| Popis: |
In order to solve the problem of automatic following of mobile robot in unstructured environment, an autonomous following method based on depth camera was proposed. A mobile platform based on ROS operating system and equipped with depth camera is designed. According to the depth-space characteristics of the depth camera, the 3D depth data and the coordinate information of the object are obtained. Secondly, according to the RGB mode of the depth camera, the specific target recognition and following problem is solved. The target is calibrated well in advance to reduce a lot of time spent in the later stage of target recognition and calibration, and the flexibility of following is improved by pasting the calibration objects on the target that needs to be followed. The improved artificial potential field algorithm is used to solve the unreachable problem of the target point and complete the path planning, so as to complete the autonomous obstacle avoidance following task. |
