Abstract

Food supply chain assurance should begin in the field with regular per-plant re-identification and logging. This is challenging due to localisation and storage requirements. A proof-of-concept solution is provided, using an image-based, super-GNSS precision, robotic localisation per-plant re-identification technique with decentralised storage and blockchain technology. ORB descriptors and RANSAC are used to align in-field stones to previously captured stone images for localisation. Blockchain smart contracts act as a data broker for repeated update and retrieval of an image from a distributed file share system. Results suggest that localisation can be achieved to sub 100mm within a time window of 18 seconds. The implementation is open source and available at: \url{https://github.com/garry-clawson/Blockchain-Crop-Assurance-and-Localisation}