Research On Active Collision Avoidance Control Of Vehicle Based On Road Adhesion

In the wake of the progress of automobile industry and economy,traffic safety becomes the shackles that restrict the further development of the industry.As an active safety technology,active collision avoidance system has attracted growing attention owing to its excellent performance in driving safety.Active collision avoidance system includes longitudinal active collision avoidance and lateral active collision avoidance.Although the former technology is more mature and widely popularized in the market,lateral active collision avoidance can avoid collision more effectively when the main vehicle speed is high or the relative distance between the front vehicle and the main vehicle is difficult to meet the needs of longitudinal active collision avoidance.Therefore,the research on vehicle longitudinal and lateral active collision avoidance control strategy is of great significance to vehicle driving safety.Based on the school-enterprise cooperation project ’Intelligent Driving and Braking Control Algorithm Development’,this paper takes the Electro-Mechanical Braking Booster and ESP system as the controlled object,studies the longitudinal and lateral active collision avoidance control strategy including road adhesion estimation,and verifies the designed control strategy based on the Hardware-in-the-Loop platform of the braking system.The main research contents are as follows :1)Design of road adhesion coefficient recognition algorithm based on state observation Firstly,the vehicle state calculator is designed by analyzing the vehicle dynamic model,and the tire force observer is constructed based on Kalman filter theory.Secondly,a road adhesion estimation algorithm based on recursive least squares algorithm is constructed,which includes two estimation methods based on tire force model and slope method.Finally,this paper conducts simulation and verifies its recognition performance.2)Collision risk assessment and decision design.To comprehensively evaluate and analyze the collision risk of vehicles in the process of driving,this paper constructs a collision risk assessment method based on the safety distance model and the time-to-collision model,furthermore designs the decision logic of collision avoidance mode according to the principle of priority to use longitudinal active collision avoidance.Selecting a quintic polynomial as the trajectory prototype of lateral active collision avoidance,this paper introduces how to calculate the latest turning point.The trajectory planning based on quintic polynomial is designed.3)Research on longitudinal active collision avoidance control strategy.Based on the electronic mechanical braking booster system,the vehicle longitudinal active collision avoidance control strategy is studied,mainly including the algorithm design of the vehicle deceleration-loop control and the pressure-loop control.The former adopts a feedforward controller based on vehicle dynamics model and introduces a PI feedback controller.For the pressure-loop control,feedforward control based on PV characteristics and fuzzy PI feedback control are adopted.Finally,the rapid prototype verification of the pressure loop control algorithm shows that the control algorithm can meet the demand of vehicle longitudinal active collision avoidance.4)Research on lateral active collision avoidance control.The application of MPC control theory in lateral active collision avoidance is studied,and the MPC lateral dynamic control algorithm is constructed.Firstly,based on the small angle assumption and Taylor expansion,the nonlinear vehicle dynamics model is linearized.Secondly,the objective function and constraint conditions are designed,including the objective function considering trajectory tracking accuracy,control vector and its increment,and the constraint conditions of actuator characteristics and vehicle stability are comprehensively considered.Finally,the simulation comparison between the MPC controller and the preview-based lateral acceleration feedback controller is carried out,and the results show that the MPC controller can more accurately follow the desired trajectory.5)Hardware-in-the-loop experimental verification.Based on the dSPACE hardware platform and software tool chain,the hardware-in-the-loop experimental platform of braking system in the loop is constructed.Firstly,the pressure loop control algorithm is verified by rapid prototype,and the results show that the control algorithm can meet the demand of vehicle longitudinal active collision avoidance.Secondly,the hardware-in-the-loop verification of the active collision avoidance control algorithm is carried out.The experimental results show that :Firstly,the active braking control algorithm can follow the expected deceleration well and basically meet the collision avoidance requirements.Second,the decision-making of the collision avoidance model is reasonable;Thirdly,the designed active steering control algorithm can make the actual vehicle trajectory follow the ideal trajectory better in the hardware-in-the-loop experiment.Fourthly,based on the experimental results,the designed active collision avoidance algorithm can integrate the advantages of vertical and horizontal active collision avoidance to achieve better collision avoidance performance.