Title: Reinforcement Learning-Based Control and Safety Verification for Aerial Robotics

Date: Wednesday, April 15th, 2026

Time: 10:00 AM - 12:00 PM ET

Location: TSRB 523a

Virtual Link: Microsoft Teams

Christian Llanes

Robotics PhD Candidate

School of Electrical and Computer Engineering

Georgia Institute of Technology

 https://christianllanes.com/

Committee:

Dr. Samuel Coogan (Advisor) - School of Electrical and Computer Engineering & School of Civil and Environmental Engineering, Georgia Institute of Technology

Dr. Kyriakos G. Vamvoudakis - School of Aerospace Engineering, Georgia Institute of Technology

Dr. Panagiotis Tsiotras - School of Aerospace Engineering, Georgia Institute of Technology

Dr. Sehoon Ha - School of Iterative Computing, Georgia Institute of Technology

Dr. Glen Chou - School of Cybersecurity and Privacy & School of Aerospace Engineering, Georgia Institute of Technology

Abstract: 

Learning-based techniques have become increasingly popular as a tool for improving autonomy of aerial robotics platforms. However, these methods typically lack safety guarantees using formal verification. We contribute to autonomy for aerial robotics by using learning-based techniques and a safe verification method for supervising unverified controllers. Specifically, we propose a navigation framework for urban air mobility and aerial robotics that leverages actor-critic model predictive control  for urban air mobility, and using reachability with control barrier functions for verifying these learning-based methods. We leverage mixed monotonicity to quickly overapproximate reachable sets for aerial drone dynamics in real-time. Additionally, we explore multi-agent reinforcement learning to train drone swarms to cooperate in a multi-agent pursuit evasion problem. We also extend actor-critic model  predictive control to multi-agent reinforcement learning with a demonstration of improved robustness over standard multi-layer perceptron policies. Finally, we contribute software tools for the aerial robotics community for simulating hardware code for the Crazyflie nano quadrotor.