Interacting multiple sensor unscented Kalman filter for accelerating object tracking

Abstract

Due to limited sensing range for sensor nodes, moving object tracking has to be realized by relaying from one node to the other in a cluster. By taking object tracking in a fixed cluster as a Markov jump nonlinear system, the interacting multiple sensor unscented Kalman filter(IMSUKF) algorithm is designed to deal with distributed tracking. The proposed method can be divided into two parts: one-step unscented Kalman filter for object tracking and the fusion of the information provided by all the nodes. Finally, simulation results show the effectiveness of the proposed method.