INTEGRATED MULTI-SENSOR SYSTEM: A TOOL FOR INVESTIGATING APPROACHES FOR TRANSIT FRONTAL COLLISION MITIGATION

The paper discusses the University of California's PATH program development of a frontal collision warning system (CWS) for transit buses. Multiple sensors including two microwave radars, one laser radar, five ultrasonic sensors, four cameras, one GPS receiver as well as accelerometer, speedometer, steering angle sensor are already installed on a SamTrans transit bus to collect real life data. Sensors were calibrated and aligned to the system dimensionality. A software tool was developed to integrate multiple sensors and play back collected data. An algorithm framework was proposed to fuse multi-sensor data for collision warning. A linear warning algorithm has been developed. This paper describes the integration of multiple sensors as a primary step to develop a CWS.