Greenhouse Rail Detection Using 3D Point Clouds for Robot Docking

The patent transitions from 2D camera-based detection to 3D point cloud detection using a tilting laser scanner. This dimensional change enables accurate three-dimensional positioning of the rail, allowing the mobile robot to detect rail position with high precision in the x, y, and z directions, thereby resolving the measurement precision issue while maintaining manageable device complexity through sophisticated algorithmic processing of the 3D data.

The patent replaces conventional mechanical or simple optical sensors with a laser-based scanning system. The tilting laser scanner uses light reflection and time-of-flight measurements to detect rail position, substituting mechanical measurement methods with optical field-based detection, which provides superior measurement precision for three-dimensional spatial positioning.

The patent performs preliminary rail detection and position identification before the mobile robot attempts to dock. The system scans the environment, identifies rail positions in advance, and plans the switching path from flat area to rail area beforehand. This preliminary action ensures that the robot has accurate position information and a predetermined safe path, thereby improving reliability while managing detection complexity through advance preparation.

The patent implements a feedback mechanism where the mobile robot continuously detects rail position using the tilting laser scanner, compares the detected position with the planned docking path, and adjusts its motion in real-time. This closed-loop feedback control ensures safe switching between flat and rail areas by constantly monitoring and correcting the robot's position based on accurate rail detection data.