Research Opportunities

You don’t have to be a graduate student to do research in AeroAstro — we have a range of research opportunities in which you can work, sometimes as early as your first year at MIT. As a participant in the Undergraduate Research Opportunities Program (UROP), you can work with faculty as a junior colleague in every phase of their research: developing plans, writing proposals, conducting research, analyzing data, presenting research results. You may find yourself working on a project sponsored by the FAA, NASA, or a private aerospace company. UROPs are a great way to explore aerospace research, prepare for graduate school and future careers, connect with faculty and apply classroom learning to real-world research.

UROPs take place during the academic year and over the summer. Projects may last for an entire semester; some continue for a year or more. UROP students receive academic credit, pay, or are volunteers, depending on the project. A UROP may not be pursued for both pay and credit within a term. Ineligible for UROPs are special students, students on leave from MIT, and students who have completed their degree program (unless they stay registered to finish a double major).

If you’re interested in particular research areas, you should:

If you have general UROP questions or concerns, contact UROP coordinator Marie Stuppard.

Current AeroAstro Research Opportunities

View current AeroAstro undergraduate research opportunities below. Submit an opportunity (Note: Touchstone login required.)

Open opportunities are listed in chronological order, with the most recent submissions at the top.

Affiliated Lab/Research Group: Astrodynamics, Space Robotics, and Controls Lab (ARCLab)

Powered descent guidance algorithms play a critical role in the safe and efficient landing of spacecraft on a planetary or lunar surface. These algorithms are responsible for determining the optimal trajectory and thrust of the spacecraft during its descent. However, as missions become more complex and ambitious, the computational requirements of these algorithms are becoming increasingly demanding.

This project focuses on addressing the challenge of growing computational complexity in spacecraft guidance algorithms by developing more computationally efficient powered descent guidance algorithms. The project will involve solving a constrained optimal control problem to compute fuel-optimal trajectories for spacecraft entry, descent, and landing. The UROP will work on developing a multi-core implementation for a guidance-specific second-order conic programming (SOCP) solver and using supervised learning for optimal strategy prediction based on problem parameters. This work will involve utilizing the latest advances in parallel computing and machine learning to create more efficient and effective guidance algorithms.

Preferred, but not necessary qualifications:
– Familiarity with deep learning libraries, e.g., pytorch, tensorflow
– Experience with Julia, Python, or C
– Familiarity with optimal control theory and spacecraft dynamics
– Experience with parallel algorithms, multicores, and GPUs
– Experience with software development and version control (e.g. Git)

  • Project Duration: Spring 2023 (02/06/2023 – 06/04/2023)
  • Students who may apply: First Years, Sophomores, Juniors, Seniors
  • Compensation: For Credit or For Pay ($15/hr). 8-10 hours per week.

For more information or to apply, contact: Julia Briden, jbriden@mit.edu

Affiliated Lab/Research Group: Human Systems Laboratory (HSL)

Would a global carbon tax reduce the flood risk at MIT? The answer to this question of policy impact is critical for local policy making or climate-resilient infrastructure development, but is often hidden behind tables and scientific reports. This UROP will be part of creating The Climate Pocket: a climate education simulator that illustrates local science-based flood impacts of global climate policy decisions, as shown in Fig. 1 and http://trillium.tech/eie. The UROP will work in collaboration with experts in physics-informed machine learning from MIT and CERENA-IST, climate policy scenarios from En-ROADS, and physics-based flood hazard maps from Climate Central.
The UROP will create a global visualization layer of future coastal floods, as they would be seen from space, similar to Fig. 1. The research will include 1) assembling high-resolution satellite imagery and flood hazard maps with geospatial processing tools, such as ArcGIS, 2) exploring novel methods to combine physics-based metrics and generative vision modeling, such as generative adversarial networks (GANs) or normalizing flows, and 3) creating a shareable web-demo.
Preferred, but not necessary qualifications:
– Passion for climate issues
– Familiarity with deep learning libraries, e.g., pytorch, tensorflow.
– Familiarity with geospatial data processing in, e.g., Python-rasterio, ArcGIS, GDAL, GEE.
– Experience of large-scale data proccessing in python with, e.g., xarray, pandas
– Collaborative spirit
– Very optional, but fun if there‘s experience in UI/UX, web development, climate policy

We strongly value an environment of inclusion, support, and collaboration and highly encourage students from historically excluded groups to apply. The research will be with our team at the Human Systems Laboratory, Dept. of AeroAstro, and Prof. Dava Newman and can be virtual. If you’re interested please feel free to email me with a CV and two paragraphs about your interest, experience, and long-term goals at lutjens [at ] mit [dot] edu .

