Novel Software Architecture for an Autonomous Agricultural Robotic Fruit Harvesting System

This paper introduces a novel system architecture developed for a custom built, agricultural, autonomous fruit harvesting robot and presents proof of concept results for the system using the GMapping [1] SLAM algorithm. The robot has been developed for use in a state-of-the-art indoor vertical fruit growing environment. The system architecture and robot transformation coordinate frames are described before initial proof of concept navigational experiments are conducted and results presented. A global SLAM and local path planning novel navigational system is also proposed as part of future development of the system.