大規模点群のための高速干渉判定法に基づく工場設備内の経路探索

抄録

Terrestrial laser scanners enable us to capture dense point-clouds from engineering plants. So far, we developed an efficient collision detection method for large-scale point-clouds. In our method, we created layered depth maps and evaluated collisions using depth values. In this research, we discuss methods for path planning based on dense point-clouds. We generate 2D collision maps of engineering plants, and then detect shortest paths for movable objects. When an engineering plant is measured using a laser scanner, points are typically missing in some regions because of occlusion. Our method can evaluate whether better paths potentially exist if points are captured from occluded regions.