Omniscient Planning and Control Environment for the Naval Enterprise

Abstract

We propose to develop a new Naval planning and control framework that enables un-manned underwater vehicles (UUVs) to autonomously account for top-level objectives and mission priorities as well as host vehicle intentions and capabilities. The planning and control framework will enable a human operator to pilot one or more UUVs by provid-ing high-level mission objectives, and it will ensure probabilistic guarantees of mission success and system safety when UUVs are operating collaboratively with manned and un-manned platforms. To ensure that the framework can be used with existing commercial and DoD systems, it will accommodate standard data types and formats needed by othersystems and tools. Computational limitations onboard a UUV will be addressed by utilizing high-performance computing o --board the UUV (e.g., cloud-based computing) for computationally intensive tasks. Promising machine learning approache that we will investigate can perform computationally intensive training tasks on a host platform or in the cloud, while the results of training can be implemented onboard the UUV. We undertake continuous experimentation in the ?eld to ensure that our planning approach can be reduced to practice. Experimentation culminates in full-scale sea-trials during the last yearof this project.

Document Details

Document Type
DoD Grant Award
Publication Date
Jul 27, 2018
Source ID
N000141812627

Entities

People

  • Daniel J. Stilwell

Organizations

  • Office of Naval Research
  • United States Navy
  • Virginia Tech

Tags

Fields of Study

  • Computer science

Readers

  • Defense Acquisition Program Management
  • Distributed Systems and Data Platform Development
  • Unmanned Aerial System (UAS) Autonomous Capabilities and Mission Reconnaissance.

Technology Areas

  • AI & ML
  • AI & ML - Autonomous Systems