This program has been archived.
Smart and Autonomous Systems (S&AS)
|Reid Simmonsfirstname.lastname@example.org||(703) 292-4767|
|James Donlonemail@example.com||(703) 292-8074|
|Jie Yangfirstname.lastname@example.org||(703) 292-4768|
Important Information for Proposers
A revised version of the NSF Proposal & Award Policies & Procedures Guide (PAPPG) (NSF 20-1), is effective for proposals submitted, or due, on or after June 1, 2020. Please be advised that, depending on the specified due date, the guidelines contained in NSF 20-1 may apply to proposals submitted in response to this funding opportunity.
The Smart and Autonomous Systems (S&AS) program focuses on Intelligent Physical Systems (IPS) that are capable of robust, long-term autonomy requiring minimal or no human operator intervention in the face of uncertain, unanticipated, and dynamically changing situations. IPS are systems that combine perception, cognition, communication, and actuation to operate in the physical world. Examples include, but are not limited to, robotic platforms, self-driving vehicles, underwater exploration vehicles, and smart grids.
Most current IPS operate in pre-programmed ways and in a limited variety of contexts. They are largely incapable of handling novel situations, or of even understanding when they are outside their areas of expertise. To achieve robust, long-term autonomy, however, future IPS need to be aware of their capabilities and limitations and to adapt their behaviors to compensate for limitations and/or changing conditions.
To foster such intelligent systems, the S&AS program supports research in four main aspects of IPS: cognizant, taskable, adaptive, and ethical. Cognizant IPS exhibit high-level awareness of their own capabilities and limitations, anticipating potential failures and re-planning accordingly. Taskable IPS can interpret high-level, possibly vague, instructions, planning out and executing concrete actions that are dependent on the particular context in which the system is operating. Adaptive IPS can change their behaviors over time, learning from their own experiences and those of other entities, such as other IPS or humans, and from instruction or observation. Ethical IPS should adhere to a system of societal and legal rules, taking those rules into account when making decisions. Each of these research areas requires the IPS to be knowledge-rich, employing a variety of representation and reasoning mechanisms, such as semantic, probabilistic, commonsense, and meta-reasoning.