Earlier this year I (successfully!) defended my dissertation titled “Design of Missed Thrust Resilient Trajectories Using Expected Thrust Fraction” and have now obtained my Doctorate of Philosophy. As, could have been expected, my dissertation was on Expected Thrust Fraction (Theory of Expected thrust fraction, an application to Mars Sample Return and Solar Sailing)
Dissertation Abstract
“Low-thrust propulsion methods, and the order of magnitude reduction in required fuel, is an enabling technology for deep-space exploration. Unfortunately, spacecraft with low-thrust levels require long-duration thrusting arcs, making them susceptible to missed thrust events. This investigation describes a new method for designing low-thrust spacecraft trajectories resilient to missed thrust events. The method, called expected thrust fraction, embeds the stochastic nature of missed thrust events into a time-dependent thrust penalty, which can be utilized inside a traditional optimal control solver. Additionally, the theory of expected thrust fraction is expanded to include solar sailing missions and their unique missed thrust behavior. The performance of expected thrust fraction is then characterized on a variety of low-thrust transfers, and its resilience to missed thrust events is compared to previous mitigation techniques. Finally, the technique is applied to NASA’s Mars Sample Return mission concept, where the spacecraft’s resilience to missed thrust events is improved from 65.4% to greater than 99%.”
Dissertation Statistics
Length – 135 pages
Duration to write – 537 days (I first downloaded the dissertation template on October 23rd, 2020 and submitted the final version for approval on April 12th, 2022. This was not all spent writing, in fact, the majority of the writing came in two bursts, an approximately 2 month-long period prior to proposing where I wrote approximately 60 pages, and a 3 month-long period in the summer of 2022 where I wrote the remaining pages.)
Number of versions – 12
Self-citations – 13 (Was that really it?)
Number of Citations to David Bowie’s hit song “Life on Mars?” I had to remove – 1
Longest citation – James Keane, Alexandra A. Ahern, Fran Bagenal, Amy C. Barr Mlinar, Ko Basu, Patricio Becerra, Tanguy Bertrand, Ross A. Beyer, Carver J. Bierson, Michael T. Bland, Doris Breuer, Ashley G. Davies, Katherine de Kleer, Imke de Pater, Daniella N. DellaGiustina, Tilmann Denk, Ariana Echevarria, Catherine M. Elder, Lori M. Feaga, Cesare Grava, Patricia M. Gregg, Tracy K. P. Gregg, Christopher W. Hamilton, Camilla D. K. Harris, Walter M. Harris, Hamish C. F. C. Hay, Amanda R. Hendrix, Sarah M. H¨orst, Rowan Huang, Andr´ea C. G. Hughes, Kandis Lea Jessup, Xianzhe Jia, Lauren M. Jozwiak, James T. Keane, Laura Kerber, Laszlo P. Kestay, Krishan K. Khurana, Walter Kiefer, Michelle R. Kirchoff, Edwin S. Kite, Lea Klaiber, Rachel L. Klima, Corbin L. Kling, Valery J. Lainey, Rosaly M. C. Lopes, Alice Lucchetti, Kathleen E. Mandt, Isamu Matsuyama, Christine McCarthy, Alfred S. McEwen, Melissa A. McGrath, Laurent G. J. Mont´esi, Julieanne I. Moses, Arielle Moullet, Quentin N´enon, Gregory A. Neumann, Marc F. Neveu, Francis Nimmo, John W. Noonan, Maurizio Pajola, Mark P. Panning, Ryan S. Park, Anne Pommier, Lynnae C. Quick, Jani Radebaugh, Julie A. Rathbun, Kurt D. Retherford, James H. Roberts, Elias Roussos, Paul M. Schenk, Nick M. Schneider, Joe W. Schools, Rohan Sood, John R. Spencer, Dan C. Spencer, Gregor Steinbr¨ugge, Ali H. Sulaiman, Sarah S. Sutton, Antony Trinh, Constantine C. C. Tsang, Janet Vertesi, Audrey Vorburger, Joseph H. Westlake, and David A. Williams. Recommendations for Addressing Priority Io Science in the Next Decade. Bulletin of the AAS, 53(4), 3 2021. (The TLDR of this paper is that IO is cool and we should go explore it!)
Grad School Stats
Duration – 4 Years (Starting Fall of 2018 and ending Spring of 2022)
Number of Internships – 2 (Both with JPL on topics very well aligned with my research)
Number of funding sources received – 6 (2 Grants, 2 Fellowships, 1 semester of TA’ing, and 1 mission support contract)
Number of grants/funding requests submitted – 25 (Note that this number is much larger than the number of sources received, there’s a lot of rejection in applying for grants)
Number of conference presentations – 7
Number of conferences attended in person – 1 (Thanks COVID)
Number of Journal papers submitted – 3
Number of undergrads mentored – 3 (Highly recommend this as it is very rewarding)
Number of undergrads I tricked into thinking research is cool and are now pursuing a PhD – 1 (Sorry!)
Kilo’s of coffee beans consumed – I honestly don’t even want to know the true number but my estimates place it at approximately 75 kilograms of dry beans, which means I literally consumed more than my weight in dry coffee beans over the course of my PhD.
Unsolicited Advice
My two pieces of unsolicited advice for anyone about to start or currently in a PhD program is to (1) write daily and (2) work consistently/intentionally
Write Daily – Every day I worked I tried to set aside 1 hour at a set time just for writing (or any of the tasks involved in writing like editing). It’s the method suggested by Paul J. Silvia in “How to Write a Lot“, and as the name implies, it will help you write a lot. Now there are two common criticisms of aiming to write a lot (a) “so I should just turn out trash” and (b) “but that will have me outstrip how quickly I can generate results”. To (a) I say, obviously not, there should be a minimum quality to anything that you publish, but that minimum should be far below “perfect”. Additionally, in my opinion, the ability for you to tell how good an idea you came up with is very rare, and by writing a lot (of at least a minimum quality) you’ll end up producing more great papers than if you simply tried to write only great papers. You’ll also produce more meh papers, but I’ve found that you’ll be judged by your best work, rather than your average work. To (b) I say, if your writing output exceeds your results and grant needs, then start a blog and place your excess there. This blog benefitted greatly from my excess output for the first year and half of grad school. I found that once you hit on a productive research approach, your production of results can far outstrip your ability to write them up (which was a misfortune for this blog).
Work Consistently/Intentionally – Some grad students love to tell you how much they work. They claim they’re at the lab all day to the point where they don’t have a life outside of it. I’ve found that the sheer volume that a person claims to work is, in general, inversely proportional to the amount they actually produce (Note: some people actually do work a lot, and produce a lot, but they don’t brag about hours worked because they are too busy working to have time to brag). This is because when they’re in the office they generally are doing light to no work (they may have a paper pulled up on one monitor, but they are actually reading the New York Times or watching YouTube). Over my four years, I rarely was in the lab more than 6 hours a day, but for the first 2-3 hours of every day, I did focused work. In the beginning, I would even turn my phone off and blocked websites like Twitter and YouTube so as not to get distracted. After these first few hours, I would start getting less productive and would hit little to no productivity around or after lunch which is when I would leave. This left me with about half the day to do other stuff and generally enjoy life. I would spend far fewer hours “in the lab/working” than some of my friends, but because I put in 2-3 hours every day of productive work, I ended up outproducing most people I know. How you spend your time in the lab is a lot more important than the sheer quantity of time in the lab.
Want More Gereshes?
If you want to receive new Gereshes blog post directly to your email when they come out, you can sign up for that here!
Don’t want another email? That’s ok, Gereshes also has a twitter account and subreddit!