Problems We Know

What we know, and how we learned it.

Research Programs

  • Developed Parameter Informed Reinforcement Learning (PIRL) — a novel AI-driven framework for system identification that enables autonomous vehicles to learn and adapt to their own dynamics in real time

  • Led multiple competitively funded aerospace R&D programs as principal investigator

  • Designed and validated adaptive control laws for autonomous underwater vehicles operating under nonlinear, time-varying conditions

  • Developed high-fidelity 6-DOF dynamics models for autonomous vehicle programs from first principles through hardware validation

  • Conducted UAS collision avoidance research in collaboration with the FAA

  • Designed robust model-based control architectures for autonomous systems operating in uncertain environments

  • Researched fault-tolerant GN&C architectures for autonomous vehicles

  • Developed software automation tools during a research engagement at NASA Langley Research Center

Vehicle Platforms

  • Unmanned aerial vehicles (fixed-wing and multi-rotor)

  • Autonomous/Unmanned underwater vehicles (AUVs/UUVs)

  • Unmanned surface vessels (USVs)

  • Manned fixed-wing aircraft

Technical Methods

  • System identification — physics-informed and data-driven

  • AI/ML-augmented guidance, navigation, and control

  • Reinforcement learning applied to dynamic systems

  • Model Reference Adaptive Control (MRAC)

  • Classical and modern control design (PID, LQR, H-inf.)

  • Nonlinear control for time-varying systems

  • Digital twin development and high-fidelity simulation

  • Sensor fusion and state estimation

  • Kalman filtering and observer design

  • 6-DOF rigid body dynamics modeling

  • GPS-denied navigation

  • Fault detection, isolation, and recovery (FDIR)

  • Flight dynamics analysis and stability margin assessment

  • Independent technical evaluation and due diligence

Software & Tools

  • MATLAB / Simulink

  • C++ — embedded control algorithm implementation

  • ROS 2 — robotic middleware and hardware integration

  • Python — data analysis, simulation, machine learning

Publications & Presentations

Dr. Nathan Schaff

  • PhD Dissertation: "Parameter Informed Reinforcement Learning for Vehicle System Identification" — Embry-Riddle Aeronautical University, Dec 2025

  • "Online Training and Implementation of Parameter Informed Reinforcement Learning for Aircraft System Identification" — AIAA SciTech 2025 (AIAA 2025-1727)

  • "Indirect Adaptive Control for Autonomous Flight Vehicles Using Parameter Informed Reinforcement Learning" — AIAA SciTech 2025 (AIAA 2025-0347)

  • "Online Aircraft System Identification using Parameter Informed Reinforcement Learning" — AIAA SciTech 2024 (AIAA 2024-0344)

  • Master's Thesis: "Online Aircraft System Identification Using a Novel Parameter Informed Reinforcement Learning Method" — Embry-Riddle Aeronautical University, Dec 2023

Dr. J. Kyle Zelina

  • "Fault Tolerant Adaptive Stabilization for Multirotor Systems with Actuator Failure" — AIAA SciTech 2024

  • "Adaptive Stabilization of Multi-Rotor Systems with Actuator Limits and Transient Mass Distribution" — AIAA SciTech 2023

  • "Dynamic Inversion with Adaptive Augmentation for a High-Speed Guided Projectile" — AIAA SciTech 2023

  • "Adaptive Control for Time Varying Systems with Actuator Dynamics and Sensor Noise" — AIAA SciTech 2022 (AIAA 2022-2029)

  • "Adaptive Control for Nonlinear Time-Varying Rotational Systems" — AIAA SciTech 2021 (AIAA 2021-1120)