This study deals with simultaneous localization and mapping problem by using unscented Kalman filter to compensate for observation outliers. In solving simultaneous localization and mapping problem using algorithms such as EKF or UKF, robot observations play a crucial part in determining its position estimation in any environment. If the robot observations obtained unexpected or fault values, the accuracy of the estimation will be deteriorated. In this research, an enhanced method based on UKF is developed to overcome the fault observation by assigning a weights to the observations. By comparing the observation values with its own estimate to detect the fault observations and then the weights of these observations are determined. Simulations were carried up to investigate the performance of the new method by comparing it with EKF-SLAM, UKF-SLAM and H∞ SLAM. The algorithms are compared in terms of parameters such as the RMSE and the runtime of the algorithm by using MATLAB. Results show that proposed method can performed better compared to other in dealing with observation outliers.


    Access

    Download


    Export, share and cite



    Title :

    Range and Bearing-based Simultaneous Localization and Mapping of Unmanned Ground Vehicle using Unscented Kalman Filter



    Publication date :

    2016-12-02


    Remarks:

    Journal of Telecommunication, Electronic and Computer Engineering (JTEC); Vol 8, No 11: Artificial Intelligence and Robotics; 119-123 ; 2289-8131 ; 2180-1843


    Type of media :

    Article (Journal)


    Type of material :

    Electronic Resource


    Language :

    English



    Classification :

    DDC:    629



    Unscented Kalman Filter and Gauss-Hermite Kalman Filter for Range-Bearing Target Tracking

    Barragán, Gabriel / Infante, Saba / Hernández, Aracelis | Springer Verlag | 2021


    Unscented Kalman Filter

    Zarchan, Paul / Musoff, Howard | AIAA | 2015


    Three-dimensional positioning of unmanned vehicle in field based on unscented Kalman Filter

    Li, Xiaolong / Zhang, Chi / Jin, Ma et al. | British Library Conference Proceedings | 2022