Assured Autonomy in Contest Environments (AACE)
Fall 2021 Review
Venue
The program review will be held through a video conference (zoom) format on April 30th. The meeting is an open meeting. The presentations will be recorded and posted.
Agenda for April 30th (times are EST)
- 10:00-10:15 Welcome, Center Overview, Workforce Development and AFRL Collaboration
- 10:15-10:35 Topologically Inspired Path-Planning with Intermittent Feedback
- 10:35-10:55 Probabilistic Conformance for Cyber-Physical Systems
- 10:55-11:15 Characterization of Satellite Swarms Under Non-Keplerian Dynamics – Model Development
- 11:15-11:35 Exploiting environmental effects in LEO for attitude and orbit control with the Drag Maneuvering Device: current and future challenges
- 11:35-11:45 Break
- 11:45-12:00 Adaptive Safety for Hybrid Systems
- Ricardo Sanfelice (Slides), (Video for this talk and next talk)
- 12:00-12:10 Adaptive Safety With Multiple Barrier Functions Using Integral Concurrent Learning
- Axton Islay, Masoumeh Ghanbarpour, Ricardo Sanfelice, and Warren Dixon (Slides)
- 12:10-12:30 A Hybrid Gradient Descent Algorithm
- 12:30-12:40 Prototype Validation in Multi-Agent Communication
- 12:40-13:00 Preserving Privacy in Human-Machine and Agent Interaction
- 13:00-13:20 Safe learning and verification of neural network controllers
- 13:20-13:50 Lunch
- 13:50-14:20 Asynchronous Distributed Gradient Descent with Exponential Convergence Rate via Hybrid Methods
- 15:00-15:20 Learning Optimal Strategies for Temporal Tasks in Stochastic Games
- 14:20-14:30 (Deep)Learning, State Estimation, and Event-Triggered Communication in Multi-Agent Systems
- 14:40-15:00 Online Policy Learning for Unknown and Varying Tasks in Adversarial Environments
- 15:20-15:40 Secure Planning Against Stealthy Attacks via Model-Free Reinforcement Learning
- 15:50-16:10 Deep, Cooperative, and Modular Learning for Approximate Dynamic Programming
- 16:10-16:40 Open discussion with AFRL/AFOSR
- AFRL/AFOSR program managers