Vehicle Roll Angle Estimation On Rough Road

Intelligence is one of the trends in the future of automotive technology.Vehicle roll angle is an important parameter in the high-precision control to improve the driving safety and ride comfort of the smart car.Limited by the cost and technical conditions,the current vehicle roll angle cannot be directly measured by vehicle on-board sensors and needs to be obtained through an estimation algorithm.This paper will consider the random characteristics of the road excitation and the nonlinear dynamic characteristics of the suspension,and study the high-precision vehicle roll angle estimation method based on low-cost sensors,which has important theoretical significance and practical value.Road excitation causes the vehicle to roll by affecting the vehicle suspension force.For this reason,this paper firstly establishes the time domain model of road excitation,analyzes the dynamic characteristics of the suspension spring and damping under the influence of road excitation,and then establishes a vehicle rolling dynamics model based on the structural of suspension.This paper uses the high-precision vehicle dynamics simulation software veDYNA to verify the effectiveness of the model.The road roughness is the disturbance input of the vehicle roll dynamics system,but it cannot be directly measured by the onboard sensors.Therefore,this paper designs a road roughness estimator based on the Q-parameters method,expounds the algorithm structure and the estimation principle,and gives an adaptive method for the estimator parameters.The method of selecting sampling time based on Q-parameters method is given in this paper,and the validity of the estimator is verified in the quarter vehicle model and the vehicle model respectively.On this basis,the state observer for vehicle roll angle was designed for the known and unknown conditions of road roughness.First,the estimated road excitation is taken as the input of the roll angle state observer considering the road surface excitation.Then,considering the segmentation characteristics of the vehicle rolling model,the vehicle roll angle PWA estimator is designed based on the global Lyapunov function of the switch system.This paper demonstrates the effectiveness of the PWA estimator in estimating the vehicle roll angle through simulation of multiple conditions on C and D roads and bulging roads.Then,taking into account the difficulty of obtaining road surface excitation information,this paper designs a sliding mode observer for vehicle roll angle with unknown road surface excitation information,and compares the estimation results with the PWA estimator to verify the effectiveness of the sliding mode observer.Therefore,different state observer design methods are provided for vehicle roll angle estimation.At the same time,the simulation results also reflect the necessity of estimating the excitation of the road surface and considering the dynamics of the suspension when observing the roll state caused by the unevenness of the road surface.