Dynamic Plants

Researchers: Olivier Pieters, Francis wyffels in collaboration with Tom De Swaef (ILVO), Michiel Stock (UGent-KERMIT), Peter Lootens (ILVO) and Isabel Roldàn-Ruiz (ILVO)

Plants are complex and dynamic organisms that continuously sense changes in their environment and adapt their responses accordingly. However, these changes do not occur uniformly. We are investigating to what extent these differences occur, how they relate to each other, and how they can be used to improve plant productivity.

Plants are highly complex dynamical systems that react to a wide range of external and internal stimuli. Consequently, they exhibit emergent intelligent behaviour due to their sophisticated cells and cellular networks. These cells form the basis of plants’ extensive set of signal receptors and downstream physiological networks. In our research we consider plants as computing units, able to sense and interpret these signals and provide adequate responses that maximise fitness to the prevailing environmental conditions. More specifically, we are investigating whether the framework of physical reservoir computing is applicable to plants. Physical reservoir computing is the extension of reservoir computing to physical media. Instead of a simulated reservoir, a physical medium is used. A wide range of media is already explored within and outside the AIRO group. We now want to extend this to living organisms.

The analogy between a plant-based physical reservoir and general reservoir computing is depicted in the figure above. A wide range of biotic and abiotic factors form the input to the reservoir. Biotic factors include plant-plant interactions, pests, diseases, and temperature. Abiotic factors are precipitation, nutrient availability, for example. The instantaneous plant response is a combination of the current and past environment and is partially observable using sensors such as are RGB cameras, leaf length sensors and bio-impedance sensors. The target variables are formed by linearly combining the readout values from the sensors. Possible target functions include water flow, stomatal conductance and stresses.

Our goal is to show that by exploiting the computational capabilities of plants one can better understand complex plant behaviour and ultimately improve (the conditions for) crop growth. Additionally, it also offers an entirely new way of looking at plant responses, at a much more integrated scale.


  1. Gloxinia—an open-source sensing platform to monitor the dynamic responses of plants
    Pieters, Olivier, De Swaef, Tom, Lootens, Peter, Stock, Michiel, Roldàn-Ruiz, Isabel, and wyffels, Francis
    SENSORS 2020
  2. Development of a quantitative comparison tool for plant models
    Pieters, Olivier, De Swaef, Tom, and wyffels, Francis
    In FSPM2020: Towards Computable Plants; 9th International Conference on Functional-structural Plant Models 2020
  3. Limitations of snapshot hyperspectral cameras to monitor plant response dynamics in stress-free conditions
    Pieters, Olivier, De Swaef, Tom, Lootens, Peter, Stock, Michiel, Roldàn-Ruiz, Isabel, and wyffels, Francis