Copyright © 2001, HRL Laboratories, LLC. All Rights Reserved Progress in Pheromone Robotics David Payton, Regina Estkowski, Mike Howard HRL Laboratories, LLC 3011 Malibu Canyon Road, Malibu CA 90265 {payton, regina, mhoward}@hrl.com Abstract: In our Pheromone Robotics research, we are developing techniques for coordinating the actions of large numbers of small-scale robots to achieve useful large-scale results in surveillance, reconnaissance, hazard detection, and path finding. This paper describes progress in the areas of world- embedded computation and world-embedded displays that are enabled by our use of the biologically inspired notion of a “virtual pheromone.” We describe the pheromone logic primitives that enable our robot collective to become a distributed computing mesh embedded within the environment, while simultaneously acting as a physical embodiment of the user interface. 1. Introduction Emerging technologies in micro machining and MEMs hold the promise of creating extremely small robots. Although limited in size and power, such robots can work together in large numbers to conceivably accomplish a wide range of significant tasks [5],[7],[10],[15], including surveillance, reconnaissance, hazard detection, path finding and payload conveyance. Coordinating and interacting with very large numbers of robots involves issues not encountered when dealing with one or a few robots [2],[8],[11]. Coordination schemes that require unique identities for each robot, explicit routing of point-to-point communication between robots, or centralized representations of the state of an entire swarm can be overwhelmed when dealing with extremely large numbers. To assure that our multi-robot system is both scaleable to large robot swarms and tolerant to individual robot failures, we focus on techniques that effectively limit each robot’s interactions to its own local neighborhood. We are inspired by techniques used by ants and termites for communication and coordination [4],[9],[17]. We implement “virtual pheromones” using simple transceivers mounted on each robot. Like their chemical counterparts, our virtual pheromones facilitate simple communication and emergent coordinated movement with only minimal on-board processing. But virtual pheromones go a step further, transforming a robot swarm into a distributed computation grid embedded in the world. This grid can be used to compute non-local information about the environment such as bottlenecks and shortest paths, in ways that are foreign to insect colonies. In our approach, we envision a scenario in which a rescue team enters an unfamiliar building after a disaster, and needs to quickly locate any survivors. We imagine multiple tiny robots released at the entrance to the building. Using simple attraction/repulsion behaviors, these robots quickly disperse into the open spaces. Upon detection of a survivor, a robot emits a virtual pheromone message signaling the discovery. This message is propagated locally between robots only along unobstructed paths, producing a gradient as it is propagated. Ultimately, the message makes its way back to the entrance where rescue team members can now follow the pheromone gradient to the survivor. To do this, the robots themselves serve as a distributed display of guideposts leading the way along the shortest unobstructed path. We are pursuing this vision with 20 small robots specially designed to support our virtual