Key points are not available for this paper at this time.
Digital twins are a core industry 4.0 technology enabling the virtual replication of real-world objects, mimicking behaviours and states throughout their lifespan. While digital twins have shown significant benefits in industries such as manufacturing, transportation, and healthcare, their application in agriculture is still within its infancy. Their realisation also poses significant challenges, such as the creation of dynamic agricultural objects (e.g., plants). Existing literature on digital twins in agriculture identifies their limited ability to monitor physical objects without predictive capabilities and that there is a significant lack of 3D representations of plants with functional attributes. Yet, incorporating 3D representations of plants with underlying functionality in a digital twin can greatly improve growth, yield, and disease prediction accuracy. This enhancement enables various applications, such as assessing and developing pruning strategies, providing education to growers, guiding pruning robots, and optimizing spraying techniques. To that end, Functional Structural Plant Modelling presents a potential solution by representing the 3D architecture of plants and incorporating the functionality of different plant parts. By conducting a domain analysis of 3D plant phenotyping and FSPM, this study addresses the specific needs of digital twins in agriculture regarding FSPM. The investigation bridges the existing knowledge gap by identifying crucial concepts, including 3D plant modelling with underlying functionality and 3D plant phenotyping for digital twins. Specifically, a framework for 3D FSPM integration into agricultural digital twins is proposed. The framework not only acknowledges the associated requirements and challenges identified in existing literature but also lays foundation for the advancement of digital twins in the agricultural domain.
Mitsanis et al. (Tue,) studied this question.