Adaptive sampling based on the motion

Motion based adaptive sampling method is presented in this paper. The distribution of particles can be adaptively changed according to motion information so that particles could track object in higher accuracy when the state transition function is unknown. The advantage of proposed method is examined by doing strict simulation tests in 1D, 2D and 3D space. The result was compared with the result from the standard particle filter in each step. The simulation results show that it would improve the prediction accuracy using the proposed method. Keywords-particle filter; adaptive; motion; simulation