A Driving Behavior Detection Based on a Zigbee Network for Moving Vehicles

Abstract

In this paper, a scheme of moving-vehicles behavior detection based on a Zigbee network is proposed. Three-axis accelerometers are installed on vehicles to capture the moving vehicle postures. A fuzzy inference system is developed to infer the six basic states of vehicle posture, such as normal driving, left/right turning, departure, accelerate, braking and bumping. Based on the recognition of vehicle postures, the dangerous driving behaviors of vehicle such as serpentuate will be detected. In this paper, the design and development of hardware, vehicle posture measurement and dangerous driving behavior inferences are presented and realized. Additionally, an Android APP is developed to offer human-machine interface. The detection results and GPS information are showed in this developed system. The system sends message to related user if dangerous driving behavior is detected. The detected data is stored to cloud for further application.