Autonomy of Mixed Initiative Large Teams

What Is It?

Autonomy of Mixed Initiative Large Teams Program is an active research program focused on maturing technologies needed to automatically track entities (i.e., persons, vehicles and groups) using signatures discoverable in disparate data. 

How Does It Work?

The program is developing signature-based entity detection and tracking from imagery, unstructured text and network data. Additionally, research is ongoing on signature disambiguation and signature association.   

What Will It Accomplish?

This enabling capability program has the potential to provide the Navy with efficient data storage, retrieval and presentation of everything that is known about an entity, regardless of the information’s source.

The Autonomy of Mixed Initiative Large Teams research initiative is focused on optimizing sensor fields with supporting applications services to automatically track low-level entities in complex feature space.

This enabling capability research program ill mature technology needed to automatically complete valued work tasks on behalf of the analyst and the commander.  Specific work is ongoing in several areas.

Work continues on technology that will enable a sparse disparate sensor field to continuously track dismounts in areas of interest using standoff face-matching, soft biometric feature extraction, voice recognition and TTL. Vehicle tracking over even larger areas—based on appearance, color and motion reasoning—has been matured using wide area and tower based imagers. 

The development of sensor fields that are able to work together to serve the specific needs of an ongoing mission has also been supported by this program. Behavior recognition has been supported by several programs at the Office of Naval Research and remains an interest area and an investment area of this program. Many behaviors of interest can now be automatically detected in real time. The discovery of hidden weapons, explosives and other equipment of interest continues to be a critical information requirement of commanders, so ONR has maintained an investment in through wall sensors.

Additionally, the sheer volume of useful but unstructured data sources requires consideration of and maturation of natural language processing to automate the extraction of entities, word frames, associations and themes. All of the sensor fields described are matured during each year and evaluated as part of the yearly Empire Challenge Experiment Series, also referred to as Green Devil.

Research Challenges and Opportunities:

  • Automated natural language understanding
  •  Adaptive tracking algorithms
  •  Video analytics
  •  Soft biometrics
  •  Link/network discovery
  •  Autonomous task driven sensor fields

Point of Contact:

Martin Kruger
(703) 696-5349
martin.kruger1@navy.mil

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