Termite colonies build tremendous, complicated mounds, acting with no central control or careful advance planning. These social insects provide a fantastic proof of principle that limited agents, acting independently with access only to local information, can build amazing things. How could we build and program robot swarms—artificial termite colonies—to build things for us? We want a human user to be able to give such a swarm a high-level description of something they want built, and have a guarantee that the system will build it, without the user having to be involved in the details of how it’s done.
Termites have inspired our swarm robotics work in a high-level way—systems of many independent builders limited to local information, coordinating their activity by manipulating a shared environment, and building things much larger than themselves by climbing over partial structures in progress—but the details of what the insects do are in many ways still very much unknown. We’re studying the detailed behavior of mound-building termites and other insects, with the goal of establishing in effect what program the insects are running, and connecting these individual behaviors with the architecture of the structures they build.
Future habitats to support human space exploration will spend most of their time uncrewed, during which periods any number of problems (both anticipated and unanticipated) will inevitably occur, yet it’s critical they be functioning correctly when crews arrive. Teams of autonomous robots will need to be able to perform ongoing maintenance and repair of a variety of systems, potentially including initial deployment and later expansion of the habitat, with very limited support from Earth; and they will need to be able to work effectively and safely with humans during crewed periods.
If evolution is driven by competition, and an organism’s main competitors are others of its own species, why should cooperation be as widespread and successful as it evidently is? We use spatial models to help understand mechanisms behind the evolution of altruistic and self-restraining behaviors. Long-term feedback through changes to the environment, as organisms shape their environment and that of their descendants, can have counterintuitive results: Behaviors against the self-interest of individual organisms—up to and including death—can be favored by selection in the long term.
Technologies for manipulating DNA and other biomolecules have reached a point where we can use them to construct nanoscale devices. This is swarm robotics on another scale altogether—extremely limited agents, in truly enormous numbers. We’re working on ways to advance the capabilities of both the individual agents and the collectives that comprise them, looking to extend traditionally macroscale swarm-robotic tasks into the nanoscale domain, including exploration, mapping, and construction.
We’ve been involved with various educational robotics projects, including Root, a whiteboard-based robot designed to make learning to code tangible and accessible for all ages; and AERobot, a platform for introductory programming and robotics classes, inexpensive enough to let every student have (and keep) their own robot.