SMAART proposes an intelligent, low-cost system for 24\/7 autonomous and continuous monitoring based on recent developments in artificial intelligence, sensing, Internet of Things, edge computing and robotics, leveraging advanced Artificial Intelligence of Things technologies and deep learning-based decision support tools.
\n\u00a0
\nThe proposed platform is designed to integrate with existing technology, extending its limits and lowering its cost, rather than to replace it. The expected result is a virtual representation as accurate as possible and updated in real time (digital twin, or digital copy) of the condition of the crop field and\/or herd that includes a time-varying, high-precision 3D model of each plant (feasible for low\/medium density crops) or individual animal moving through space and often exhibiting aspects of dynamic morphology over the course of a productive career.
\n\u00a0
\nThe 3D model will allow, among other things, the evaluation of physical and geometric parameters of interest, such as canopy volume or leaf area for plants, or morphological traits of the animal, useful for livestock enhancement; near-real-time detection of health emergencies or indications of animal physiological parameters (fattening status, behavioral changes, interaction with wild animals, herd\/mandria dynamics) as well as veterinary issues (lameness, parturition, trauma); production of detailed prescription maps for precision agriculture; optimized grazing planning; and estimation of spatial distribution of animals.
\n\u00a0
\nSMAART involves the design and implementation of an overall platform that includes; \u25cf the implementation of an agricultural Unmanned Ground Vehicle prototype for multi-temporal monitoring and real-time management of emergencies in agriculture; \u25cf a demonstration system of a multisensor robotic platform equipped with AI algorithms for the acquisition, classification, and generation of Agridata (agricultural and livestock) \u25cf an Unmanned Aerial Vehicle system for continuous semi-automated crop and pasture surveillance, prescription map generation, and grazing condition assessment; \u25cf a smart platform for managing existing large-scale datasets (e.g. weather or satellite time series) or to be acquired during the course of the project (e.g. images and videos); \u25cf a comprehensive set of optimized AI algorithms for field\/grazing\/planting data classification and early detection of plant diseases (entomofauna for optimized and precision management of allowed plant protection products and weed control in pre- and post-emergence in the integrated setting) or detectable changes on the individual animal or herd level (nutritional status, motor activity and space occupation, social dynamics, etc.,); \u25cf deterministic tools for risk assessment in animal feeding and parameters of interest for optimal pasture and animal management; \u25cf digital twin of the culture plant and herdincluding evolutionary 3D representation of the individual animal\/plant, quantitative evaluation of morphological and growth parameters;\u00a0\u25cf front end for real time updating of the field\/grazing situation on the end user’s device, reporting of emergencies and possible initiation of treatments.
\n\u00a0
\nSMAART also provides for an articulated activity of field trials. The Sa Marigosa Producers’ Organization will provide sites where demonstration actions will be conducted on different types of horticultural plants (medium density plants). The livestock activities will make use of the collaboration of a support farm that will be selected according to certain inclusion criteria, about the animal species and breed, the animal categories present, semi-extensive farming and the availability of pasture.
\n\u00a0
\nThe results emerging from the proposed project activities will be able to return the integration of the production process supported by smart decision-making strategies with general and scalable value, which can potentially be extended and appropriately declined to other crop types and forms of farming.<\/p>\n","protected":false},"featured_media":13456,"template":"","meta":{"_acf_changed":false},"gruppo":[99],"class_list":["post-12568","projects","type-projects","status-publish","has-post-thumbnail","hentry","gruppo-tecnologie-digitali-en"],"acf":{"title":"Sostenibilita' e management per agricoltura e zootecnia di precisione con intelligenza artificiale, robot e tecnologie IoT","acronym":"SMAART","project_uuid":"bc6c8d11-0c41-11ef-8ceb-506b8da9258c","start_date":"01\/01\/2024","end_date":"31\/12\/2026","funder":"MIMIT - Fondo per la crescita sostenibile al sostegno di progetti di ricerca e sviluppo - Accordi Innovazione 2^ sportello","partners":"\n
| Nome<\/th>\n | Nazione<\/th>\n | Ruolo<\/th>\n <\/tr>\n\n |
|---|---|---|
| Abinsula Srl<\/a><\/td>\n | Italy<\/td>\n | Coordinator<\/td>\n <\/tr>\n\n |
| Greenshare S.r.l.<\/a><\/td>\n | Italy<\/td>\n | Partner<\/td>\n <\/tr>\n\n |
| Universit\u00e0\u00a0degli Studi di Sassari<\/a><\/td>\n | Italy<\/td>\n | Partner<\/td>\n <\/tr>\n\n |
| Sa Marigosa Societa'\u00a0 Consortile a r.l.<\/a><\/td>\n | Italy<\/td>\n | Partner<\/td>\n <\/tr>\n<\/table>","team_members":[],"project_image_header":false,"proj_website":"","project_status":"Execution","project_coordinator":[{"ID":10264,"post_author":"4","post_date":"2025-06-30 17:23:51","post_date_gmt":"2025-06-30 15:23:51","post_content":"Fabio Maggio is a senior researcher and Permanent Project Leader (PPL) of the Digital Technologies for Aerospace sector. His focus lies within digital agriculture and precision farming, which he explores through three foundational and interconnected ICT technologies: AI, big data, and robotics.Data pertaining to agriculture reaches CRS4 through various channels. Agro-meteorological data is automatically delivered daily, thanks to our collaboration with Arpa Sardegna; additionally, images, time series, and other forms of information are sourced from agencies for agriculture of Regione Sardegna (Laore and Agris), as well as from end users who collaborate with us. Machine learning methods play a pivotal role in data analysis and processing, such as optimal interpolation of agro-meteorological data to obtain crop-scale information. Moreover, cutting-edge deep learning models enable us to automatically classify images for early detection of field emergencies, such as occurrences of phytopathologies, weed\/pest infestations, and deficiencies in water and\/or nutritional elements. With the aid of state-of-the-art edge devices, able to mitigate the necessity for cloud computing, and optimized AI techniques, real-time classification is now achievable. This opens the way to a new generation of smart rovers (unmanned ground vehicles, UGV) capable of continuous and quasi-autonomous operation. This interdisciplinary expertise forms the foundation of a structured initiative led by Fabio and his colleagues at CRS4.Fabio also boasts decades of professional experience in numerical simulation (i.e., computer-based solutions for large-scale partial differential equations), high-performance computing (HPC) utilizing parallelism and general-purpose GPUs, and leading edge numerical methods (such as spectral elements and domain decomposition).Between 2018 and 2023, Fabio delivered courses on \"Numerical Simulation and HPC\" and \"Machine Learning\" for the Master's Degree program in Mathematics at the University of Cagliari.","post_title":"Fabio Maggio","post_excerpt":"","post_status":"publish","comment_status":"closed","ping_status":"closed","post_password":"","post_name":"fabio-maggio","to_ping":"","pinged":"","post_modified":"2026-04-05 19:12:01","post_modified_gmt":"2026-04-05 17:12:01","post_content_filtered":"","post_parent":0,"guid":"https:\/\/www.crs4.it\/people\/fabio-maggio\/","menu_order":0,"post_type":"people","post_mime_type":"","comment_count":"0","filter":"raw"}]},"yoast_head":"\n |