A Study of BP Neural Network for Bus Unsafe Driving Behavior

Monitor for safe operation of the bus is one of the main content of the intelligent transportation system (ITS). Among them, the real-time safety monitoring for driving behavior is a hotspot of research in this field, based on the traditional monitoring means such as video, thermal infrared, there exist some problems, especially under the condition of insufficient light at night and the car crowded conditions effect is poor. In this paper based on driving behavior of active safety monitoring and early warning method, through the vehicle attitude sensor real-time acquisition vehicle running posture, suggesting driving behavior, when found unsafe for early warning, designed the corresponding BP neural network algorithm. The experiment result compared with the results of video footage, and the absolute value of error is less than 8%.