Using Plants As Sensors To Create A Global Monitoring System

With the “Internet of plants,” the divisions between the human and natural world will start to blur.

First there was an Internet of computers. Then came the “Internet of things.” Is the Internet of plants next?


It may be years off. But PLEASED, or Plants Employed As Sensing Devices, is developing some of the early building blocks. Just as humans are likely to be increasingly connected to machines, creating one seamless sensing network, in the future we may connect plants as well. The division between the natural world and the human world may begin to disappear.

Plants give off electrical signals just like humans do when they move. By analyzing and classifying these signals, it’s possible to tell what stimulus may have been the cause. So, there’s a signal for, say, acid, or a chemical, or fire, or many hundreds of things. Once you can read the signals accurately and tell them apart, you can begin to use plants as biosensors. The trees come into the network.

See PLEASED’s video below:

“Say I apply some acid to the plant, I see the signal generated and can say ‘Okay, an acid has been applied.’ This is the first step to use pants as biosensors to provide feedback on the environment in which we live,” says Andrea Vitaletti, who leads W-Lab, a spin-off from the University of Rome. His company is working with several European universities on PLEASED.

So far, researchers have worked with tomatoes, cucumbers, eggplants, and tobacco plants, gradually building up a wider data set. In time, they want to develop algorithms for tiny devices that they’ll attach to plants. That could be useful for a whole range of things, from precision agriculture and organic food certification to environmental monitoring.

“The vision of the project is we will have very tiny sensor nodes that we will plug into plants,” Vitaletti says. “They will recognize signals from the plants, then communicate wirelessly to other devices and create a global picture.”


It’s still early days. “But we’ve shown that the process is possible,” he adds. “Now it’s a question of increasing the quality and size of the data set to test the algorithm.”

About the author

Ben Schiller is a New York staff writer for Fast Company. Previously, he edited a European management magazine and was a reporter in San Francisco, Prague, and Brussels.