by Michael Saint-Guillain, PhD Student
It is in the exciting context of space exploration that I spent two weeks during March’18 isolated in the Utah Desert (U.S.), as executive officer and computer scientist, in a crew composed of eight students, all from UCLouvain.
I mainly wanted to provide my own opinion about the need for computer scientists in outer space manned missions, such as Mars exploration. In corollary, I also wanted to point outthe following fact: Building your own life on working on computers and algorithms may sound quite disappointing if you hope for exciting projects and extraordinary experiments. That’s wrong, and I’ve never as much enjoyed life as since I started my Phd in Computer Science.
Computer Science, and consequently computer scientists, are cool !!
Throughout the discussion I will rely on the modest experience I gathered on space missions, provided to me by the UCL to Mars project.
Before the mission
Our Martian adventure started exactly one year ago, in April’17. I had been selected by both the 2016 and 2017 UCL to Mars crews for being part of the future 2018 crew. I met the other selected crew members, my crew members, aware that they would be now part of my life during a year and half. That’s quite a strong commitment. I didn’t even knew them, neither did I knew how important they will actually become in my own little short term personal universe.
It is after a short period of euphoria that began an infinite-looking sequence of sets of subsets of painful tasks. Nobody told me that I would spend more than 90% of the time I devoted to the project, actually not being working on that project. Well, that’s not completely true. During all these weeks and months, we were indeed working on it that is, making the project feasible. Sponsors… I hated this word during eight months. Nowadays, I thank them for putting their faith
on us and making this fantastic adventure happen. Truly, their all great, so take a minute to look at them on the project’s website.
I get the sense that our Martian adventure truly started only two months before the mission, when we met our very esteemed compatriot Frank De Winne . Meeting such a Belgian hero was an amazing moment for all of us. It also definitively made things concrete and tangible. Afterwards, things went a bit too fast for me to describe it. And so we landed on Mars.
The role of computer scientists on Mars
There is not a single chance that future first manned mission on Mars occurs without being extensively simulated. That’s in this context that the Mars Society hosted us during two weeks of full isolation in their beloved Mars Desert Research Station (MDRS).
« The Mars Desert Research Station (MDRS), owned and operated by the Mars Society, is a fullscale analog facility in Utah Desert that supports Earth-based research in pursuit of the technology, operations, and science required for human space ploration. » source: mdrs.marssociety.org
The MDRS is such an amazing place. The Mars Society is full of mind-blowing people, and a Mars analog mission is really something. Amongst the most frequent questions I usually receive about this mission, the following appears quite very often:
« And, well, what does a computer scientist do on such a mission ? U no astrophysics, chemistry ?… »
In what follows, I will provide that question with a few answers. If you are interested in Mars analog missions and their major issues in general (they are many other analog research stations such as the MDRS, around the Earth), you should feed your curiosity with the excellent (and recent) Springer book « On to Mars ! Chronicles of Martian adventures », by Vladimir Pletser
So, how could a computer scientist like myself be of any help in such mission ?
Let us start with the research project I carried at MDRS. You can read the abstract here , otherwise in a few words: it is about how to schedule on-the-spot operations in the light of uncertainty (and, as a side question, how to model every crew member’s esearch project as a finite set of atomic constrained jobs). In Maximizing the value of a space mission , Nicholas G.Halla and Michael J.Magazine insist on the importance of mission planning/scheduling, as 25% of the budget of a space mission may be spent in making these decisions beforehand. As an example, the a priori scientific schedule involving the more or less 175 experiments to be conducted by Voyager 2 space probe required 30 people during six months. Nowadays of course, hardware and
techniques evolved and it is likely that a couple of super-equipped (i.e. with a brand new laptop) brains may suffice in that case. Yet, the problems evolved too. Instead of the single machine Voyager 2, space missions have to deal with teams of astronauts. For instance, that is currently the main responsibility of the EPIC team concerning the ISS Columbus module. Whereas the scheduling problem I tackled at MDRS involved 8 operators, future missions on Mars are likely to involve much more people.
Why should we start sending computer scientists in the outer space? Couldn’t they simply stay on earth and solve our problems down here, remotely ?
In the case of manned missions as far as Mars, the main problem will be the communication delays. From the Moon or the ISS, a signal takes less than a second to reach the Earth. Astronauts access high bandwidth Internet on the ISS. In the very near future, we’ll enjoy 4G on the Moon. FromMars, depending on the positions of the two planets and the Sun, the delay varies between three and twenty one minutes. If the Sun is between both, then communicating is even impossible. Sounds impossible to have a conversation in such context. The classical « Houston, we’ve had a problem » would probably take around ten minutes to be heard, as well as each question or answer in any of both directions. We’re therefore not anymore in a context where all decisions are taken in realtime down on Earth and astronauts are « only » super-skilled operators implementing these.
On Mars or further, realtime decisions such as operations management, but also exceptional measures in case of unpredictable events (injuries, critical system breakdown, …) will have to be determined locally. Standard deviations, such as operational issues requiring rescheduling, can be tackled by dedicated algorithms. In case less-standard events occur, such algorithms may have to be modified urgently. But when critical issues or anomalies will happen, there will be no dedicated software. The safety of the rew will then entirely (and how desperately) rely on some Artificial Intelligence, and consequently on the computer scientist behind, ready to interact or, whenever needed, to fix things. In fact, nobody wants the only solution option of undesirable scenarios to be left without anyone capable to repair or adapt it. There is an interesting article (Fr.) discussing the major issues of long scale space missions, from the computer science point of view.
There will always be work for computer scientists
Computer Science will also be of significant utility on planetary manned missions. You won’t have GPS on Mars. You won’t even have any detailed map on which to position yourself. In the beginning, ground solutions, such as UWB positioning through ground antenna grids, will be preferred to expensive GPS satellite networks. During the UCL to Mars 2018 analog mission, our crew engineer Bastien Baix was working on simulating such Martian conditions, by working on the very first step of a Mars
positioning system, that is the 3D mapping of the areas around the MDRS planetary habitat.
In fact, there is almost no scientific experiment that could potentially be conducted on Mars which does not involve Computer Science at some level, if not critically depends on it. There is nothing more useful than a computer scientist. Apart from basic troubleshooting activities, such as fixing the MDRS Phoenix remote-controlled rover, I also had the chance to provide a modest but valuable help to our crew physicist Sophie Wuyckens in her research. In fact, her master thesis as well as her project at
MDRS concerns the detection on the ground of « muons », which are particles produced from cosmic microwave background heavier particles disintegrated when entering the atmosphere. As Sophie was using a bench of computer programs to gather informations of the muons hitting the detector, we worked together at modifying these programs in order to fit better her experimental requirements during the mission.
Learning by sharing my own skills to someone from a totally different field is something really inspiring and fulfilling.
Take-away messages :
• Computer Science is cool.
• Come at our conference on Space exploration ( Fr) , May 22sd: