Summaries - Office of Research & Innovation
Back Automatic Generation of System Plans for Autonomous Unmanned Systems
|Division||Research & Sponsored Programs|
|Department||Naval Research Program|
|Investigator(s)||Berzins, Valdis A.|
|Sponsor||NPS Naval Research Program (Navy)|
The objective of this study is to leverage automatic programming in support of SSC-P strategies for developing unmanned systems autonomy software.
Prior software engineering research gave us breakthroughs in technology that improved the development of computer software and benefited real-time embedded systems. The automatic programming approach pioneered by the Naval Postgraduate School utilized formal representations of system requirements, and in particular timing relationships among operations. Teams of autonomous systems, such as the USN is envisioning deploying in the future, share many of the characteristics that made these formal mathematical methods, scheduling, and code generation so useful. Additional factors include uncertainty and a larger set of possible actions and states.
To apply AI planning algorithms to this problem, research on suitable representations for mission plans and system plans is needed, along with planning architectures and principles for automatically deriving system plans from mission plans. Mission plans should be described at a conceptual level natural for the human operators of autonomous systems and suitable for enabling one operator to control a team of multiple autonomous systems. System plans must be sufficiently clear and detailed to be carried out by fully automated systems.
Mathematical formulations of the problem need to be explored and a rigorous approach needs to be applied. There are set theoretic, statistical, and information theoretic aspects to be considered.
This project addresses a broad set of questions whose solution will contribute to the third offset strategy by speeding up mission planning for autonomous systems, making it more flexible, and less labor intensive.
Proposed work includes several challenging tasks that will take multiple people and two full years to complete. Requested budget is for the first year.
|Publications||Publications, theses (not shown) and data repositories will be added to the portal record when information is available in FAIRS and brought back to the portal|
|Data||Publications, theses (not shown) and data repositories will be added to the portal record when information is available in FAIRS and brought back to the portal|