Small Unit Mobility Enhancement Technologies

What Is It?

The Small Unit Mobility Enhancement Technologies (SUMET) program aims to increase the platform capability and affordability of unmanned, ground vehicle-enabling technologies to include low-cost, videobased perception systems, advanced video processing techniques, cognitive reasoning architectures and novel algorithm coding methodologies.

How Does It Work?

The SUMET perception system and video processing effort classifies objects by examining certain material characteristics in a spatial scene at the pixel level while utilizing advanced processing techniques. The program develops a cognitive reasoning architecture emphasizing unified cognition theory and learning working memory. Finally, significant emphasis is also placed on developing advanced behavior generation and novel coding techniques such as physicomimetics and genetic programming.

What Will It Accomplish?

SUMET will decrease the Navy and Marine Corps’ reliance on costly sensors; enable systems to maneuver with greater autonomy in complex terrain and decreases algorithm development time; and address the need for advanced unmanned systems to perform logistic functions in the distributed battlespace, removing the logistic burden from the individual Marine.

The Small Unit Mobility Enhancement Technology (SUMET) science and technology effort aims to introduce increased unmanned ground platform autonomous capability into the Marine Corps at an affordable cost.

Commenced in October 2009, the collaborative program leverages robotic efforts taking place in every department, or code, within the Office of Naval Research (ONR).

Key ONR technology programs contributing technology research include Image Understanding and Robotic Perception (Code 31); Unmanned Underwater Systems (Code 32); Developing Autonomy for Unmanned Surface Vehicles by Using Virtual Environments (Code 33); Natural-language Dialogue with Autonomous Systems (Code 34), and; Intelligent Autonomy for Unmanned Autonomous Systems Mission Control Interfaces (Code 35).

Additional collaboration occurs with several other organizations. Performers include the Naval Research Laboratory, University of Maryland, Southwest Research Institute, SOAR Technologies, Nevada Automotive Test Center and Carnegie Mellon University.

Research Challenges & Opportunities

  • Distributed computing networks to process “at-the-sensor” reduce the computational burden on the central processing unit
  • Reasoning algorithms to discriminate between objects and apply context to a near-field spatial scene (e.g., rock-bush, puddle-hole, door-window)
  • Dynamically generated high-level situation awareness model incorporating information not organic to the vehicle—such as threat areas, road and terrain connectivity and traversability—and real-time events and intelligence
  • Advanced autonomous behaviors which integrate bottom-up perception and top-down reasoning to execute doctrinally correct tasks with no human
    intervention

Point of Contact

Jeff Bradel
(703) 588-2552
jeff.bradel@navy.mil

* Some pages on this website provide links which require Adobe Reader to view.