top of page
R. Gary Raham

Plantoids and GrowBots

By R. Gary Raham:


Robot Explorers Inspired by Plants


I confess: even biologist/gardeners can be a little zoocentric—biased towards creatures with (mostly) organized brains that move around, doing interesting things. We tend to think about plants as attractive and useful objects or tasty consumables that squelch hunger pangs at meal times. We fail to remember that plants are living brethren that have run evolution’s survival gauntlet for eons using—even pioneering—many of the same biochemical pathways we enjoy. Plants, rooted in place, just live in the slow lane. Now, however, a few biologists like Barbara Mazzolae, are beginning to try and see the world from a plant’s perspective. The result may soon be robots that explore the universe in a much more phytocentric way.

Barbara Mazzolai holds the title of Associate Director for Robotics and Director of the Bio-inspired Soft Robotics Laboratory at the Istituto Italiano di Tecnologia located in Pontedera, Italy. In the early 2010s Mazzolai was excited by efforts of engineers to create robots inspired by nature. These robots took the form of creatures like geckos, dragonflies, and octopuses—all animal forms that could walk, swim, or fly across assorted landscapes. But Mazzolae realized that plants had unique skills of their own. They possess root systems that actively explore the soil for water and nutrients using far less energy than worms, moles or drill bits. Why not design robots based on plants’ unique brand of intelligence?


Yes, intelligence. The concept is still somewhat controversial, but not as outrageous as one might think. Researchers Frantisek Baluska and Stephano Mancuso argue that plants “possess a plant-specific intelligence with which they manipulate both their abiotic and biotic environment, including climate patterns and whole ecosystems.” In his 2013 book, What a Plant Knows, Daniel Chamovitz discussed many of the ways plants interact with us and other living creatures to promote their own survival.


Dodder plants (Cuscuta pentagona), for example, must find tomato or other plants to parasitize or they die. They do so by growing toward the odor of their prey in lazy spirals that eventually contacts a host stem. Willow trees injured by tent caterpillars release chemicals that serve as alarm signals for other downwind plants. Although plants don’t have eyes, they do have primitive photoreceptor systems that lock them into the same kind of circadian (daily) rhythms we notice when we experience jet lag. Mimosa pudica, the so-called sensitive plant, along with certain other plants, respond dramatically to touch by collapsing their leaves quickly. But they can learn to squelch that response, if poked repeatedly with no adverse results. And mineral-starved plants like the Venus flytrap use trigger hairs that depolarize like a nerve cell when touched by an insect. As a result, lobe-shaped, nastily barbed leaves snap shut on prey to leach them of valuable nitrogen compounds.


In Planta Sapiens, The New Science of Plant Intelligence (W. W. Norton & Co., 2022), Paco Calvo with writer Natalie Lawrence make their case that studying plant neurobiology is a real thing. They say “Exploring plants’ ways of doing things, seeing into their worlds, potentially as active participants in these projects, not just as passive tools, can help us explore worlds beyond Earth.” This idea, of course, leads directly to Mazzolae and her plantoid robots. Exactly how could such a robotic device work?


Dario Floreano and Nicola Nosengo describe Mazzolae’s work in their article “The Plant-Inspired Robots That Could Colonize Mars,” adapted from their 2022 book, Tales from a Robotic World, How Intelligent Machines Will Shape Our Future. Mazzolae argues that plant roots perform two jobs exceptionally well: they anchor and support plants while also probing through soil in search of water and nutrients. Robots with similar skills might serve as excellent explorers on planets like Mars with sandy soil and low gravity.


Mazzolai envisions a probe striking Mars like a dart. “The impact would dig a small hole in the planetary surface, inserting a ‘seed’ just deep enough in the soil, not too different from what happens to real seeds. From there, a robotic root would start to grow by pumping water into a series of modular small chambers that would expand and apply pressure on the soil.” Mazzolai knew that she had to figure out how to create a growing robot (a hardware problem) such that its “roots” could collect and share information and use it to make collective decisions (a software problem.)


The growth mechanism, rather like a 3D printer that can add material continuously behind the root tip as it probes the soil, works its way into natural fractures. “A plastic wire is wrapped around a reel stored in the robot’s central stem and is pulled toward the tip by an electric motor. Inside the tip, another motor forces the wire into a hole heated by the resistor, then pushes it out, heated and sticky, behind the tip.” The structure behind the moving tip hardens to become a solid tubular structure.


GrowBots, also plant-like robots, work against gravity and probe space for an anchor point, like dodder plants. The upper trunk of a GrowBot might extend soft, foldable leaves filled with a synthetic chlorophyll that could trap light for energy. Together, plantoid robots and GrowBots might even be able to build infrastructure on a planet ahead of colonization efforts. The technology isn’t there yet for creations like these, but Mazzolai does have the attention of the European Space Agency (ESA). Engineers are becoming less blind—as should we—to the talents of plants. The result could help us explore alien worlds with plant-like efficiency.

R. Gary Raham writes and illustrates both science fact and science fiction. His book, Confessions of a Time Traveler (Penstemon Publications, 2015) contains examples of both, including award-winning contributions to Colorado Gardener magazine. Check out his website at www.rgaryraham.com.

55 views

RECENT ARTICLES:

bottom of page