Fuel & Find: A KTP Breakfast Forum
Room 304 - Business Hub, Sandyford Building
-
International
Ideally situated in the 5th best student city in the UK (QS Best Student Cities 2026), Northumbria University is a UK Top 40 University (Complete University Guide 2026) with a diverse community of 34,500 students from over 140 countries.
Discover moreBusiness
Northumbria University is proud to offer a range of Professional, Statutory and Regulatory Body (PSRB) approved & accredited courses and programmes. Explore our list of courses and programmes under our Education and Training page.
Discover moreResearch
Northumbria is a research-rich, business-focused, professional university with a global reputation for academic quality. We conduct ground-breaking research that is responsive to the science & technology, health & well being, economic and social and arts & cultural needs for the communities
Discover moreAlumni
Northumbria University is renowned for the calibre of its business-ready graduates. Our alumni network has over 253,000 graduates based in 178 countries worldwide in a range of sectors, our alumni are making a real impact on the world.
Discover moreTheme: Smart and Resilient Networks
Up to 40% of crops are lost to disease and pests before harvest. In North East England, persistent cloud cover makes optical satellite monitoring unreliable for much of the growing season. This project delivers a Digital Twin crop surveillance system that fuses optical satellite imagery (Landsat, Sentinel, Pléiades) with Sentinel-1 radar and Met Office weather data to monitor fields continuously regardless of cloud conditions. Uniquely, it uses a Tsetlin Machine, a transparent, rule-based AI, to detect crop threats and explain every alert in plain English via an Agentic LLM interface, enabling Farm Officers to query threats conversationally and act before yield is lost.
More about the Project Lead:
Kabita is the Principal Investigator of NESCA (Explainable Digital Twin Crop Surveillance) at Newcastle University, where she holds a Senior Lectureship in Signal Processing and Machine Learning. Her research portfolio spans a breadth of AI-focused interdisciplinary challenge areas, including explainable AI, medical digital twins for trustworthy clinical decision-making, smart sensing, and AI-augmented non-destructive testing and evaluation. Kabita is committed to developing lightweight and explainable machine learning tools where every decision can be reasoned, audited, and acted upon in real time. When applied to unconventional domains, these tools have the potential to tackle complex, high-impact societal and industrial challenges where decisions must be fast, transparent, and fully accountable.
In this project, Kabita aims to build a prototype explainable digital twin that combines satellite imagery, radar, and weather data to deliver transparent, real-time crop protection insights. By fusing multi-source space data with logic-based AI, the system will enable farm officers to detect and respond to crop threats early through auditable, natural-language conversations. She has secured competitive research funding as Principal Investigator and Co-Investigator on several prestigious grants, including EPSRC Programme Grants and a Horizon Europe award, and has published widely in reputed venues.
Room 304 - Business Hub, Sandyford Building
-
Page last updated 05/06/26