LAUSR

Five-axis sweep scanning is an emerging surface inspection technology that has shown great potential to boost inspection efficiency for complex surfaces. The existing five-axis inspection methods rely on computer aided design (CAD) models to plan the entire sweep scanning path prior to measurement. Thus, they cannot be applied to measure deformed surfaces for repair or remanufacturing, since deformed surfaces lack valid CAD models. To solve this problem, an adaptive five-axis inspection approach is proposed for inspecting deformed surfaces based on real-time path planning, with no use of any CAD model. The proposed approach uses an iterative strategy that implements through three steps: surface prediction, sweep scanning path planning on the predicted surface, and conducting the inspection. The surface prediction method is accurate enough and computationally fast, so it is suitable for online applications. The path planning algorithm can improve inspection efficiency by avoiding empty strokes of measurement. Experiments conducted on three complex free-form surfaces verify the feasibility, accuracy, and efficiency of the proposed approach.