Guillem Casadesús Vila

An aerospace engineer from Barcelona passionate about robotic space exploration, autonomous and distributed space systems, optimization, and machine learning

I am currently a PhD Student at Stanford Aero&Astro, working on lunar Positioning, Navigation, and Timing at the Navigation and Autonomous Vehicles (NAV) Lab. In 2022, I completed my Bachelor's Thesis on dynamic resource management in satellite communications as a visiting student in the System Architecture Group at the Department of Aeronautics and Astronautics at the Massachusetts Institute of Technology (MIT). In parallel to my thesis research, our MIT team developed a mission concept for producing propellants on Mars from in-situ resources. I obtained a BSc in Aerospace Engineering and a BSc in Telecommunications Engineering at Universitat Politècnica de Catalunya (UPC) under the program offered by the Interdisciplinary Higher Education Center (CFIS). During my undergraduate studies, I collaborated with the UPC Nano-Satellite and Payload Laboratory on projects encompassing satellite constellations. I also completed an internship at Barcelona Supercomputing Center, working on a machine-learning distributed library for high-performance computing. During high school, I started the Gadget Robotics team, check it out here!

Soccer, skiing, scuba diving, and traveling are my favorite activities to take a break from work.

Research and Projects

Positioning, Navigation and Timing for Space Robotics

Navigation and Autonomous Vehicles – Stanford

Accurate Position, Navigation and Timing (PNT) will be crucial in the new era of Moon exploration. However, current vision-based methods are unreliable for the Moon's polar regions and permanently shadowed craters. I will be working on the architecting of lunar satellite navigation systems that take advantage of broadcasted Earth-GPS signals to provide PNT services and reliable communications.

[NAV Lab Lunar GPS]

Deep Reinforcement Learning for Trajectory Design in Cis-Lunar Space

Stanford Computer Science

There is a growing focus on populating the region between the Earth and the Moon, cis-lunar space. Finding low-thrust trajectories for cis-lunar space in real-time is a challenging problem, as traditional optimization methods require high computing power. Therefore, new approaches are needed to compute optimal trajectories and their control inputs in real time, improving spacecraft autonomy and efficiency.

[Stanford CS229]

Dynamic Resource Management for Communication Satellites

System Architecture Group – MIT

Modern high-throughput satellites with digital payloads and phased arrays have thousands of configurable variables. This makes it necessary for satellite operators to move from manual resource allocation to autonomous and dynamic resource management tools. The project aims at developing a holistic and AI-based approach to dynamically allocate resources for satellite constellations.

[Publications] [System Architecture Group] [ESL]

Martian In-Situ Resource Utilization from Subsurface Water Ice

Engineering Systems Laboratory – MIT

Fifty years after the first steps on the Moon, NASA is planning on sending crewed exploration missions to the Moon and Mars. Leveraging in-situ resources to produce propellants and consumables is a key enabler for these space exploration missions. The concept developed during NASA's RASC-AL competition consists of trucks working in tandems and focuses on reliability and scalability to produce large amounts of methalox.

[Publications] [MIT News] [MIT AA News] [NASA story]

Distributed Satellite Systems Simulator and Autonomous Operations

Nanosatellite and Payload Laboratory – UPC

Complex Earth Observation satellite systems are called to satisfy higher requirements than ever before, requiring specific simulation tools to plan their design and management. The goal of this project is to develop a Distributed Satellite Systems Simulator aimed at facilitating the conception and validation of new mission architecture concepts and advancing satellite networking and autonomous operations.

[Nanosat Lab]

Machine Learning and Distributed Computing Models

Workflows and Distributed Computing Group – BSC

In recent years, machine learning (ML) has proven to be a breakthrough in solving complex problems. However, ML libraries are difficult to set up in high-performance computing clusters. In this project, I developed ML algorithms for the Distributed Computing Library, a Python library built on top of PyCOMPSs, a model and runtime for distributed infrastructures, such as Clusters and Clouds.

[dislib] [COMPSs]

Publications

G. Casadesus Vila and G. Gao, “Markov Decision Processes for Scheduling Lunar PNT Services”, Proceedings of the Institute of Navigation GNSS+ Conference (ION GNSS+ 2024), Baltimore, MD. Abstract accepted.
K. Iiyama, G. Casadesus Vila, M. Cortinovis, K. Coimbra, and G. Gao, “System-Level Comparison of Lunar Orbit Determination and Time Synchronization Methods”, Proceedings of the Institute of Navigation GNSS+ Conference (ION GNSS+ 2024), Baltimore, MD. Abstract accepted.
K. Iiyama*, G. Casadeus Vila*, G. Gao, “Contact Plan Optimization and Distributed State Estimation for Delay Tolerant Satellite Networks (accepted),” 2024 IEEE Aerospace Conference, Big Sky, MT, USA, 2024.
C. Agia, S. Bandyopadhyay, G. Casadeus Vila, et al., “Modeling Trades for the Design of Deep Space Autonomous Spacecraft and Simulators (accepted),” 2024 IEEE Aerospace Conference, Big Sky, MT, USA, 2024.
K. Iiyama*, G. Casadeus Vila*, and Grace Gao, “LuPNT: Open-Source Simulator for Lunar Positioning, Navigation, and Timing,” Proceedings of the Institute of Navigation GNSS+ conference (ION GNSS+ 2023), Denver, CO, Sep 2023.
P. Morrissey, . . . , G. Casadeus Vila, et al., “The High-Energy X-ray Probe (HEX-P),” SPIE UV, X-Ray, and Gamma-Ray Space Instrumentation for Astronomy XXIII, San Diego, CA, USA, 2023.
G. Casadesus Vila, E. Alarcon, “Toward Autonomous Cooperation in Heterogeneous Nanosatellite Constellations using Dynamic Graph Neural Networks (accepted),” in International Astronautical Congress, Baku, Azerbaijan, 2023.
J. J. G. Luis, S. Aliaga, G. Casadesus Vila, N. Pachler, E. F. Crawley, and B. G. Cameron, “Frequency Plan Design for Multibeam Satellite Constellations Using Integer Linear Programming,” IEEE Transactions on Wireless Communications, 2022. [link]
G. Casadesus Vila, J. J. G. Luis, N. Pachler, E. F. Crawley, and B. G. Cameron, “Dynamic Frequency Assignment for Mobile Users in Multibeam Satellite Constellations,” International Astronautical Congress, Paris, France, 2022. [link]
C. Gentgen, G. Casadesus Vila, G. C. Lordos, J. Posada, M. Long, L. Nagareddy, J. Kambhampaty, M. T. Have, M. Hoying, J. A. Hoffman, and O. L. D. Weck, “BART & MARGE: A Resilient and Scalable Architecture for In-Situ Propellant Production on Mars Using Large Trucks,” AIAA ASCEND 2022, Las Vegas, NV, USA, 2022. [link]

