IN Brief:
- University of Lincoln’s LIAT team showcased the PhenAIx robotic phenotyping system at RHS Chelsea Flower Show.
- The system uses robotics, AI, and advanced imaging to detect hidden signs of stress, disease risk, and plant performance.
- Earlier crop insight can support breeding, yield protection, input efficiency, and more resilient ingredient supply chains.
The Lincoln Institute for Agri-Food Technology has showcased an AI-driven robotic crop-scanning system designed to detect plant health issues before visible symptoms appear.
The University of Lincoln’s RoboCrops: Plant Selection, Beyond the Visible exhibit won a Silver Gilt medal at RHS Chelsea Flower Show 2026. Presented by LIAT in the show’s GreenSTEM zone, the exhibit demonstrated how robotics, artificial intelligence, imaging, and plant science can be combined to assess crop performance in greater detail.
At the centre of the exhibit was PhenAIx, an advanced robotic phenotyping system that captures information about plant growth, structure, and health. The system can identify subtle signs of stress, disease risk, and hidden differences in plant performance that are not visible to the human eye.
By giving researchers and breeders more detailed plant data, the system is intended to support faster selection of crops with stronger resilience traits. These may include better tolerance to heat, drought, disease, or lower-resource production conditions, depending on the crop and growing environment.
Professor Elizabeth Sklar, director of LIAT, said: “We’ve had tremendous interest from visitors regarding our interdisciplinary approach and received many compliments on the ways in which our stand highlights the need for collaboration across different STEM disciplines in order to address complex problems related to food security.”
Crop health insight is becoming more closely connected with food manufacturing performance. Processors depend on consistent crop volume, quality, size profile, storage stability, and specification. Disease or stress that reduces yield, changes dry matter, affects sugar levels, weakens shelf life, or increases defects can move quickly from field risk into factory risk.
Robotic phenotyping is likely to be most useful in breeding programmes, trial plots, high-value horticulture, controlled-environment farming, seed development, and intensive crop monitoring. Compared with manual assessment, automated systems can capture more consistent measurements, while AI can identify patterns across growth and stress response that would otherwise be difficult to detect.
Agricultural uncertainty will not disappear through imaging and robotics alone. Disease pressure still depends on weather, soil, variety, farm management, region, and seasonal conditions. Better detection can, however, help growers intervene earlier, improve trial selection, and support procurement planning with more accurate upstream information.
Technology-led growing systems are becoming a stronger feature of food supply. Oishii’s $150m smart farming scale-up reflects the same wider movement toward data-rich production, even though vertical farming and robotic phenotyping occupy different parts of the agri-food system.
Plant health data also sits close to food safety and quality assurance. A crop disease issue does not automatically create a safety risk, but it can affect grading, storage behaviour, processing quality, contamination control, and waste. Earlier visibility allows better decisions on segregation, harvest timing, transport, and processing allocation.
The industrial challenge now sits in deployment. Robotic systems need to operate reliably in real growing environments, generate data that growers can act on, and connect with supply-chain decisions without adding unnecessary complexity. Their value will be measured through better forecasts, fewer specification failures, improved yield stability, and stronger crop resilience.
As climate, disease, and input pressures intensify, plant health is becoming a data problem as well as an agronomic one. The earlier that information is captured, the more useful it becomes across breeding, growing, processing, and supply continuity.
