No abstract

No abstract

An efficient path planning algorithm, for multi degrees of freedom manipulator robots in dynamic environments, is presented in this paper. The proposed method is based on a local planner and a boundary following method for rapid solution finding. The local planner is replaced by the boundary following method whenever the robot gets stuck in a local minimum. This method was limited to 2-DoF mobile robots and in this work we showed how it can be applicable for a robot with n degrees of freedom in a dynamic environment. The path planning task is performed in the configuration space and we used a hyperplane in the n dimensional space to find the way out of the deadlock situation when it occurs. This method is, therefore, able to find a path, when it exists, no matter how cluttered is the environment, and it avoids deadlocking inherent to the use of the local method. Moreover, this method is fast, which makes it suitable for on-line path planning in dynamic environment. The algorithm has been implemented into a robotic CAD system for testing. Some examples are presented to demonstrate the ability of this algorithm to find a path no matter how complex is the environment. These examples involve a 5-DoF robot in a cluttered environment, then two 5-DoF robots, and finally three 5-DoF robots. In all cases, the proposed method was able to find a path to reach the goal and to avoid the dynamic obstacles.

No abstract

In this paper we propose a new path planning method for robot manipulators in cluttered environments, based on lazy grid sampling. Grid cells are built while searching for the path to the goal configuration. The proposed planner acts in two modes. A depth mode, while the robot is far from obstacles, makes it evolve toward its goal. While a width search mode becomes active when the robot gets close to an obstacle. This method provides the shortest path to go around the obstacle. It also reduces the gap between pre-computed grid methods and lazy grid methods. No heuristic function is needed to guide the search process. An example dealing with a robot in a cluttered environment is presented to show the efficiency of the method.