleration,the lateral offset distance and the braking distance were selected as the evaluation index of braking performance to analyze the influence of the demand parameters of the road surface on the braking stability of the vehicle.According to the simulation results,the orthogonal test design theory was used to analyze the influence of perception parameters on braking stability of autonomous vehicles.In addition,under typical braking scenarios and different road conditions,the braking strategies for autonomous vehicles were both investigated based on the real-time adhesion characteristics between the tire and the asphalt pavement.Comfort is the key to show the advantage of autonomous vehicles with respect to the braking performance.Based on the international standard ISO,the calculation model of driving comfort index with weighted acceleration as a variable was simplified and the evaluation levels for driving comfort were determined.Then,based on the braking characteristics of autonomous vehicles under different braking scenarios,the variation characteristics of the vehicle speed collected was analysed and the weighted acceleration was corresponding calculated to obtain the comfort index for various braking conditions.Moreover,the evaluation models for braking comfort of autonomous vehicles under different braking scenarios were developped based on the Logistic regression theory.The change law of driving comfort and the requirements for road environment parameters were both analysed under different braking scenarios.Thereby,the braking strategies for improving comfort of autonomous vehicles were determined in different road environments.Based on the comfort evaluation method,an evaluation system for braking comfort of the autonomous vehicles was established,which can provide theoretical guidance for safety braking and the accurate identification of sensing systems for autonomous vehicles in complex road environments in the future.