Covariance intersection multirobot object tracking algorithm based on self-adaption SR-CKF

Abstract

Multiple autonomous mobile robot system has attracted increasing attention from scholars for its spatial and functional distributivity, high fault tolerance, strong robustness and many other advantages. Aiming at numerical instability, huge calculation amount, poor precision and other problems exis ting in the synergetic dynamic object tracking of multiple mobile robot in unknown complex environment, this paper proposes the covariance intersection multirobot object tracking algorithm based on self-adaption SR-CKF. The algorithm is distributed and it can improve the accuracy of the evaluation on relevant objects without independence assumption for data information, and thus avoids the evaluation of the cross correlation among objects’ status. In addition, targeting at bad observation information, the self-adaption SR-CKF is built on the basis of the information covariance matching principle, which has improved the robustness of the whole system. The simulation result has proved that this algorithm can effectively solve the problems in multirobot synergetic objects tracking in unknown environment.