Sentient Engineering Autonomy: Learning, Intelligent and Optimal Naval Systems (SEALIONS)
Abstract
Publicly Releasable Abstract Sentient Engineering Autonomy: Learning, Intelligent and Optimal Naval Systems (SEALIONS) is a framework to evaluate the utility and performance of diverse sets of artificial intelligence and machine learning (AI/ML) tools to autonomously operate unmanned systems used by the US Navy and the Marine Corps. The key objective of the SEALIONS framework is to enable an automated prescription of the AI/ML ingredients to carry out the communication, control, guidance and navigation functions of the autonomous system. This is accomplished by implementing and integrating the component AI/ML tools on a given platform to perform a mission function, generating performance data from representative model simulations and using flight experiments, and using datadriven modeling approaches to quantify the performance of the component selection. The conditional performance measure then supports an outer-loop learning process to generate prescriptions for future autonomous system design. In addition to providing a wholistic tool to carry out sensitivity analyses of the mission performance as a function of the component algorithm choices, the SEALIONS framework also serves as a first-cut performance certification mechanism for autonomous systems. Recent advances in data-driven modeling of dynamical systems are proposed to evaluate the performance of the autonomous system being operated by a given selection of AL/ML tool options for guidance, perception and control. While the proposed work aims to use the missions related to the autonomous deployment and recovery of unmanned systems as a use case for the study, the SEALIONS framework is broadly applicable to a variety of other contexts of interest to the US Navy. Advanced AI/ML tools for guidance, perception and control will be implemented using representative experiments at the Land, Air and Space Robotics (LASR) laboratory to support the development of the SEALIONS framework. The proposed work will provide a stimulating educational environment for the next generation workforce.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Mar 11, 2025
- Source ID
- N001742310016
Entities
People
- Manoranjan Majji
Organizations
- Texas A&M University
- United States Navy