Bionic vision inspired on-road obstacle detection and tracking using radar and visual information

Inspired by the cooperation between the cone cell and the rod cell in human vision system, this paper presents a novel method for fusing millimeter wave (MMW) radar and monocular vision sensor information for on-road obstacle detection. The MMW radar performs similarly as the cone cell to indicate potential target region in the view, while the camera plays the same role as that of the rod cell to obtain precise target contour and to allow tracking over frames. Using real-world data collected from on-road vehicles, experimental evaluations have shown that, our proposed system could detect on-road obstacles with high detection rate and low false-alarm rate, yet the overall image processing time is significantly reduced compared to conventional approaches. In addition, the system has been applied on our unmanned vehicle for the past two years' “Future Challenge” autonomous vehicles competition.