UAV Guidance on GPUs by Nominal Belief-State Optimization

Award Information
Agency:
Department of Defense
Branch:
Air Force
Amount:
$99,950.00
Award Year:
2010
Program:
STTR
Phase:
Phase I
Contract:
FA9550-10-C-0135
Agency Tracking Number:
F09B-T06-0329
Solicitation Year:
2009
Solicitation Topic Code:
AF09-BT06
Solicitation Number:
2009.B
Small Business Information
Apolent Corporation
3333 Bowers Avenue, Suite 130, Santa Clara, CA, 95054
Hubzone Owned:
N
Socially and Economically Disadvantaged:
Y
Woman Owned:
N
Duns:
790217728
Principal Investigator
 Sanjay Rajopadhye
 Associate Professor
 (970) 491-7323
 Sanjay.Rajopadhye@colostate.edu
Business Contact
 Brijendra Sharma
Title: Managing Partner
Phone: (408) 203-6828
Email: bsharma@apolent.com
Research Institution
 Electrical & Computer Engineering
 Edwin Chong
 Colorado State University
Engineering Room B104
Fort Collins, CO, 80523
 (970) 491-6600
 Nonprofit college or university
Abstract
We apply the theory of partially observable Markov decision processes (POMDPs) to the design of guidance algorithms for controlling the motion of unmanned aerial vehicles (UAVs) with on-board sensors for tracking multiple ground targets. While POMDPs are intractable to optimize exactly, principled approximation methods can be devised based on Bellman’s principle. We introduce a new approximation method called nominal belief-state optimization (NBO). We show that NBO, combined with other application-specific approximations and techniques within the POMDP framework, produces a practical design that coordinates the UAVs to achieve good long-term mean-squared-error tracking performance in the presence of occlusions and dynamic constraints. Although the POMDP/NBO combination exemplifies increased tracking performance, this performance gain can be hindered by computational complexity. Implementing computationally intense subroutines intrinsic to the POMDP/NBO approach in highly parallel graphics processing units (GPUs) will allow the realization of our approach on complex systems in near real time. BENEFIT: Improved UAV surveillance technique, Optimal sensor resource management, High Performance GPU library

* information listed above is at the time of submission.

Agency Micro-sites

US Flag An Official Website of the United States Government