Zhu Qiguang, Peng Yingchun, Yuan Mei, and Chen Weidong
SLAM, cubature Kalman filter (CKF), extended H∞ filter (EH∞F), robustness
H∞ filter-based simultaneous localization and mapping (SLAM) algorithm does not make any assumptions about the noises, but it has the drawback of calculating Jacobian matrices; CKF-SLAM variants are Jacobian free, but they tend to show low robustness to non-Gaussian noises. For these problems, a cubature-extended H∞ filter-based SLAM algorithm (CEH∞F-SLAM) is proposed to possess advantages of both algorithms. In CEH∞F-SLAM, cubature transform is embedded into EH∞F by using statistical linear error propagation, and parameter γ is adjusted to its minimum at each iteration for better robustness. CEH∞F-SLAM not only avoids the calculation of Jacobians but also shows strong robustness facing unknown disturbance. The performance of UKF-SLAM and cubature Kalman filter-SLAM is compared with the proposed algorithm by computer simulations. The results show that CEH∞F- SLAM algorithm, which has better stability and accuracy, is an effective SLAM algorithm.
Important Links:
Go Back
