Alignment Detection Technology of Chang’e-6 Primary Package Container

The Chang’e-6 mission achieved the first successful sample collection and return from the Moon’s far side. Accurate alignment detection of the primary packaging container is critical for the success of this mission, as it ensures proper retrieval of lunar soil. To address challenges such as complex backgrounds, uneven lighting, and reflective surfaces, this paper introduces an alignment detection method that integrates YOLO object recognition, Devernay subpixel edge detection, and the RANSAC fitting algorithm. By employing both linear and elliptical fitting techniques, the method accurately determines the median line of the primary packaging container, ensuring precise alignment detection. The effectiveness of this approach is demonstrated by an average alignment distance of 0.28 mm with a standard deviation of 0.03 mm in lunar surface images, underscoring its accuracy and reliability.