A Farm Management Information System for Semi-Supervised Path Planning and Autonomous Vehicle Control

This paper presents a farm management information system targeting improvements in the ease of use and sustainability robot farming systems. The integrates functionalities field survey, path planning, monitoring, controlling agricultural vehicles real time. Firstly, Grabcut-based semi-supervised registration method is proposed for arable detection from orthoimage taken by drone with an RGB camera. It partitions complex into simple geometric entities user interaction. average Mean Intersection over Union about 0.95 when size ranges 2.74 ha to 5.06 ha. In addition, desktop software web application are developed as entity FMIS. Compared existing FMISs, this provides more advanced features farming, while providing simpler interaction better results. allows clients invoke services receive responses independent programming language platforms. Moreover, compatible other services, users, devices following open-source access protocol. We have evaluated 5 tractors 2 Hz communication frequency. protocols will be publicly available protentional users.