Robotic grinding based on point cloud data: developments, applications, challenges, and key technologies

Robotic grinding based on point cloud data is considered an alternative solution for efficient and intelligent machining of complex components by virtue of its flexibility, intelligence, and cost efficiency, particularly in comparison with the current mainstream manufacturing modes. Over the past two decades, the development of robotic grinding techniques based on point cloud data has evolved from an independent measurement and machining operation to an integrated “measurement-machining” approach. The total grinding cycle time was reduced by 43% compared to manual grinding. Currently, the average measurement error of the robotic grinding system based on point cloud data is about 0.06 mm, the average machining error is 0.1 mm, and the average roughness of the surface is 0.286 µm, which can meet the requirement of complex components. The relevant research in the field of robotic grinding based on point cloud data in the past 20 years was organized in this paper. Then technical difficulties, specifications, and breakthrough advances of robotic grinding were summarized. Online measurement and path planning were analyzed on robotic grinding for complex components. Finally, some research interests and potential application areas were proposed to improve the accuracy, quality, and application range.