An automatic extrinsic parameter calibration method for camera-on-vehicle on structured road

This paper proposes a novel self-calibration method for getting the extrinsic parameters of the camera on the autonomous vehicle. Basing on the model of the highway and the extrinsic parameters of the camera, the lane marking can be described parametrically in the vehicle coordinate system. The position and headings of the vehicle can be measured by itself in real time. Thus the Extended Kalman Filter can be applied to the kinematical model of the vehicle. By this means, it is possible to refine the original rough extrinsic parameters of the camera with an iterative way. Experiment results show that this method has a good performance in the structural scenery on highway.