Development And Experimental Verification On Decision Algorithm Of Lane Departure Warning System

Lane Departure Warning(LDW)system is such a system that alerts the driver to correct the vehicle's heading direction when the vehicle is departing from the lane due to the driver's negligence.LDW is one of the Driver Assistance(DA)Systems,aiming to improve vehicle's lateral safety and to reduce lane departure accidence.LDW uses varies of sensors to get the lane information and the vehicle's status information,and decides if the vehicle is departing from the lane,and finally publishes warning information through the means of vision,haptic and auditory.The driver will correct the vehicle's heading direction after receiving the warning information.Automobile Simulate and Control Laboratory(ASCL)of Jilin University started to undertake considerable research on LDW and proposed some lane departure decision algorithms a long time ago.They also tried to test the proposed algorithms on the Driver In Loop(DIL)bench.Based on previous achievements we undertook further research on LDW.In this paper,the following aspects are addressed:1.After fully study and analysis of the existing lane departure decision algorithms,a novel algorithm which uses the surrounding lane information to execute the safety assessment has proposed and also described the feasibility and advantages of the proposed algorithm in detail at the theory level.2.Referring to the international mainstream means of publishing warning information,three warning publish methods through vision,haptic and auditory was designed and finally carried out on the test vehicle.Choosing different method to publish warning information so as to improve the system's acceptance was undertaken.Measurement of driver's reaction time was implemented and the moment of when to publish warning information was based on driver's response characterizes.3.Data collection,departure decision and warning information publishment were fulfilled by using LabVIEW to program and modified the test vehicle to meet the system requirement.Some experiment conditions which were designed by taking the national standard into consideration was designed to test the proposed LDW.Results showed that the proposed LDW performed very well during the test.In order to illustrate the system's effectiveness from data level,the test data was recorded and analyzed off-line.Then modified and optimized the key parameter of the algorithm.After the optimization,the proposed LDW performed more precisely.