| As the number of cars continues to increase,traffic safety problems have become particularly prominent.Among them,traffic accidents caused by lane departure are increasing year by year.In order to reduce the occurrence of safety accidents,intelligent assisted driving technology has become a research direction.Based on the research results of lane departure control systems at home and abroad,the thesis explores and studies the lane departure avoidance system control in assisted driving technology.The main research contents are as follows:Firstly,obtain road information and identify and detect lane lines to provide a basis for subsequent vehicle departure decisions.In response to the requirement of real-time and robustness in the recognition of curve lane lines,a lane line recognition algorithm based on the least square method is proposed,which extracts the characteristic points of the curve lane line,and performs spline fitting on the extracted characteristic points,real-time detect road lane lines.Secondly,in view of the problem that the existing lane departure avoidance system cannot adapt to drivers of different driving styles,a fuzzy controller is designed to determine the road environment safety according to the radius of curvature of the lane,the standard deviation of the lateral position of the vehicle under different driving styles,and the mean value of the local lateral position.The virtual lane line of the boundary,and the vehicle dynamics safety boundary is obtained according to the vehicle 2-DOF model,so as to establish the lane departure decision criterion based on the virtual lane crossing time and the vehicle dynamics safety threshold.The decision method is used to reduce the frequency of the controller unnecessary intervention improves the acceptance of drivers with different driving styles.Then,based on the 2-DOF vehicle dynamics model,the influence of the tire cornering stiffness uncertainty boundary on the vehicle dynamics is analyzed,and the lane departure avoidance bottom layer H∞ robust controller is designed according to the tire cornering stiffness boundary.The controller does not depend on the change of the vehicle speed and realizes the decoupling of the vehicle speed.Through CarSim/Simulink co-simulation,the results show that the proposed lane departure decision criterion reduces unnecessary intervention of the controller and improves the acceptance of drivers with different driving styles.The designed controller is highly robust.Finally,on low-coefficient roads,the tire force is easily saturated,and lane departure prevention only through steering control cannot correct the vehicle in time.Therefore,the paper proposes a hierarchical control algorithm.The upper system obtains the lateral position offset of the vehicle based on the visual information of the vehicle,and calculates the expected yaw rate and the side slip angle of the center of mass through the PI controller.The steering and braking integrated MPC controller takes the desired yaw rate and the center of mass side slip angle as the tracking target,and calculates the front wheel angle and brake pressure through the tire inverse model,and then the hierarchical integrated control of the vehicle is realized.The simulation analysis show that the designed integrated controller can solve the problem of tire force saturation,thereby ensuring the stability and safety of the vehicle.Figure[55]table[7]reference[84]... |
