This subject aims to study the detection and tracking of mobile robots using RGB‐D cameras. Because ordinary cameras have difficulty in distance measurement in practical applications, and inaccurate positioning results in poor tracking results, the application of RGB-D cameras in ROS mobile robots has been studied on the problems of target detection and tracking. According to the problems in the actual operation, the deep learning network model is optimized through comparison to improve the accuracy of detection, and finally the tracking strategy is formulated and the tracking test is performed on the mobile robot.