Research On Bending Mode Algorithm Of Adaptive Front-lighting System With Simulation Of Driver Preview Behavior

Adaptive Front-lighting System(AFS) could modify lighting pattern and directionautomatically on the basis of information from vehicle state and road condition, improvingroad illumination, and enhancing the safety of night driving.In recent years, the core problem of AFS research is how to make drivers obtain correcttraffic information in time from the area he pay attention to by modifying the lightingpattern actively when driving. But the existing AFS bending mode algorithms mostly swivelthe lighting pattern basing on steering wheel angle and rarely take driver preview behaviorand eye movement in different driving conditions into account.Aiming at above problem, the thesis conducts a driver preview behavior experiment,using eye tracking system, measures driver’s preview point at different vehicle speed androad curvature radius, and establishes a bending mode algorithm of AFS with the simulationof driver preview behavior, taking drivers’ preview point as input. During the research, firstly,the thesis establishes a dynamic predicting algorithm of future trajectory, according tovehicle state, drivers’ control input, and the accompanying change of vehicle future state;Secondly, the thesis analyses the mechanism of driver preview behavior, conducts aexperiment that measures driver preview distance at different speed and curvature radius,and establishes an empirical formula of driver preview time; Thirdly, the thesis takes driverpreview time as a parameter that represents driver eye movement into account, andestablishes a bending mode algorithm of AFS with the simulation of driver preview behavior;Lastly, the thesis conducts a simulation analysis, and verifies the control effects of the AFSalgorithm proposed, by means of analyzing the changing process of lighting area controlledby the algorithm during curve driving.