icon-carat-right menu search cmu-wordmark

Multi-Agent Decentralized Planning for Adversarial Robotic Teams

Poster
This poster, prepared for the SEI 2016 Research Review, visualizes technology development to allow one person to command an entire swarm of UAS.
Publisher

Software Engineering Institute

Abstract

For the past four years; the SSD CPS-ULS group has been working on technologies to enable one human operator to control and interact with a team of autonomous; unmanned systems. In FY16 MADPARTS; we focused on defensive algorithms that protect a human operator or an important asset from a mobile adversary. We demonstrated our line-of-sight prevention algorithms in simulated quadcopters and in real-world demonstrations with unmanned surface vehicles in lakes near Pittsburgh. The algorithms resulted in line-of-sight prevention at over 99% success rates in simulations against mobile adversaries