Tag Archives: D. Hofstadler

Flora Robotica: Exploring Bio-Hybrids

Flora Robotica is tagged,  “Societies of Symbiotic Robot-Plant Bio-Hybrids as Social Architectural Artifacts”

This EU funded project seeks to create “robot-plant bio-hybrids” that are, and I quote, “an embodied, self-organizing, and distributed cognitive system.

Essentially, the plants guide the “growth” of modular robots, and the robots scaffold the growth of the plants. The emphasis is to have an “equal” role for robots and plants, not to be some form of robot gardener. The system operates at “plant speed”, and grows “meaningful architectural structures”, a version of Morphogenetic Engineering unguided by human intervention.  (Or mostly unguided.  I’m not totally sure.)

This project is said to be basic research, AKA, “curiosity driven” research. The goal is to see what can be done and what happens when you do it, not to solve a specific problem. For that reason, it isn’t completely fair to question what the point is. It doesn’t have to have a point.

Nevertheless, I have to really wonder what could come from this. In other word, this project succeeds in making me think.

Part of the game is to make a robot swarm that behaves like a “robot plant”. I’m not sure how original this concept is, but I can see that pursuing this metaphor might lead to some insights into self-assembly. We’ll see.

There are some vague notions of a “social garden”, where the robots are connected to the Internet, and humans somehow interact digitally, subtly influencing the plant-robot structure through some kind of crowd sourcing. (This idea is far from clear to me from their descriptions.) It’s hard to say what this might accomplish, though is it clear that you could do it without the bio-hybrid system.

The investigators are interested in greener cities (in several senses of the word “green”). They mention the possibility of architecture that is self-healing and in tune with the environment. That sounds fine, but why are bio-hybrids the right approach? We already have green areas sustained by human-animal-plant synergy. Why replace humans with robots?

Their plant-plus-woven-structure is kind of cool, and its neat for plants to build their own trellis (and for the trellis to feed its plants). But how does this do anything interesting to anything outside the symbiosis. For example, are these structures suitable for human use? Do they create a vibrant ecology? Do they even create a sustainable bio-hybrid?

Sustainability is actually and interesting question here. Plants generally sustain themselves through a variety of biological processes. Robots generally have completely different sustenance (usually human assisted). Just how would a robot-plant bio-hybrid sustain itself? How would plants sustain the robots? How would robots acquire inputs to sustain the plants? I think something is missing from this picture.

I can also see philosophical and strategic questions here, assuming the idea can be made to work.

What is the benefit to plants from such a symbiosis? Plants are already self-sufficient and highly successful. Adding the complexity of robot cognition and “steering” might enable plants to survive slightly better, but mostly it serves the needs of the unseen human masters.

On the other hand, how does this symbiosis benefit the robots? Anything the plants do could be done by adequate human assistance, or human plus digital systems. What do the slow moving, very dumb plants have to offer the robots? I’m not sure.

Overall, this project looks like a solution to a range of non-problems. As Evan Ackerman put it, “It’s hard to see where all this is going, to be honest, but the vision is intriguing“. But it made me think, so it is a success in that. We’ll see what else comes from this work.


Project publications:  here.  Videos: here

  1. H. Hamann, M. Wahby, T. Schmickl, P. Zahadat, D. Hofstadler, K. Stoy, S. Risi, A. Faina, F. Veenstra, S. Kernbach, I. Kuksin, O. Kernbach, P. Ayres, and P. Wojtaszek. Flora robotica-mixed societies of symbiotic robot-plant bio-hybrids. In Proceedings – 2015 IEEE Symposium Series on Computational Intelligence, SSCI 2015, 2015, 1102-1109. https://www.scopus.com/inward/record.uri?eid=2-s2.0-84964937762&partnerID=40&md5=ddf147e65f827b13630a666a4b7b3271

 

Robot Wednesday