Distributed Learning and Controller Design for Assured Autonomy
Abstract
Systematic ways to integrate diverse, heterogeneous, and possibly time-varying components into complex autonomous systems while guaranteeing system level properties define a holy grail in the science of assured autonomy. With much work being done already on topics such as safe machine learning or reinforcement learning to obtain guarantees on performance and safety of learning enabled autonomous systems (including through this program), this research effort focused on the next challenging step: how to provide guarantees on assured autonomy in a multi-agent system where multiple learning components are interacting. The project successfully completed design and analysis of new algorithms for distributed learning in both competitive and cooperative environments.
Document Details
- Document Type
- Technical Report
- Publication Date
- Aug 01, 2022
- Accession Number
- AD1176161
Entities
People
- Vijay Gupta
Organizations
- University of Notre Dame