Other intelligences is an artist research that aims to explore the communication between plants remotely by using AI tools. Through this project we seek to know more about the plants’ language and behavior and try to understand a better these living beings with we co-habit in the Earth.
This is a processual work in which we have created a neuronal plant network connected to the Internet, composed of plants located in different places around the world. At the same time, each of this living beings is being monitored through the recording of electrophysiology data and environmental data of their surrounding; thanks to environmental sensors that allow us to measure and link the changes in the surrounding of the plant with their behavior.
Several scientific studies have demonstrated that plants have chemical processes that have similar features with the process that take place in the nervous system of animals, thus plants have capacities like memory and therefore they can learn about their environment..Based on plant intelligence and their capacity of learn we created or neuronal plant network. We designed an algorithm makes possible the remotely communication between plants through the training and learning to different stimuli.
In this way we investigate, about the recognition of patterns by the plant, which allows it to identify with which plant of the network they are communicating and thus allow two way dialogue. To do that we have provided each of the plants with tools that allows them sending and receiving signals of light, sound and movement. The entire communication process is being recorded, to be analyzed afterwards with an Artificial Intelligence , in order to recognize patterns that humans cannot perceive with the naked eye.
Through the use of technology as a tool, we have created a network of plants connected to the Internet. A network analogous to the network of roots and mycelia of fungi that take place in the forest that the Professor of Forest Ecology Suzanne Simard began to study almost three decades ago. This woods’ network allows trees and plants to communicate with each other by sharing information and nutrients. In Other Intelligences, our network, is made up of data, algorithms and actuators, and it allows us to investigate alternative relationships with nature and technology through the use of artistic methodologies.
All species on earth are linked by symbiotic and interdependent relationships; the project proposes an innovative point of view on the transformation processes that are shaping the collaboration between species, ecosystems and technologies. To know more about the language and behavior of plants will allow us to learn more about nature and thus better understand our environment.
THE DEVICE: SENSORS AND ACTUATORS
3D printed device with sensors and actuators
A sensor developed by ourselves called Clorofila 3.0 is included in the device. This sensor is based on the electrophysiology of the plants, and allow us to measure the electrical activity that takes place into the plant, depending of the surrounding of this living organism , like changes in light, temperature, sound, atmospheric pressure, etc.
Additionally we placed environmental sensors that allow us to monitor the surrounding of the plants by measuring changes in temperature, sound, proximity,CO2, humidity, etc (see figure 3). This sensors allowing us to compare and link this measures with the biochemical state of the plant, to try to figure out what is affecting to the plant. Furthermore all the data is sending to the Internet every two seconds and recording in a server, which allow us to analyze it.
Also three kind of actuators were placed in the device in order to provide tools to the plants to make possible the remote communication. We designed the actuators by thinking in the well being of the plants.
One of the actuator is to alert the plant that something is happening with itself, the other two actuator allowing the plant to send signals to others that warm the other that something happened.
The actuators are: A CONTACT SPEAKER
To create low frequency sounds that reminds to sounds in the nature, by generating soft vibrations in the soil. This actuator is placed to warn the plant that something is happening with itself
TWO SERVO MOTORS
Integrated with two soft spikes inspired in the branches moved by the wind, it make movements to the plant. This actuator is placed to receive movement signals from other plants.
GROW LIGHT LEDS
Integrated in the device, next to the soil. This actuator is placed to receive light signals from other plants.
CONNECTING PLANTS
Kind of plants connected to the neuronal plant network
According to our experience working with plants, we choose just three kind of house plants, that we consider environmentally sensitive. We asked all the people who joined to the experiment to host on of this plants at home:
We have now 25 plants connected in total, in Norway,Spain, Mexico, Belgium, Uk and Denmark.
Monstera in Oslo receiving light signals from another plant located in other part of the world
We also created two tutorials with some easy steps to connect the device with the sensors and actuators to the plant and to the internet, in order to share all the data in real time.
However Clorofila 3.0 sensor need to be adjusted in each plant, due each plant has different features. We do it remotely through a control panel we developed with this purpose.
Control panel to change the settings of each sensor. For internal work.
Control panel for internal work to see live data of the elctrical activity of theplant and their environment.
Live electrical data from one of the connected plants.
COMMUNICATION BETWEEN PLANTS
Based on plant intelligence and their capacity of learn we created our own algorithm in order to set up our neuronal plant network and we trained the plant to achieve it. All the data of the plants and the activation values are stored for analysis of the evolution of the communication of the plants. Hundreds of gigabits of data are being generated.
In order to achieve an improved functioning we four groups of plants were created. In this way just plants for the same group can communicate each others. In this way, the activation ranges are clearer by giving the plant clear responses to specific values.
After two weeks we could detect patters with the naked eye in the communication between plants. Now we are using an Artificial Intelligence to analyze and figure out this patters, that we like humans cannot detect.
Control panel to see the live communication between plants. For internal work.
How does it work?
All the data recorded by Clorofila 3.0 sensor about the activity of the plant and the environmental data from the sensors are sent via wifi to the Internet every two seconds. The data is sent separated in groups of files ready to processing and analysis to a serve , where is processed and stored.
A program monitors the electrical activity of all plants at the same time and processes the data with a gap of two seconds. At the same time the program is creating the necessary actions for the actuators of each plant.
Through an algorithm we calculate the threshold values that activate the actions in a plant. This algorithm is the same for all plants, thus for equal values we obtain equal outputs.
Once it has been decided whether a plant has the specific measurement intervals for an activation, the values to activate the actuators of that plant are calculated and it is also determined with which plant it must communicate, deciding whether to activate one or another actuator. For similar activation values we have similar results, thus recognizable patterns can be easily created. In other words, if you can observe your plant at home, receiving signals from others; lights, movements. According to the frequency of this lights, how they blinking for how the motors are moved, you can figure out which plant is sending signals.
Now we are working in the use of an Artificial Intelligence in order to detects patterns that we as humans cannot detect between the plants communication
DATA VISUALIZATION
We are using Artificial Intelligence tools in order to process the data and establish the values of the entropy of the processes that are taking place in the communication between plants. An increase in the variation of the data entropy, indicates a beginning of communication preferences between specific plants, thus obtaining the learning keys of the plants and their preferences between the members of the group.
Now, in the last phase of this processual research, we are working in the visualization of the plants communication and the programming of the AI. Meanwhile the network of plants is running and evolve everyday
Project
Work produced with the support of a 2020 Leonardo Grant for Researchers and Cultural Creators, BBVA Foundation.
The Foundation takes no responsibility for the opinions, statements and contents of this project, which are entirely the responsibility of its authors.