A plant reacts intricately but predictably to its environment
Plants interact with their environment and adapt to achieve a specific goal. For example, a plant can extend beyond a shady spot to get as much sunlight as possible on its leaves. Behind him, a complex system enters into force in the plant, but despite this complexity, its reaction to the “shadow” state is rather predictable.
This predictive ability provides perspectives to “calculate” plant behavior. ILVO-UGent researcher Olivier Pieters installed leaf thickness sensors and cameras on strawberry plants and developed a sensor platform with related hardware and software that was accurate enough to effectively measure environmental variables and plant properties (leaf thickness and leaf elongation). Then apply successfully Reservoir Computing (People’s Republic of China) to its manufacturer data.
Computing physical reservoirs
PRC comes from computer science and is used, among other things, to allow a robot to modify its movements in real time in response to visual stimuli. At first glance, a plant looks “simpler” than a robot, but that’s certainly not true in this context. For example, a plant does not always react in exactly the same way to a stimulus, and its behavior also varies depending on other factors such as age and season. Unlike the robotic body, plants change regularly. Just think of the leaf buds in the spring and wither in the fall. This makes it difficult for the PRC, because this method assumes that reactions and memory do not change over time.
The first step, continue with the desired research (and start)
For these reasons, Peters conducted research on fully grown plants for only one week. But he was effectively able to use PRC and the new sensor network to calculate from a measured plant characteristic such as leaf thickness to the (complex) interaction between a plant and its environment, such as light intensity, as well as moisture and photosynthesis rate.
This research on PRC in plants is exploratory research, but it can be called pioneering. Professor Weifels: “This PhD is a good example of robust interdisciplinary research in which a concept from computer science is being tested on plants. This has given us new insights into at least three areas: computer science, electrical engineering and physiology.”
Pieters and his colleagues will now work with the developed platform and some other sensing technologies (including contactless technology) to see how efficiently they map plant characteristics to different plant species. The goal is to be able to monitor as many properties as possible in an easy-to-use way.
Interesting application views
Olivier Peters: “The innovative thing is that our view of plants is changing: from an organism that undergoes changes in the environment (passive) to an organism that processes information (active). The great thing is that the People’s Republic of China It allows for a standardized approach to the response of plants. This means that all plant processes can be studied in a similar way, which is interesting for breeders. It may also be interesting in the long run for precision farming starting from a lot of data and even for greenhouse gardening with climate control. For example, a plant can control its (micro-)climate in a greenhouse across the People’s Republic of China.”
