Multi-robot collaborative SLAM Based on optimized ORB

A multi-robot vision collaborative SLAM system based on the updated ORB algorithm was provided to respond to the needs of large error and poor sensitivity in the multi-robot vision collaborative SLAM system. Firstly, histogram equalization is carried out on the collected image, which improves the overall contrast of the image. Shi-Tomasi corner detection algorithm is used to supplement the loss of feature points during camera rotation. Second, Lucas-Kanade (LK) sparse optical flow is employed to enhance the ORB algorithm's feature matching process. Finally, the improved ORB algorithm is applied to the CCMSLAM multi-robot cooperative system. The experimental findings demonstrate that after using the revised method, the system has enhanced accuracy and resilience.