Jon Arrizabalaga

Last updated: 03-December-2024

Hi! I am a PhD student at the Munich Institute of Robotics and Machine Intelligence (MIRMI) at the Technical University of Munich (TUM) under the supervision of Prof. Markus Ryll. Since January 2024, I am located at the Robotics Insitute (RI) at Carnegie Mellon University (CMU) advised by Prof. Zachary Manchester.

Before pursuing my doctoral studies, I obtained my MSc. degree in Robotics-Mechatronics at KTH Royal Institute of Technology (Sweden, 2020) and wrote the MSc. Thesis at the robotics department of Bosch Research (Germany, 2020).

My research interests lie at the intersection of system dynamics, control theory, numerical optimization and machine learning, with a special focus on their applications to robotic and autonomous systems. More specifically, aiming to enhance the agility, efficiency and robustness of such systems, I develop novel methods that unify perception, planning and control in a transparent and understandable manner. For further details, check out my publications!

You can find more about me in an interview with undergrad students! If you find my work appealing or would like to connect, please get in touch. I am always looking for new colaborations.

 

I expect to complete my PhD early 2025, and
I am actively seeking (academic/postdoc/industry) job opportunities !

PUBLICATIONS

A Universal Formulation for Path-Parametric Planning and Control

J. Arrizabalaga, M. Ryll
Preprint, 2024

 

Differentiable Collision-Free Parametric Corridors
Best Paper Finalist

J. Arrizabalaga, Z. Manchester, M. Ryll
International Conf. on Intelligent Robots and Systems (IROS), 2024

     

PHODCOS: Pythagorean Hodograph-based Differentiable Coordinate System

J. Arrizabalaga, F. Vega, Z. Sir, Z. Manchester, M. Ryll
Under review, 2024

 

Learning for CasADi: Data-driven Models in Numerical Optimization

T. Salzmann, J. Arrizabalaga, J. Andersson, M.Pavone , M. Ryll
Learning for Dynamics and Control Conference (L4DC), 2024

   

Geometric Slosh-Free Tracking for Robotic Manipulators

J. Arrizabalaga, L. Pries, R. Laha, R. Li, S. Haddadin , M. Ryll
International Conference on Robotics and Automation (ICRA), 2024

     

Pose-Following with Dual Quaternions

J. Arrizabalaga, M. Ryll
Conference on Decision and Control (CDC), 2023

   

SCTOMP: Spatially Constrained Time-Optimal Motion Planning

J. Arrizabalaga, M. Ryll
International Conf. on Intelligent Robots and Systems (IROS), 2023

   

Spatial motion planning with Pythagorean Hodograph curves

J. Arrizabalaga, M. Ryll
Conference on Decision and Control (CDC), 2022

   

Towards Time-Optimal Tunnel-Following for Quadrotors

J. Arrizabalaga, M. Ryll
International Conference on Robotics and Automation (ICRA), 2022

 

Real-time Neural MPC:
Deep Learning Model Predictive Control for Quadrotors and Agile Robotic Platforms

T. Salzmann, E. Kaufmann, J. Arrizabalaga, M. Pavone, D. Scaramuzza, M. Ryll
Robotics and Automation Letters (RA-L), 2022

 
Norway

A caster-wheel-aware MPC-based motion planner for mobile robotics

J. Arrizabalaga, N. van Duijkeren, M. Ryll, R. Lange
International Conference on Advanced Robotics (ICAR), 2021

   

SOFTWARE

I believe that open-source code is a major driving force in accelerating research and technological progress overall. To account for this, I always intend to release the code resulting from my work. See here a list of repositories that I have created or contributed to:

Norway

l4casadi
Data-Driven Models in Numerical Optimization

A Python package to embed PyTorch models and functions into CasADi. Together with Tim Salzmann, I wrote the fish and NeRF examples.

   
Norway

pydecomp
Python-based Convex Decomposition

A Python package for fast convex decomposition of obstacle-free spaces.

Norway

corrgen
Differentiable Collision-Free Parametric Corridor Generation

A differentiable collision-free corridor generator.

   
Norway

gsft
Geometric Slosh-Free Tracking for Robotic Manipulators

A real-time slosh-free tracking method for robotic manipulators.

   
Norway

pfdq
Path-Following with Dual Quaternions

A pose-following framework based on unit dual quaternions for simultaneously controlling the translation and rotation of a rigid body.

   

I hope that this code paves the ground for other researchers. If you benefit from it, please leave a !

TEACHING

During my doctoral studies I have had the privilege to lecture the following courses:

  • Engineering Mechanics 1: Introductory theoretical mechanics course offered throughout multiple semesters.
    • Winter 21-22, TUM, BSc. Aerospace Engineering, Enrolled students: 200
    • Winter 22-23, TUM, BSc. Aerospace Engineering, Enrolled students: 350
    • Summer 22-23, TUM-Asia, MSc. Aerospace Engineering, Enrolled students: 20
    • Winter 23-24, TUM, BSc. Aerospace Engineering, Enrolled students: 450
    • Summer 23-24, TUM-Asia, MSc. Aerospace Engineering, Enrolled students: 25
  • Introduction to Robot Operating System (ROS): Offered as an elective accross multiple MSc. programmes at TUM.
    • MSc. Computer Science and Robotics
    • MSc. Aerospace Engineering
    • MSc. Mechanical Engineering
I very much enjoy teaching and have always received great feedback from the students. In fact, for my teaching services in Engineering Mechanics 1, I was awarded with the Best PhD Lecturer Award !

CONTACT ME

If you are interested in reaching out, please send me an email. In case you want to receive the latest updates on my research, follow me on social media!

jarrizab@andrew.cmu.edu

Jon Arrizabalaga

@jonarriza96

jonarriza96