Education

Sept. 2022 - Present

Stanford University

MSc Aeronautics & Astronautics

Coursework: Machine Learning, Multidisciplinary Design Optimization, Multi-Robot Control and Distributed Optimization, Decision Making under Uncertainty, Distributed Space Systems.

Sept. 2021 - July 2022

Massachusetts Institute of Technology

BSc Thesis in AeroAstro

Title: Dynamic Frequency Assignment for Mobile Users in Multibeam Satellite Constellations

Advisor: Prof. Edward F. Crawley

Sept. 2017 - July 2022

Universitat Politecnica de Catalunya – CFIS

BSc Aerospace Engineering

BSc Telecommunications Engineering

Coursework: Systems Engineering, Aerospace Propulsion, Control Theory, Flight Mechanics, Communication Systems, Information Theory, Signal Processing, Network Protocols

Speaking & Media Appearances

Stanford University [Spotlight]

Robotics projects, lunar PNT at NavLab, and martian ISRU

#SpaceExploration

Stanford, CAMarch 2023

Participant at [ASCEND]

Publication on Martian ISRU and propellant production concept

#SpaceExploration

Las Vegas, NVOctober 2022

Presenter at the [Int. Astronautical Cong.]

Publication on resource optimization for satcomms

#Satcomms

Paris, FranceSeptember 2022

Featured in [elSegre], [TotLleida], [laMañana]

"laCaixa" Fellowship award reception from the king Felipe VI

#Fellowships

Madrid, SpainJuly 2022

Interview in [la Veu de la Segarra]

Personal projects, "laCaixa" Fellowship award, and grad school

#Fellowships

Madrid, SpainJuly 2022

Featured in [NASA News], [MIT News] and [MIT AeroAstro News]

First place at the NASA RASCAL competition

#SpaceExploration

July 2022Cocoa Beach, FL

Presenter at the NASA RASC-AL Final Forum

Publication on Martian ISRU architecture

#SpaceExploration

Cocoa Beach, FLJune 2022

Presenter at [Malakabot Robotics]

Design and construction of a humanoid robot

#Robotics

Malaga, SpainJune 2017

Featured in [elSegre, video] & [Comarques]

Design and construction of a humanoid robot

#Robotics

Lleida, SpainApril 2017

Creator of [Gadget Robotics Team]

Team that participated in the Spanish National Robotics Competition

#Robotics

Lleida, SpainMarch 2017

Featured in [UPF News]

First place at the XI UPF Research and Engineering Awards

#Robotics

Lleida, SpainMarch 2017

Featured in [Ara] & [laPedrera Foundation]

First place at the Exporecerca Jove research fair

#Robotics

Lleida, SpainMarch 2017

Project Images and Videos

Video_optimized_global_AdobeExpress.mp4

MEO communication satellite constellation providing service to fixed and mobile users such as airplanes and cruise ships. The satellites are represented by triangles on the equator, the beam footprints by ellipses, and the optimized frequency assignments by colored boxes. Generating a baseline frequency plan and re-optimizing in real time enables autonomous operations under uncertain user demand.

Autnonomous races in one of National Robotics League events in the Vasc Country. Our robot featured a PD controller to follow the black line enhanced with speed and turbine power schedules learned during track reecognition before the race. The robots have sensors that allow them to asses the possibility of performing overtakes.

Component breakdown of Dr. Gang, our racing robot. It features a small turbine in its center to lower the pressure underneath the robot and achieve high speeds during turns, as well as forward and side sensors to autonomously perform overtakes.

Autonomous race between two robots featuring a risky but successful overtake during one of the matches of the Spanish National Robotics League in October 2020.

Friendly fight between our humanoid robot Timo (in blue) and Raider (in red) during the Madrid Gaming Experience in November 2016. No one got hurt.

Getting ready for an autonomous mini-sumo battle between two robots, where they will try to push one another off the tatami. Our opponent's robot features a piece of paper to create confusion, whereas ours has a pattern hard to detect by IR sensors.

More about me

Fixing a village water supply in The Gambia

Admiring SLS after its wet rehearsal

Visiting Percy at the Intrepid Museum in NY

Making friends in the Red Sea

Playing for the MIT club soccer team

Skiing with my brother in the Pyrenees

Contact