Luis A. Pabon

Contacts:

Email: lpabon at stanford dot edu

Luis A. Pabon

Luis is a Ph.D. student in the Department of Aeronautics and Astronautics. His research lies at the intersection of machine learning and control theory, targeting advancements in robotics and autonomous systems. Luis focuses on developing algorithms that enable robots to efficiently learn from their own experience, safely interact with the world around them, and adapt to unfamiliar environments.

Prior to joining Stanford, Luis earned his B.S. in Mechanical Engineering with an Aerospace minor from the California Insitute of Technology. There, he conducted research in time-delayed control, space robotics, and autonomous exploration with both the Burdick Group and the Autonomous Robotics and Control Lab. He founded Caltech Air and Outer Space (CAOS) and held leadership roles on teams that won over $360,000 in the NASA BIG Idea Challenge (2021 and 2022). Luis has also interned at Honeywell Aerospace and the NASA Jet Propulsion Laboratory.

Outside the lab, Luis is passionate about soccer, snowboarding, reading, and making music.

Awards:

  • Stanford School of Engineering Fellowship
  • EDGE Fellowship

ASL Publications

  1. L. Pabon, J. Köhler, J. I. Alora, P. B. Eberhard, A. Carron, M. N. Zeilinger, and M. Pavone, “Perfecting Periodic Trajectory Tracking: Model Predictive Control with a Periodic Observer,” in IEEE/RSJ Int. Conf. on Intelligent Robots & Systems, Abu Dhabi, 2024.

    [BibTeX] [Abstract]

    Abstract: In Model Predictive Control (MPC), discrepancies between the actual system and the predictive model can lead to substantial tracking errors and significantly degrade performance and reliability. While such discrepancies can be alleviated with more complex models, this often complicates controller design and implementation. By leveraging the fact that many trajectories of interest are periodic, we show that perfect tracking is possible when incorporating a simple observer that estimates and compensates for periodic disturbances. We present the design of the observer and the accompanying tracking MPC scheme, proving that their combination achieves zero tracking error asymptotically, regardless of the complexity of the unmodelled dynamics. We validate the effectiveness of our method, demonstrating asymptotically perfect tracking on a high-dimensional soft robot with nearly 10,000 states and a fivefold reduction in tracking errors compared to a baseline MPC on small-scale autonomous race car experiments.

     
    @inproceedings{PabonEtAl2024,
  author = {Pabon, L. and Köhler, J. and Alora, J.I. and Eberhard, P.B. and Carron, A. and Zeilinger, M.N. and Pavone, M.},
  title = {Perfecting Periodic Trajectory Tracking: Model Predictive Control with a Periodic Observer},
  year = {2024},
  booktitle = {{IEEE/RSJ Int. Conf. on Intelligent Robots \& Systems}},
  address = {Abu Dhabi},
  url = {https://arxiv.org/abs/2404.01550},
  owner = {lpabon},
  timestamp = {2024-07-01}
}
  2. J. I. Alora, L. Pabon, J. Köhler, M. Cenedese, E. Schmerling, Z. M. N., G. Haller, and M. Pavone, “Robust Nonlinear Reduced-Order Model Predictive Control,” in Proc. IEEE Conf. on Decision and Control, Singapore, 2023.

    [BibTeX] [Abstract]

    Abstract: Real-world systems are often characterized by high-dimensional nonlinear dynamics, making them challenging to control in real time. While reduced-order models (ROMs) are frequently employed in model-based control schemes, dimensionality reduction introduces model uncertainty which can potentially compromise the stability and safety of the original high-dimensional system. In this work, we propose a novel reduced-order model predictive control (ROMPC) scheme to solve constrained optimal control problems for nonlinear, high-dimensional systems. To address the challenges of using ROMs in predictive control schemes, we derive an error bounding system that dynamically accounts for model reduction error. Using these bounds, we design a robust MPC scheme that ensures robust constraint satisfaction, recursive feasibility, and asymptotic stability. We demonstrate the effectiveness of our proposed method in simulations on a high-dimensional soft robot with nearly 10,000 states.

     
    @inproceedings{AloraPabonEtAl2023,
  author = {Alora, J.I. and Pabon, L. and Köhler, J. and Cenedese, M. and Schmerling, E. and N., Zeilinger M. and Haller, G. and Pavone, M.},
  title = {Robust Nonlinear Reduced-Order Model Predictive Control},
  year = {2023},
  keywords = {pub},
  booktitle = {{Proc. IEEE Conf. on Decision and Control}},
  address = {Singapore},
  url = {https://arxiv.org/abs/2309.05746},
  owner = {jjalora},
  timestamp = {2023-09-11}
}