Driver lane changing behavior analysis based on parallel Bayesian networks

Driver behavior model is one of the key technologies for the driver assistance and safety system which can provide useful priori knowledge for detecting the abnormal behaviors effectively. The paper introduces the Driver behavior model which is established by the parallel Bayesian networks with steering angles and their difference and the final status of driver behavior is decided by the largest probability of the each status during the time of lane changing. In addition, the transition time of the status is automatic tagged by the multi-variable Gaussian model of steering angles and their difference. Finally, the parallel Bayesian network model is compared with the Gaussian Bayesian Network of the steering angles. The state transition diagram of driver behavior and the analysis of experiment results are given.