ＷＳＮｓ下移动机器人ＨｕｂｅｒＭ-ＣＫＦ离散滤波定位.

According to problem of low accuracy in traditional mobile robot localization, this paper designed the discrete HuberM-CKalmam filter (HCKF) based mobile robot positioning algorithm. It made use of the cost function of HurbM maximum likelihood estimation to solve the linearized CKF observation matrix, which improved the estimation accuracy for CKF filter algorithm with unknown-non-Gauss white noise disturbance. Then, it constructed WSNs based experiment environment for mobile robot at gymnasium, and combined with mobile robot dynamics model, it compared the positioning accuracy for the HCKF and the CKF algorithm. The results show that, for the two cases of non interference and unknown noise interference, the proposed method respectively improves the accuracy of HCKF algorithm about 7% and 15% more than CKF algorithm. [ABSTRACT FROM AUTHOR]