Autonomy for Large Teams of Collaborating Unmanned Systems

There is currently a wide range of research efforts examining the control of groups of unmanned vehicles to carry out cooperative or collaborative missions. However, many of the approaches used in these efforts are highly centralized, designed for a narrow set of requirements, and have significant scalability limitations.

Because of the threat that IEDs have posed to troops in Iraq, and the importance that urban and littoral operations have assumed in recent conflicts, unmanned systems are being rushed into service in large numbers. Current systems lack sufficient autonomy and require close human supervision. Moreover, there is very limited capability to share resources or use these systems collaboratively.

ONR’s effort in this area utilizes biological models of social and collaborative behavior by intelligent predators, as well as physics-based (potential fields) models for experimentation and development of principles and methodologies.

This research will enable autonomous systems to reliably maintain persistent surveillance and force protection of complex areas with many heterogeneous platforms/sensors and a small number of humans.

• Ensuring that such technologies are scalable to large and complex areas is critical for operation in complex naval areas such as coastal areas, large ports, riverine, and urban environments.
  
• Ensuring that humans can interact with large numbers of unmanned systems at the appropriate level of abstraction will be critical for ensuring that these systems are flexible and responsive for operational adaptation against unconventional threats by forward-deployed naval forces with limited manning.