In this paper, based on Kalman filtering algorithm, a method of target vehicle motion state radar estimation with radar (or lidar) is presented. The state equations is established based on rigid plane dynamics theory, and then with a Kalman filter to do radar data processing, the position, velocity and acceleration of the target vehicle can be estimated at the same time, so that to cover the shortage that acceleration information can not be gained with radar system. Through simulation and field tests it is verified that the detection accuracy of position and velocity of target vehicle is increasing, and the acceleration of target vehicle can be estimated effectively and accurately.