  • Students who may apply: First Years, Sophomores, Juniors, Seniors
  • Compensation: For Credit or For Pay

For more information or to apply, contact: Bjorn Lutjens, lutjens@mit.edu

Affiliated Lab/Research Group: Human Systems Laboratory
PI/Faculty Sponsor/Program Director: Professor Jeffrey Hoffman
PhD Student Supervisor: Maya Nasr

MOXIE, the Mars Oxygen In-Situ Resource Utilization Experiment, is one of the payloads that is on the Mars 2020 Perseverance rover. MOXIE was developed by MIT and NASA’s Jet Propulsion Laboratory (JPL) to demonstrate, for the first time, In-Situ Resource Utilization (ISRU) on another planet by extracting O2 from CO2 in the Martian atmosphere using solid oxide electrolysis (SOE). In order to inform and control its system, MOXIE has a set of temperature, pressure, and composition sensors that measure its internal gas flows. The four composition sensors are commercial off-the-shelf (COTS) hardware and include an oxygen sensor (0 – 100%) and a carbon dioxide sensor (0 – 5%) for the output gas stream from the SOE anode (expected to be pure oxygen) and another carbon dioxide sensor (0 – 100%) and a carbon monoxide sensor (0 – 100%) for the cathode (a mixture of CO2 and CO). Except for the luminescence oxygen sensor, all of these composition sensors are Non-Dispersive Infrared Radiation (NDIR) sensors produced for Earth-ambient-conditions. A series of tests under a range of pressures and compositions have been conducted in order to properly calibrate and characterize (C&C) the sensors to understand their behavior on Mars. The UROP work will involve data analysis of laboratory-collected experimental data for the composition sensors. Matlab data analysis skills needed. 

*This opportunity is open to all, but preference for Juniors/Seniors (or students with good Matlab data analysis experience)

  • Project duration: Spring 2022 (02/01/2022 – 05/31/2022)
  • Students who may apply: First Years, Sophomores, Juniors, Seniors
  • Compensation: For Credit or For Pay

For more information or to apply, contact: Maya Nasr, mayanasr@mit.edu.

Affiliated Lab/Research Group: Human Systems Laboratory

Antarctica has already experienced a 3°C rise in air temperature and is melting at an ever-increasing pace. Yet there exists no daily high-resolution satellite imagery to monitor, understand, or publicize the melting processes in Antarctica. In our previous research, we have advanced the physical consistency of generative adversarial networks (GANs) to synthesize trustworthy satellite imagery of future coastal floods. This UROP project will extend the method to temporal data and generate a trustworthy synthetic satellite product that visualizes melting Antarctic sea ice.

Preferred, but not necessary qualifications:
– Experience working with satellite data, e.g., Sen-1/2, Planet, MODIS
– Geospatial data processing, e.g., python rasterio, Google Earth Engine, ArcGIS, etc.
– General understanding of numerical methods for solving differential equations
– Cryospheric dynamics, e.g., ability to understand the MAR model
– Understanding of generative deep learning models, e.g., VAEs, GANs, normalizing flows, diffusion models
– Programming experience with machine learning libraries, e.g., pytorch, tensorflow
– Passion for climate issues
– Collaborative spirit

We strongly value an environment of inclusion, support, and collaboration and highly encourage students from historically excluded groups to apply. The research will be with our team at the Human Systems Laboratory, Dept. of AeroAstro, and Prof. Dava Newman. If you’re interested please feel free to email me until 09/17 with a CV and two paragraphs about your interest, experience, and long-term goals at lutjens [at ] mit [dot] edu.
Happy to hear from you,
Björn Lütjens

  • Project duration: Friday, September 17, 2021 to Friday, February 4, 2022
  • Students who may apply: First Years, Sophomores, Juniors, Seniors
  • Compensation: For Credit or For Pay

For more information or to apply, contact: Björn Lütjens, lutjens@mit.edu

Affiliated Lab/Research Group: Human Systems Laboratory

Climate models can be unwieldy beasts of computing. Simulating a year of climate can take up to two weeks on a 5000 GPU node supercomputer and if being powered by fossil fuels emit two railcars worth of coal.

The computational expense of climate models makes them difficult to use for exploring policy decisions, planning climate-resilient infrastructures, or quantifying uncertainties. A lightweight global climate model that can run on a tablet could revolutionize the way we interact with climate predictions.

