Fusing Lidar, Radar, and Camera Using Extended Kalman Filter for Estimating the Forward Position of Vehicles

As radar and lidar sensors' precision varies with distance, this paper proposes an extended Kalman filter that reflects the precision of the sensors as the distance changes. Majority of previous studies did not consider the measurement errors of the sensors. Thus, the Kalman filter is important in defining the relationship between the measurement vector and the state vector well. We added a reliability function to express the relationship between the state vector and the measurement vector. For the experiment, we took data from the actual vehicle for greater accuracy in estimating the position although the measurement error is reflected. Results show that the proposed method estimates the position more accurately and the estimate process is smoother.