In today’s dynamic battlespace, human operators are consumed by the supervision of autonomous vehicles, constantly “tweaking” the system to obtain the required information. The Smarts project will bridge the currently disparate fields of machine learning and control theory to enable the development of novel, more intelligent, and more capable autonomous agents.
Smarts seeks to develop new algorithms and systems that will significantly increase:
(1) the ability of autonomous agents to interpret and reason about their environments in complex surveillance missions(2) the ability of groups of autonomous agents to coordinate for tasks of larger scope than individual agent tasks(3) the ability of humans to interact with the system at higher cognitive levels
This will provide the warfighter with time-accurate, continuous, and cognitive-level views of large and complex environments.