Optimal path determination for a robot in a 2D workspace using quadtree modeling

Avoiding collisions is the primary objective for a robot moving in a workspace with obstacles. The conventional obstacle avoidance algorithms suffer from the drawbacks due to considerable approximations and lack of flexibility. The proposed obstacle avoidance algorithms are developed based upon the geometric modeling of robot workspace using quadtree data structure. By examining the obstacle-quadtree, nodes that can allow safe passage for the mobile robot are identified. By connecting all such nodes, a set of collision-free paths are generated and subsequently the optimal path is chosen