PATH PLANNING FOR MOBILE ROBOT USING SKELETON OF FREE SPACE

This paper proposes a new path planning method for obstacle avoidance of mobile robot. In order to achieve easy planning of the path, the free space of the environment is represented by a skeleton consisting of wires and nodes, whereas the skeleton of free space is obtained based on the thinning algorithm which is widely used in the image processing technique. In addition, a path graph is introduced to represent the connectivity between nodes and wires, where the number of nodes, a measure of computational complexity, depends on the number of obstacles rather than the number of vertices of obstacles. The path graph facilitates the determination of feasible paths from the skeleton of free spaces. But the feasible paths may have any unnecessary turns along which the mobile robot cannot follow, a path improvement process is performed with respect to the feasible path in order that robot may follow the path smoothly. Comparing with the visibility graph (V-graph) method, the proposed method is optimized in terms of searching time. The effectiveness of the proposed method is verified by a series of simulation studies.