Aiming at the problems of low accuracy, poor real-time performance and large cumulative error in traditional depth camera visual synchronization positioning and map construction algorithms, this paper proposes an improved visual SLAM algorithm for mobile robots based on depth camera. According to the realization process of robot vision, a feature point extraction algorithm based on feature extraction and image segmentation watershed algorithm is proposed in the front-end process, which improves the real-time performance of the algorithm. Then the RE-RANSAC algorithm is used to eliminate the mismatched feature points to improve the matching accuracy, and then the accumulated error is eliminated through closed loop detection, and finally the process of robot mapping is completed. After simulation experiments, the feasibility of the improved algorithm is proved, and the robot’s mapping and trajectory estimation are completed.
本站僅提供期刊文獻檢索。
