Design and Implementation of Intelligent Orchard Robot

This paper addresses the labor shortages and high costs in traditional agriculture by designing an intelligent orchard picking robot. The robot leverages the ROS operating system, NVIDIA Jetson Nano, and a tracked chassis to navigate complex terrains. Key technologies include SLAM for real-time mapping, A* for path planning, and YOLOv5 for fruit detection. Experiments conducted in a simulated orchard environment demonstrate the robot's effectiveness in autonomous navigation and accurate fruit recognition, achieving 95% recognition accuracy in clear conditions and 75% under occlusion. The results showcase the robot's potential to significantly enhance productivity and reduce labor demands in orchard harvesting.