Fast and Efficient Motion Planning Algorithm for CFETR Multipurpose Overload Robot in Narrow Workspace

The China Fusion Engineering Test Reactor (CFETR) multipurpose overload robot (CMOR) has nine degrees of freedom (DOF), and the redundant DOF contributes to its ability to avoid collisions in the narrow workspace. However, the redundant DOF and narrow workspace poses significant challenges in the calculation of collision-free trajectories. Traditional motion planning algorithms have problems, such as low computational success rate and long computational time. In this article, a fast and efficient motion planning algorithm was proposed. The motion planning was divided into two stages: coming out of the cask and moving in the vacuum vessel (VV). In the first stage, the fixed trajectories are used. For the second stage, the shortest paths without collision are used as the training datasets, and a supervised neural network is trained by the training datasets to form the motion planning model, and the model can calculate the collision-free trajectory quickly with a high computational success rate. Compared with the traditional motion planning methods, the proposed motion planning method performs well, the computational time is short to 2 s, and the success rate is up to 96%.