Rapid planning of industrial robot grasping path in unstructured environment

In order to overcome the problems of low smoothness, long time, and low success rate of traditional industrial robot grasping path planning methods, a rapid planning method of industrial robot grasping path in unstructured environment is proposed. Unstructured scene data is collected through LiDAR and voxel filtering is performed on the collected point cloud data. Based on the processed point cloud data and RBPF-SLAM algorithm, an environmental map is constructed, and the industrial robot grasping path is rapid planned in the environmental map through the deep deterministic policy gradient algorithm. The experimental results show that the maximum smoothness of the grasping path of the industrial robot proposed by the method is 0.98, the minimum path planning time is 0.57 s, the grasping success rate is between 96.2% and 97.9%, and the grasping path planning effect is good.