You are here
Harvard scientists develop a self-organizing thousand-robot swarm
This self-organizing swarm was created in the lab of Radhika Nagpal, Fred Kavli Professor of Computer Science at the Harvard School of Engineering and Applied Sciences (SEAS) and a Core Faculty Member at the Wyss Institute for Biologically Inspired Engineering at Harvard University. The advance is described in the August 15 issue of Science (http://www.sciencemag.org/content/345/6198/795).
Following simple programmed rules, autonomous robots arrange themselves into vast, arbitrary shapes. They use the same kind of robots that we use in Swarm-Organ, kilobots, to build two-dimensional shapes with a thousand-robot swarm. In fact, only a few robot swarms to date have exceeded 100 individuals, because of the algorithmic limitations on coordinating such large numbers, and the cost and labor involved in fabricating the physical devices. The research team overcame both of these challenges through thoughtful design.
Four robots mark the origin of a coordinate system, all the other robots receive a 2D image that they should collectively mimic, and then using very primitive behaviors—following the edge of a group, tracking a distance from the origin, and maintaining a sense of relative location—they take turns moving towards an acceptable position. They have demonstrated a mathematical proof that the individual behaviors would lead to the right global result. You can see a video of this swarm in https://www.youtube.com/watch?v=xK54Bu9HFRw&feature=youtu.be.
This success is encouraging for our project. However, the implementation of this swarm is different from our goals. The objective of Swarm-Organ is to create large swarm shapes and patterns, but without any global positioning information. Instead we make our swarms dependent only on biologically-inspired concepts of local self-organisation. The control system is also biologically-inspired - gene regulatroy networks - and we hope that through this combination, our swarms will be capable of self-healing (for example, automatically making the same shape smaller if fewer bots are available). By focusing on GRNs we are developing a theoretical framework about distributed adaptive control, which will be equally informative to both natural biological morphogenesis, as well as technology.