The Uncertainty Modeling And Simulation Of Motor Vehicle Driver Behavior

With the development of modern society, the vehicles have already been the indispensible transportation. However, with the increasement of the vehicle amount, the corresponding problems have also appeared and the road accidents occurrence rate has increased. In the road accidents, drivers play a very important role, so it is necessary to improve drivers’ driving abilities. Based on the above elements, the Driver Assistance System. It can provide guide and support for drivers. When potential danger appears, it can provide warning message and the auxiliary control to the vehicles for drivers. However, as for the driving characters are absolutely different among the drivers, it is very hard for the design of the assistant seats. Make the system adapt the driver differences under the precondition of safe are to be a key problem of system design.According to the above problems, the paper mainly solves the following problems:(1) Through collecting and analyzing driving behavior and driving operation dates, study the relation between driving behavior and driving operation with PCA.(2) When an individual completes a specific driving behavior, its driving operations have some cohesion, timing and individuality. The driving operation model based on directed digraphs was set up in this paper.(3) Based on the extracting driving characteristics from PCA method, design a neural network aimed at main characteristics. The method can distinguish the driving behavior reversely. And it provides technology support for the design of the Driver Assistance System based on driving behavior and driving operation characteristics.The result shows:(1) PCA method can make sure the key driving operation actions for the specific driving behavior.(2) It is possible to build the cohesion driving operation mode by MATLAB simulink software. And it can also help build automotive driving behavior mode, being good for the building of the self-adaption system. (3) The driving operation distinguishment method has a high accuracy rate and a low failure rate. The method can be an effective solve scheme in the driving behavior distinguishment.