Using Evolving Curves to Track Dynamic Boundaries

Period of Performance: 08/31/2004 - 08/31/2005

$750K

Phase 2 STTR

Recipient Firm

Intelligent Automation, Inc.
15400 Calhoun Dr, Suite 190
Rockville, MD 20855
Firm POC
Principal Investigator

Research Institution

Duke University
2200 W. Main St, Suite 710
Durham, NC 27705
Institution POC

Abstract

In the phase I effort, IAI successfully demonstrated the feasibility of co-operative swarm based approach to boundary tracking. This approach is relevant to applications such as tracking oil spills using a swarm of robots, tracking enemy troop movement using a swarm of UAV's (Future Combat System (FCS) type scenario), and tracking of static and moving mine fields . In Phase II, we continue our efforts to further develop and optimize Phase I algorithms and develop software for distributed, multi-robot detection and surveillance of rapidly changing perimeters. The Phase II effort will specifically focus on the following tasks (i) Further developing the architecture for optimally distributing the tracking algorithms across a robot swarm (ii) Swarm control to optimize boundary observability (iii) Cooperative resource allocation schemes for ad hoc communication in the swarm (iv) Tracking under lack of information (v) Evaluation of alterantive scalable simulation approaches; (vi) Agent-based software development and Integration (vii) Simulation testing and tuning; (viii) Development of a hardware test-bed with swarm of robotic platforms for advanced technology demonstration and (ix) Commercialization and technology transition.