Kalman filter based on-line calibration of laser scanner for vehicle navigation

In this paper, we address the calibration of laser scanner attached to ground vehicle. The purpose of this method is to estimate the laser scanner position and attitude in body coordinate from laser scanner measurement and unknown landmarks position. In order to realize automated calibration, we simultaneously estimate the sensor calibration error and landmarks position using Extended Kalman filter. The calibration is performed based on the relationship between global and sensor coordinates of landmarks position. This method can be applied in unknown place without positioning of landmarks. To evaluate and verify the accuracy of proposed method, we perform the simulation in Matlab. Also, we analyze the calibration results with vehicle movement and number of landmarks to find more accurate calibration method.