Determining Robot Trajectory Planning Using Image Processing for Wood Painting

This study proposes a new system for painting complex-geometry parts moving on a conveyor belt using a spray gun-equipped robotic arm. The system utilizes a Time of Flight (TOF) sensor to capture the 3D geometry of the parts, followed by a hierarchical edge detection algorithm to identify features such as holes and contours. The number and position of holes, as well as the distinction between perforated and non-perforated pieces, can be determined. Additionally, parts with flat or curved surfaces, machined using Computer Numerical Control (CNC), can also be detected. Using these features defined by image processing and classification methods, trajectory planning is performed for each part. A paint spray gun connected to the robotic arm applies paint to the surfaces and edges of the parts. The system leverages libraries such as OpenCV, Math, and Numpy for trajectory planning with affinity transformations. This novel system provides a reliable and efficient solution for painting complex parts in various manufacturing industries, including wood, automotive, and aerospace.