What is Dynamically Controlled Environment Agriculture (DCEA)?

Combining mechanistic models with machine learning for advanced dynamic control in CEA.

Agriculture practices often include inefficiencies imprecise input control, which has caused severe consequences to the environment. Controlled environment agriculture (CEA) could be a potential solution to mitigate the impacts of agriculture, with a tighter control of all inputs. Presently, however, current CEA practices are not always dynamically controlled. The integration of machine learning and mechanistic modelling with biofeedback into hybrid, modular models could facilitate a transition to a more dynamically controlled environment for agriculture (DCEA).

A recent paper by Abigail Rae Cohen, discusses the several pathways to hybridisation using modern sensing, latent state monitoring, and technological interventions to optimise cultivation and efficient use of resources. This integration will require collaboration between researchers with physiological and mechanistic understanding and professionals with phenotyping and sensing expertise. Through these types of collaborations, it is our hope that we can generate robust, scalable, and efficient prediction and control models for sustainable produce cultivation.

UKUAT has brought together the author of the paper, Abigail Rae Cohen, and Tony Pridmore, Professor of Computer Science at the University of Nottingham, to talk about this exciting idea of DCEA and how this could be brought to life.

At the end of the panel discussion, the panellists will answer questions from the audience.

About the speakers:

Abigail Rae Cohen

Abi is a researcher and entrepreneur that has worked in sustainable AgriTech for over 15 years. She’s started multiple ventures and received funding from the Gates Foundation to merge waste and agriculture in rural Kenya in 2011. Her work in hydroponics automation was featured in the PFSK in 2013. In 2019, Abi won the Ideas 2 Serve competition at Georgia Tech’s Scheller College of Business exploring entrepreneurial opportunities at the sanitation-water-food nexus. Now a PhD student in Environmental Engineering, Abi’s research focuses on circular food systems design and implementation with a focus on upcycling human-derived nutrients to create closed-loop sanitation and agriculture systems. 

In an effort to create these sustainable, circular food-waste-food loops, Abi’s focus involves the integration of machine learning and mechanistic modelling with biofeedback to create hybrid, modular predictive models that can facilitate a transition to a more dynamically controlled environment for agriculture. Her research outlines several pathways to hybridization using modern sensing, latent state monitoring, and technological interventions to optimize cultivation and efficient use of resources.

Prof Tony Pridmore

Tony Pridmore is Professor of Computer Science at the University of Nottingham, where he is Director of Research and leads the Computer Vision Laboratory. His research interests centre on the development of image-based plant phenomics technologies and, increasingly, the creation and operation of large-scale phenotyping infrastructures.

Tony is co-Founder of and serves on the Management Board of the Hounsfield Facility, a unique installation providing automated extraction of 3D structural descriptions of plants from X-ray data. He is a member of the Imaging and Image Analysis Working Group of the International Plant Phenotyping Network and the International Scientific Advisory Committees of the University of Saskatchewan’s Plant Phenotyping and Imaging Research Centre (P2IRC) and the German PhenoRob Cluster of Excellence.

Tony Pridmore is Associate Editor of Plant Phenomics and Director of the UKRI Technology Touching Life Network PhenomUK.