Watch these termite-inspired robots in action
Termites are some of nature’s most magnificent architects that can easily build complex mounds that exceed their own size by several orders of magnitude – occasionally reaching up to nine meters in height. Paradoxically enough, each of these tiny insects does not have even a remote idea of what kind of structure it is building, nor does it receive orders from any termite authority. In fact, the termites’ architectural prowess makes no sense except in the context of swarm intelligence.
This unusual approach to construction inspired researchers at Harvard University’s TERMES project to build little robots that mimic the behaviour of termites. Similarly to individual termites, each of these zealous robots is an autonomous unit that can carry out the required work independently from its siblings. And although it may not seem intuitive – given the modern context of interconnectivity – the robots have no wireless communication or localisation capabilities, but rely exclusively on information available from their immediate surroundings.
In other words, the robots’ behaviour is based on the concept of stigmergy, which implies that complex self-organisation may arise from a very simple mechanism, in this case allowing each independent robot to be stimulated by traces left in the environment by its own previous activity or that of other robots.
Thus, the robots have no clue as to the state of the structure they are building or what other robots are up to; they just do what seems right according to the established traffic rules (which vary depending on the structure). Basically, the number of robots in the swarm can be increased or decreased at any moment without affecting the outcome: the remaining robots will continue working until the job is done.
This bio-inspired approach can flourish in hazardous construction scenarios such as in disaster areas or extraterrestrial environments. The research can also shed new light on the very subject of swarm construction, which is not yet fully understood by science.
You can also learn about other research projects carried out by Harvard’s Self-organising Systems Research Group HERE.
