Practical path planning and path following for a non-holonomic mobile robot based on visual servoing

摘要

In this paper, a practical real-time path planning and robot navigation algorithm for a non-holonomic indoor mobile robot based on visual servoing is implemented. The proposed algorithm is divided into three parts; the first part uses Multi-Stencils Fast Marching (MSFM) as a path planning method. But the generated path results from fast marching methods when used directly, not guaranteed to be safe and smooth. Subsequently, the robot can touch corners, walls and obstacles. The proposed algorithm uses image processing methods to solve this problem. The second part estimates the position and orientation of the robot, from the visual information, to follow the desired path with avoiding obstacles. The third part proposes a decentralized PD-like Fuzzy Logic Controller (FLC) to keep up the robot on the desired path. Experimental results show that the developed design is valid to estimate shortest-path by avoiding obstacles and able to guide the robot to follow the path in real-time.