Path planning and obstacle avoidance for omni-directional mobile robot based on Kinect depth sensor

摘要

To enable robots to autonomously move and skirt obstacles in the field, the robots must be designed so that they can exercise self-localisation and path planning functions based on the results of sensor signal processing or image analysis. This research combines the Kinect depth sensor with the omni-directional camera to construct the vision system of FIRA RoboSot. We use the above two complementary devices to enable robots to position themselves and to detect their distance from obstacles. The data captured can be processed and used to construct the required digital map, assisting robots in completing path planning. For testing the performance of our approach, some experiment results show the ability of the proposed system to perform robust obstacle avoidance under indoor conditions. All these results have been proven for their excellence and feasibility.