Learning Estimates of Aggregate Performance Phase II (LEAP II)

Award Information
Agency: Department of Defense
Branch: Air Force
Contract: FA8650-10-C-3006
Agency Tracking Number: F083-180-2340
Amount: $750,000.00
Phase: Phase II
Program: SBIR
Awards Year: 2010
Solicitation Year: 2008
Solicitation Topic Code: AF083-180
Solicitation Number: 2008.3
Small Business Information
12 Gill Street, Suite 1400, Woburn, MA, -
DUNS: 967259946
HUBZone Owned: N
Woman Owned: N
Socially and Economically Disadvantaged: N
Principal Investigator
 Nathan Schurr
 Human-Agent Collaboration
 (781) 496-2453
Business Contact
 Margaret Clancy
Title: Chief Financial Officer
Phone: (781) 496-2415
Email: clancy@aptima.com
Research Institution
Intelligence, Surveillance and Reconnaissance (ISR) systems are becoming more and more complex, with ever increasing fidelity of data and ever increasing numbers of deployable sensors, such as those found on Unmanned Air Vehicles (UAV’s). Furthermore, the current theater of war is becoming increasingly unforgiving, amplifying operator errors into critical mistakes that must be avoided at all costs. LEAP is a tool for system designers that facilitates the rapid, low-cost production of operator models that predict probable performance based on past behavior, current behavior, and environmental factors. These operator models, once integrated into a UAV Control System, inform the automation that evaluates system performance, and enable more effective error mitigation to minimize the consequences of operator error. The LEAP approach, once achieved, has the ability to create more accurate operator models with fewer human experimental subjects than are necessary with traditional learning approaches. Aptima, with the support of Infoscitex Corporation, believes that the LEAP system will provide accurate performance estimates for UAV control systems, as well as other sociotechnical systems, to realize optimal system performance. BENEFIT: • Low-cost rapid creation of operator models that predict probable performance based on past behavior, current behavior, and known environmental factors. • Enhanced overall system performance from informed mitigation of the consequences of human error. • Using innovative machine learning algorithms, model development requires fewer human experimental subjects than traditional approaches. • Increased target detection accuracy and enhanced information collection for UAV systems within the ISR domain. • Robust, generalizable approach for various socio-technical systems which will benefit from an accurate prediction of operator error (e.g., airport luggage screening, monitoring of complex control stations, unmanned vehicle control systems within additional domains).

* Information listed above is at the time of submission. *

Agency Micro-sites

SBA logo
Department of Agriculture logo
Department of Commerce logo
Department of Defense logo
Department of Education logo
Department of Energy logo
Department of Health and Human Services logo
Department of Homeland Security logo
Department of Transportation logo
Environmental Protection Agency logo
National Aeronautics and Space Administration logo
National Science Foundation logo
US Flag An Official Website of the United States Government