Field navigation method of phenotypic robot based on ground-air cooperation

• Main Authors: LI ZHENGDA, YANG WANNENG, YANG MENG, SONG PENG, SHAO YANG, YUAN GUANG, ZHANG ZIKANG
• Format: Patent
• Language: Chinese; English
• Published: 13.08.2024
• Subjects: CALCULATING; COMPUTING; CONTROLLING; COUNTING; GYROSCOPIC INSTRUMENTS; MEASURING; MEASURING DISTANCES, LEVELS OR BEARINGS; NAVIGATION; PHOTOGRAMMETRY OR VIDEOGRAMMETRY; PHYSICS; REGULATING; SURVEYING; SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES; TESTING

The invention discloses a field navigation method of a phenotypic robot based on ground-air cooperation, which comprises the following steps: constructing a digital orthoimage DOM for a target crop area through an unmanned aerial vehicle equipped with a real-time motion compensation RTK technology and a camera module, and obtaining a crop row subject semantic segmentation map through a semantic segmentation model in combination with an artificial intelligence technology; according to the method, an image is preprocessed through an image processing method, key points are obtained in combination with a Skelegonize algorithm, a Douglas-Peucker algorithm, a Shii-Tomasi method and a linear extrapolation mode, the key points are grouped and sorted according to a proper strategy and then imported into a navigation system, a coherent navigation path is generated, a pure tracking algorithm controller is used for tracking the path, and the path tracking speed is high. And completing a navigation task. The system is use