Optimising robotic operation speed with edge computing via 5G network: Insights from selective harvesting robots

Selective harvesting by autonomous robots will be a critical enabling technology forfuture farming. Increases in inflation and shortages of skilled labor are driving factorsthat can help encourage user acceptability of robotic harvesting. For example,robotic strawberry harvesting requires real‐time high‐precision fruit localization,three‐dimensional (3D) mapping, and path planning for 3D cluster manipulation.Whilst industry and academia have developed multiple strawberry harvesting robots,none have yet achieved human–cost parity. Achieving this goal requires increasedpicking speed (perception, control, and movement), accuracy, and the developmentof low‐cost robotic system designs. We propose the edge‐server over 5G for SelectiveHarvesting (E5SH) system, which is an integration of high bandwidth and low latencyFifth‐Generation (5G) mobile network into a crop harvesting robotic platform, whichwe view as an enabler for future robotic harvesting systems. We also considerprocessing scale and speed in conjunction with system environmental and energycosts. A system architecture is presented and evaluated with support fromquantitative results from a series of experiments that compare the performance ofthe system in response to different architecture choices, including imagesegmentation models, network infrastructure (5G vs. Wireless Fidelity), andmessaging protocols, such as Message Queuing Telemetry Transport and TransportControl Protocol Robot Operating System. Our results demonstrate that the E5SHsystem delivers step‐change peak processing performance speedup of above 18‐foldthan a standalone embedded computing Nvidia Jetson Xavier NX system.