Robotic Trajectory Planning for Non-Destructive Testing Based on Surface 3D Point Cloud Data

Robotics has been widely used in the field of non-destructive testing in recent years. However, for complex surfaces, manual teaching or offline programming is time-consuming and difficult to ensure high precision for non-destructive testing robot trajectory planning. Therefore, this work proposes a new method to generate non-destructive testing trajectory of the robot based on the pre-processed point cloud data. The workpiece surface is measured by 3D sensor to obtain the point cloud data. The trajectory line on workpiece surface is obtained by slicing pre-processed point cloud data. The dense trajectory points are obtained by isometric discretizing trajectory lines, and then they are compressed by Douglas-Peucker algorithm. The Principal Component Analysis (PCA) method is used to estimate the normal vector of the optimized trajectory points and unify their orientation. The pose of non-destructive testing robot can be obtained by biasing the trajectory points along their normal vectors finally.