Juan Alvarez-Padilla

I am a second-year PhD student at the Electrical and Computer Engineering Department at Carnegie Mellon University, where I am advised by John Dolan and Zachary Manchester and work with my talented labmates at DRIVELab and RexLab. I am interested in developing control algorithms that enable robots to adapt to their environment and navigate towards their destinations. Therefore, I employ techniques at the intersection of trajectory optimization, control systems, and machine learning.

Prior to joining CMU, I held various positions in the industry, ranging from Artificial Intelligence Engineer at Robert Bosch to Data Scientist and CTO at GestaLabs Industry 4.0 Studio. Since 2021, I have been an Associate Professor at the Cyber-human Integration Technologies Division at Universidad de Guadalajara, currently on leave of absence for my doctoral studies. In addition to my research, I also serve as a NVIDIA Deep Learning Institute Certified Instructor and am a member of the NVIDIA University Ambassador Program, where I deliver workshops for free to academics and students of Latin America in Spanish to promote the learning of Robotics and Artificial Intelligence.

Education

	Carnegie Mellon University PhD, Electrical and Computer Engineering Department Driverless Intelligent Lab and Robotics Exploration Lab 2022 ─ Present
	University of Durham MSc (Research), Engineering Department Durham Energy Institute 2016 ─ 2018
	Universidad de Guadalajara BSc, Electronics Department Laboratorio de Robots Moviles 2011 ─ 2015

Research

Real-Time Whole-Body Control of Legged Robots with Model-Predictive Path Integral Control
Juan Alvarez-Padilla, John Z. Zhang, Sofia Kwok, John M. Dolan, and Zachary Manchester
In Submission, arXiv 2024

Project Page Video arXiv

Presenting a system for enabling real- time synthesis of whole-body locomotion and manipulation policies for real-world legged robots by leveraging the efficient parallelization capabilities of the MuJoCo simulator to achieve fast sampling over the robot state and action trajectories.

ROADWork Dataset: Learning to Recognize, Observe, Analyze and Drive Through Work Zones
Anurag Ghosh, Robert Tamburo, Shen Zheng, Juan Alvarez-Padilla, Hailiang Zhu, Michael Cardei, Nicholas Dunn, Christoph Mertz, and Srinivasa G. Narasimhan
In Submission, arXiv 2024

Project Page Dataset arXiv

Proposing the ROADWork dataset to learn how to recognize, observe and analyze and drive through work zones.

Course projects

On-Orbit Optimal Kinodynamic Planning for Low-Thrust Trjactory Maneuvers
Ibrahima Sory Sow, Juan Alvarez-Padilla, Fausto Vega, and Nayana Survana
16-782: Planning and Decision-making for Robotics, Fall 2023, Professor Maxim Likhachev

Report Code

Using RRT* with an LQR-based heuristic to efficiently design low-thrust spacecraft trajectories, addressing nonlinear dynamics and long-duration maneuvers.

Whole-body Trajectory Optimization and Tracking for Agile Maneuvers for a Single-Spherical-Wheeled Balancing Mobile Manipulator
Juan Alvarez-Padilla, Christian Berger, Sayan Mondal, Haoru Xue, and Zhikai Zhang
16-745: Optimal Control and Reinforcement Learning, Spring 2023, Professor Zachary Manchester

Report Code

Optimizing and tracking whole-body trajectories for a ballbot equipped with arms. By using direct collocation and Time-Variant Linear Quadratic Regulators (TVLQR), the ballbot performs dynamic tasks such as navigating complex paths and pushing off walls, maintaining balance despite changes in its center of mass.

Feel free to steal this website's source code. Do not scrape the HTML from this page itself, as it includes analytics tags that you do not want on your own website — use the github code instead. Also, consider using Leonid Keselman's Jekyll fork of this page.