This research project will advance the physical consistency of novel deep learning-based time-series models, such as neural networks, RNNs, LSTMs, transformers, GANs, or PINNs. The student will research their favorite method and large simulation-based climate datasets (e.g. CMIP6) to build a climate surrogate model, i.e., a trustworthy lightweight copy that can simulate the climate in minutes rather than weeks.
Preferred, but not necessary qualifications:
– General understanding of weather or climate data, e.g., CMIP6
– Familiarity with deep learning libraries, e.g., pytorch, tensorflow.
– Experience of large-scale data processing in python with, e.g., xarray, pandas
– Passion for climate issues
– Collaborative spirit
– Very optional, but fun if there‘s an experience in UI/UX, geospatial data, reduced-order modeling of PDEs, climate policy

We strongly value an environment of inclusion, support, and collaboration and highly encourage students from historically excluded groups to apply. The research will be with our team at the Human Systems Laboratory, Dept. of AeroAstro, and Prof. Dava Newman. If you’re interested please feel free to email me until 09/17 with a CV and two paragraphs about your interest, experience, and long-term goals at lutjens [at ] mit [dot] edu.

Thank you,
Björn Lütjens

  • Project duration: Friday, September 17, 2021 to Friday, February 4, 2022
  • Students who may apply: First Year Sophomore Junior Senior
  • Compensation: For Credit For Pay

For more information or to apply, contact: Björn Lütjens, lutjens@mit.edu

Affiliated Lab/Research Group: Space Propulsion Laboratory

The project is situated in the context of electrospray thruster engineering. Electrospraying is a technique that can be used to eject mass from ionic liquids in a very simplistic, and compact way, potentially enabling CubeSats with high efficiency and specific impulse propulsion.
During its normal operation, spacecraft could face serious charging and particle backscattering problems, if the electrospray beams are not properly neutralized. The most common way of neutralization for ionic liquid electrospray thrusters is the bipolar configuration, where two ion beams of different polarities are fired at the same time next to each other.

This project pretends to analyze how these two different polarity beams interact. The idea is to build an experimental setup to analyze particle fluxes, and all the postprocessing capability and data analysis.

UROPs on this project will get to participate in the design of the experimental setup, including CAD design, component machining, testing, and data processing of the ion plumes.

UROPs will get hands-on experience with space propulsion testing facilities, such as the operation of vacuum chambers, electric circuit design, soldering.

  • Prerequisites: Experience with Matlab, SolidWorks and machining is required. Availability for a minimum of 8-10 h per week.
  • Desirable: Knowledge of electrostatics, some particle motion dynamics, and basic electric circuit design is a plus!
  • Project duration: Wednesday, September 22, 2021 to Monday, December 13, 2021
  • Students who may apply: Junior, Senior
  • Compensation: For Credit or For Pay
  • Faculty supervisor: Paulo Lozano

For more information or to apply, contact: Ximo Gallud Cidoncha, ximogc@mit.edu

Paid/Credit UROPS

Paid UROP

To obtain a paid UROP, students must submit a proposal for direct UROP or research-sponsored funding by the Institute’s UROP deadline. During the fall and spring terms, most students typically work 12 hours per week and as many as 40 hours per week during IAP and the summer. MIT has an established hourly rate for UROP pay, however, at the discretion of a UROP supervisor and the department, students whose stipend is covered through research funds may earn a higher rate. Students are expected to keep track of their worked hours and submit electronic timesheets on a weekly basis. Students paid through direct UROP funds need to print their weekly timesheets for approval by their direct supervisor and bring a copy to the Course 16 student financial administrator Carol Niemi, Room 33-208. If students decide not to pursue a UROP after it has been approved, it is advisable that they immediately notify the research supervisor, the UROP Office, and Carol Niemi.

In accordance with U.S. immigration law, all student workers (and other MIT employees) are required to provide proof of identity and authorization to work in the United States. Before they can be paid, new student workers must complete Section 1 of the I-9 form electronically in MIT’s online system. The employee/student worker has three business days from hire date to present original supporting documentation in person to the Atlas Service Center.

Credit UROP

Credit UROPs can be performed during the academic year and over the summer. Note that few students enroll in credit UROPs in the summer as tuition charges will be applied.

To do a UROP for credit, students submit a proposal <http://web.mit.edu/urop/apply/index.html> by the Institute’s UROP deadline and register in WebSIS <http://mit.edu/studentserve/websis/> for 16.UR (pass/fail) by Drop Date. An AeroAstro student doing a UROP for credit outside of the department registers for a UROP number in that department (e.g., 2.URG, 6.UR, 8.UR). The number of units a student earns is equal to the number of weekly hours devoted to the project. At the end of each term, AeroAstro UROP supervisors submit a grade and an evaluation of the student’s performance on the project. Credit hours may be adjusted at that time if the supervisor deems that the student has worked fewer or more weekly hours on the project.