Human Assisted UAV-UGV Cooperative Tracking of Urban Dismounts

Period of Performance: 09/29/2008 - 01/31/2011

$749K

Phase 2 SBIR

Recipient Firm

SET Assoc. Corp.
1005 N. Glebe Rd.Suite 400
Arlington, VA 22201
Principal Investigator

Abstract

SET Corporation proposes a cooperative, layered-sensing system utilizing UAVs and UGVs, assisted by human operators, to detect, track and identify elusive dismounts in urban environments. Ground sensors maneuver in close proximity to the target, opportunistically capturing high resolution images for target fingerprinting. Airborne sensors provide better field of views as the dismount moves to partially obscured areas. The human operators provide innate pattern recognition skills, guiding the system in acquiring dismounts, and directing the system on saliency of its target fingerprint features.The operator, viewing stabilized sensor streams, exercises command and control through a touchscreen data table. SET s unique, patented target fingerprinting techniques include the Double Helical Signature gait analysis, providing a capability to recognize distinguishing features such as gait frequency, stride length, and object-carrying activity. The techniques also employ pose-robust appearance recognition, based upon shape matching and geometric transform to aid in recognizing and tracking pedestrians as they maneuver and change their relative pose. Our systems approach leverages SET s expertise in dynamic sensor management and robust-performance dismount and vehicle tracking algorithms, in combination with extensive guidance, navigation, control and simulation capabilities of our partner, Brigham Young University (BYU). Intense collaboration with AFRL is incorporated to facilitate demonstrations and field experiments.