SLAM Data Association of Mobile Robot Based on Improved JCBB Algorithm

Data association is an important step in simultaneous localization and mapping (SLAM) for mobile robots. In order to ensure high association accuracy and reduce the amount of the association calculation, an improved joint compatibility branch and bound (JCBB) algorithm based on adaptive local feature association and adaptive Gauss mixture clustering is proposed in this paper. Firstly, the local association region is adaptively determined according to the current position of the robot, the measurement range of the sensor and the distribution of the current observations. Then, these observations are divided into several groups by using adaptive Gauss mixture clustering method, and standard JCBB algorithm is used to correlate the observations of each group with the features in local region. Finally, the eventual association result is obtained by integrating the optimal solutions of each group. The experimental results show the effectiveness of the proposed algorithm in this paper.