The Reserch Of Driving Behavior’s Individual Operation Pattern Modeling

In the closed-loop traffic system composed of human-car-road,the human factor has the highest correlation with the probability of occurrence of traffic accident.Based on this,in order to reduce the traffic accident rates,the Driving Safety Assistant System(DSAS)has been the key research content in the field of intelligent vehicle to coordinate the relationship between people,car and road.At present,most of existing DSAS take the same control strategy and parameters design for every driver,so they can’t meet the driver’s individual needs.If the DSAS can automatically complete the learning of the driver’s driving behavior and vehicle operating status information,then use the learning results for the designing of DSAS’s secondary control strategy,we can improve the level of DSAS’s intellegence and individuation.In order to achieve this goal,solving the problem of modeling driver’s individual operation pattern and driving behavior recognition is one of the key technologies.Based on the above problems,the main research contents of this paper are as follows:(1)By capturing the drving behavior manipulation data,we analysed the drivers’ individual characteristics with a method of combining the data and mechanism and then study the method of extracting driver’s manipulation action node to model the driving behavior manipulation pattern.(2)Based on the the directed digraphs theory,we studied the method of modeling driver’s individual driving behavior manipulation pattern after extracting accurately the driver’s manipulation action node.(3)By using the characteristics of ART2A-E adaptive resonant neural network,such as good stability,strong plasticity and self-learning on line,we studied the method of driving behavior recognition on the basis of ART2A-E algorithm and the reverse mapping of manipulation pattern.The results of the study show that:(1)Based on the analysis of the driver’s individual characteristics,the method of controling the manipulating nodes’ generating by setting the dynamic threshold,can effectively realize the precise extraction of the manipulate action node.(2)According to the manipulate action’s change law of driver in the implementation of a specific driving behavior,a method of individual driving behavior operation modeling was put forward based on the directed digraphs.The test results show that the method can effectively reflect the individual characteristics of driver.(3)A method with ART2A-E algorithm can effectively realize the driving behavior recognition based on the reverse mapping of individual driving behavior manipulate pattern and it can provide technical supports to solve the problem of the driving manipulate pattern modeling for the individual and intellectual driving safety assistance systems.