Digital Twin for the Future of Orchard Production Systems

Abstract

The evolution of orchard production systems towards higher density layouts, makes monitoring of canopy and disease increasingly important. Technological advances over the last few years have greatly increased our ability to collect, collate and analyse our data on a per-tree basis at large orchard scales. We call this the Digital-Twin Orchard. A digital-twin is a virtual model of every tree and surroundings. The pairing of the virtual and physical worlds allows analysis of data and continuous monitoring of orchards production systems to predict stress, disease and crop losses, and to develop new opportunities for end-to-end learning. Monitoring of orchards is not a new concept but the digital-twin is a continuously learning system that could be queried automatically to analyse specific outcomes under varying simulated environmental and orchard management parameters. Digital-twin enables improvement of production and dynamic prediction of disease, stress and yield gaps using an end-to-end AI platform. In this paper, we present AgScan3D+: our automated dynamic canopy monitoring system to generate a digital-twin of every tree on a large orchard scale. AgScan3D+ consists of a spinning 3D LiDAR plus cameras that can be retrofitted to a farm vehicle and provides real time on-farm decision support by monitoring the condition of every plant in 3D such as their health, structure, stress, fruit quality, and more. The proposed system has been trialled in mango, macadamia, avocado and grapevines orchards and generated a digital-twin of 15,000 trees. The results were used to model canopy structural characteristics such as foliage density and light penetration distribution.

Can a Byte Improve Our Bite? An Analysis of Digital Twins in the Food Industry

“…Many applications focus on the growth of plants [ 36 , 42 , 53 , 54 , 61 , 62 , 63 , 64 ] or monitoring the condition of animals [ 23 , 61 , 64 , 65 , 66 ]. Further, entire production systems as greenhouses or fields are twinned [ 34 , 43 , 44 , 61 , 62 , 67 , 68 , 69 , 70 , 71 ]. Several applications could be described as supportive, e.g., to monitor and control pests [ 35 , 65 ].…”

“…The predictions could be used to suggest corrective or preventive actions (39.22%). Since most of the applications found are assigned to the production stage, many prescriptive digital twin applications belong to applications only able to assist in agricultural plants to enhance the quality during growth and harvest processes [ 42 , 43 , 61 , 62 , 64 , 69 , 70 , 78 , 80 ]. Another prescriptive digital twin is applied in a pudding production system to assist in production planning [ 73 ].…”

“…Other applications were implemented to better document the processes and quality parameters of the physical objects [ 6 , 30 , 47 , 57 , 74 , 76 , 79 ]. It should be noted that due to a lack of information, many applications could not be classified in this category [ 23 , 42 , 43 , 44 , 45 , 46 , 56 , 65 , 66 , 67 , 69 , 70 , 71 , 73 , 75 ].…”

“… [ 68 ] Crop management (irrigation system) Production X X X X X [ 69 ] Aquaponic system Production X X X X X n.a. [ 70 ] Orchard production system Production X X X X X n.a. [ 71 ] Crop management Production X X X X …”

Canopy density estimation in perennial horticulture crops using 3D spinning lidar SLAM

“…Accurately estimating canopy structure on a per‐vineyard block basis and multiple times per season poses a difficult technical challenge. The temporal fidelity requires a method that has low time burden, the requirement for accuracy and spatial detail requires a full (rather than sample‐based) map of the vineyard, and the temporal longevity (analysis over multiple seasons) requires an absolute and canonical data structure that is essentially unchanged as the hardware and sensors change over time (Edwards & Moghadam, 2020; Moghadam et al, 2020).…”