Emmanuel Soumo
Research Interests
Robotic, computer vision, perception and signal processing.
Publications
Zhou, L., Ali, O., Arnaud Soumo, E., Attenborough, E., Swindell, J., Davies, J., & Fox, C. (2025). 3D Printer Based Open Source Calibration Platform for Whisker Sensors. In: Huda, M.N., Wang, M., Kalganova, T. (eds) Towards Autonomous Robotic Systems. TAROS 2024. Lecture Notes in Computer Science, vol 15051. Springer, Cham. [December 2024].
Soumo, E., Calisti, M., & Polydoros, A. (2025) Data-Driven Greenhouse Climate Regulation in Lettuce Cultivation Using Bilstm and Gru Predictive Control, Preprint. [July 2025].
Presentations
“A Data-Driven Approach for Optimal Control of Autonomous Greenhouse for Lettuce Growth” (poster) – AgriFoRwArdS CDT Annual Conference 2024: Robots in Action [July 2024] – Norwich, UK.
“A Data-Driven Approach for Optimal Control of Autonomous Greenhouse for Lettuce Growth” (oral) – Research visit to Crystal Heart Salad [October 2024] – Hull, UK.
“Greenhouse modelling and control for lettuce growth” (oral) – AgriFoRwArdS Seminar Day and Quarterly Research Progress Meeting [December 2024] – Lincoln, UK.
“Walkies” (oral) – AgriFoRwArdS CDT Summer School 2025: Going to the Dogs! [June 2025] – Lincoln, UK.
“Data-Driven Greenhouse Modelling and Control for Optimal Lettuce Growth” (oral) – Joint Robotics CDT Annual Conference 2025 [September 2025] – Oxford, UK.
“Data-Driven Greenhouse Modelling and Control for Optimal Lettuce Growth” (poster) – Joint Robotics CDT Annual Conference 2025 [September 2025] – Oxford, UK.
Other activities
- Discussion Panel member at Towards Autonomous Robotic Systems (TAROS) 2024 – Discussion topic: Challenge-driven Postgraduate Training [August 2024].
- Research Assistant in Selective Harvesting for Delicate Produce at the University of Lincoln [December 2024].
- Contributed to AgriFoRwArdS CDT news article ‘Emmanuel visits Crystal Heart Salad‘ [May 2025].
- Hosted and competed in the East Midlands Home Robotics Competition 2025 [June 2025].
- Provided planning and technical support for the LCASTOR Team entry at the International Field Robotics Event (FRE) [June 2025].
About me
I am from a community deeply rooted in agriculture. Motivated by a desire to address the pressing challenges facing his community, particularly in the agricultural sector, I decided to join the CDT. I particularly like the variety of research and networking opportunities offered by the CDT. My favourite thing about Lincoln is the friendly atmosphere. Many residents and visitors comment on the friendly and welcoming atmosphere of Lincoln. It’s often seen as a safe and pleasant place to live. I am going to be conducting my PhD at the University of Lincoln. My career goal is to become an expert in precision agriculture, where I use robotics and AI to optimize farming practices. This role involves data analysis, sensor integration, and the development of algorithms to improve crop yield and resource utilization. I am also looking forward to starting my own company or joining a startup in the agtech sector, where I can innovate and develop new agricultural technologies such as autonomous tractors, drone-based crop monitoring, or AI-driven decision support systems. Having a strong passion for teaching, I would like to work as a research scientist in academia, government research institutions, or private companies to develop cutting-edge robotic systems and AI algorithms tailored to agriculture. This role involves designing, implementing, and evaluating robotic solutions for tasks such as harvesting, weeding, and monitoring crops. In my spare time, I enjoy listening to music or reading self-development books.
MSc Project
A Data-Driven Approach for Optimal Control of Autonomous Greenhouse for Lettuce Growth
This project presents a comprehensive data-driven approach for the optimal control of autonomous greenhouses, specifically tailored for the growth of lettuce. By integrating IoT (Internet of Things) technology, advanced data driven modelling and control techniques , and big data analytics, we aim to develop an intelligent and adaptive greenhouse environment that maximizes lettuce yield and quality while minimizing resource consumption and environmental impact. Our methodology involves deploying a network of sensors for real-time monitoring of critical environmental parameters, implementing a centralized data management system, and developing machine learning models to predict optimal growing conditions. We further employ reinforcement learning algorithms to continuously refine control strategies through feedback loops.
The project also focuses on designing an autonomous control system capable of real-time adjustments, ensuring the greenhouse maintains optimal conditions. Extensive validation and testing will be conducted to measure system performance against traditional methods. The anticipated outcomes include significant improvements in lettuce yield and quality, reduced resource usage, and operational cost savings. This innovative approach not only enhances food production efficiency but also contributes to sustainable agricultural practices, offering a transformative solution for modern farming.
PhD Project
Reduced-order modelling and control of a greenhouse environment to optimize salad production
Recent advancements in robotic and autonomous system for agriculture pave the way toward the realization of autonomous greenhouse where yield could be optimized, and resources reduced. The vast amount of data that can be collect has the potential to encapsulate growers’ knowledge into a quantitative, and controllable system, however the exact response of cultivars to environmental change and to diverse sensory input is still elusive.
Within this project, the candidate has the opportunity to work on an advanced glasshouse prototype for precision agriculture. The candidate, supervised by the CDT academic team, will collaborate with a leading UK glasshouse company to:
- Collect crop data (of leafy salad) with an unprecedent level of details, and analyse them to reveal the fundamental relationships between crop growth and environmental control
- Derive a novel mathematical model of the greenhouse system, with a generic approach that can be applied for the derivation of other dynamical systems
- Apply modern control techniques to the derived model, to exploit environmental control of the autonomous greenhouse to maximise yield and minimize resources
Emmanuel’s PhD project is being carried out in collaboration with Crystal Heart Salad, under the primary supervision of Dr Athanasios Polydoros.