Abstract:

 A time to collision (TTC) warning method was proposed considering road adhesion factor and vehicle driving state to effectively avoid collision accidents and improve the accuracy and efficiency of forward collision pre-warning time (CPWT). An improved dual-exponential model was used to obtain the vehicle adhesion coefficient under different road conditions. A CPWT model was constructed considering road adhesion factor, according to the different driving states of the front vehicle, such as static, uniform speed, uniform acceleration, and variable acceleration linear motion. The CPWTs of car-following forward were simulated by using MATLAB/Simulink. The CPWT algorithm was used to analyze the CPWT for the front vehicle and the following vehicle driving on the snowy roads, wet roads and dry roads and in different driving states. The results show that the CPWT model is more reasonable and effective compared with other collision pre-warning methods. Therefore, the CPWT method can provide a theoretical support for automatic driving and vehicle warning.

Key words: vehicle active safety , vehicle collision warning ,  collision pre-warning time (CPWT) ,  time to collision (TTC) ,  adhesion coefficient of pavement , driving state