Summer ‘break’

Yay! It’s summer! You’re in school! That means you get a break and get to do whatever you want!

As Ph.D. Comics (a scarily accurate academic comic strip) eloquently puts it:

See, summer in academia is the time of no classes, no project deadlines, no undergrads around to bug you for help or pollute the line at the coffee shop, and nothing but the call of the warm summer sun to distract you. That means it’s the perfect time to get work done, and sometimes taking advantage of the summer sun by sitting on a rooftop bar with a cocktail and a big stack of research papers to read (how most of my reading has been done this summer). This also means that if you don’t get work done, there are no excuses to feed your advisor about why you’re behind…

What am I up to this summer? So far, what I’m supposed to be doing has been rather different from what I’m actually doing.

I started the summer by going to Germany for the Robotics: Science and Systems conference. This conference is being held at Oregon State next year and hosted by our program, and as a former quasi-event planner, I have been tapped to help out. As grad students, we are required to dedicate a few hours every month to ‘service’, which is basically ‘being a good person and doing something to help the program, local community, or robotics community’. This definition is very broad, so for the next year, my service requirement will be fulfilled by helping to plan this conference. Just when I thought I had escaped…. But hey, it meant a free trip to Germany so I can’t complain too much.

Once I returned from Germany, it was hardcore nose-to-the-grindstone on research time… for about 2 weeks. Thankfully, I did get quite a lot accomplished in that time.

Next, it was off to the Bay Area, where I’m writing this from. I’m here to visit family and friends, and attend all the wedding festivities (bridal shower, bachelorette party, etc) for my sister’s upcoming wedding (in less than a week!!). I’m also spending a lot of time assisting with wedding plans and details.

All of this culminates in not nearly enough research getting done.

Which leads to the question – I keep talking about this ‘research’ thing – what does that actually mean? What am I supposed to be doing this summer?

This summer I have three main goals:

Goal 0: Do enough that my advisor is happy and therefore doesn’t kick me out. Ok – this is the biggest all-the-time overarching goal in grad school, so I’m not counting it in my official three summer goals. But how I accomplish this goal is by doing the following.

Goal 1: Make marked progress on my research. My broad research goal is to enable robots to have a better understanding of the physical world. Currently, most robots use two systems to ‘see’: video cameras and LIDAR. LIDAR gives them a 2D or 3D point cloud (i.e. distances to hard surfaces around them). Combining these two systems allows robots to develop a good enough representation of the world to navigate around. However, most are still unable to distinguish the wall from a chair or a person. There has been a lot of research in recent years about providing robots with a semantic understanding of the world, i.e. the ability to recognize specific objects or attributes. My goal is to take this a step further and develop a system that allows the robot to not only recognize unique features, but create an internal 3D CAD-like representation of the world with objects. With this system, the robot would be able to understand properties of the world (e.g. that chairs move but walls don’t, which helps with navigation) and find objects (e.g. knowing that a specific wrench is typically in one of three places). This system would also enable a much more privacy-focused representation of the world; instead of storing the exact image of the papers on your desk, it would only store that there was a stack of papers and no information about what is on them.

So far this summer, I have done a lot of background research on this topic (i.e reading a whole lot of research papers) and reached out to a group working on a very similar project to see if there is opportunity for collaboration. I have also started working on designing the system from the ground up and seeing if I can leverage other code bases.

Goal 2: Prepare myself for my qualifying exam. The qualifying exam (aka ‘quals’) is an ‘exam’ that Ph.D. students take early in their studies. The purpose is to prove that you have ideas, can write cohesively, and are generally capable of doing research. This is the first step that differentiates masters from Ph.D. students. Every program runs their quals exams differently – our program runs it for six weeks in February to March time frame in the student’s second year. Each student receives three topics specific to their research at the start of the quals period. They then have five weeks to put together a comprehensive 15 page paper that outlines background research and future research directions on each topic. After that paper is submitted, they put together a 20 minute presentation and present it to faculty approximately one week later. My goal this summer and fall is to do much of the background research and compile resources and ideas, as this is the bulk of the work for quals and will help immensely during that very intense six week period. This can be accomplished in tandem with goal one, as I have already done much of the background research in order to start my project anyway. The main difference is making sure I’m organized, keeping track of papers I’m reading, taking lots and lots of notes, thinking up ideas for future research directions, and thinking about them in terms of my quals exam.

Goal 3: Apply to the NSF Graduate Research Fellowship Program. This is a highly competitive STEM fellowship that is awarded to graduate students across the nation every year. Typically there are about 16,000 applicants and 10-12% are awarded fellowships. The fellowship is basically ‘free money’ – it provides students three years of funding including tuition, a stipend, and (optionally) additional funding for other research activities. The main benefit (other than free money) is that it allows the student to be decoupled from their advisor’s funding, giving them much more freedom in their research. The fellowship focuses on two main aspects: intellectual merit and broader impacts. Basically, you need to prove that you are able to advance your specific area of research and our scientific knowledge, and are able to make a positive impact on the world around you. Applications include a research statement, personal statement, and reference letters. This will probably take me 4 weeks of full time work to put together. So far, I have started compiling information and requirements, reading (a whole ton of) advice about how to apply, reading prior successful and unsuccessful applications, and brainstorming what I want to include in my statements. Once my application is submitted, I will likely include these statements as individual blog posts.

I’m working to carve a few hours out of each day to push these goals forward bit by bit. Once I get back from the bay, it’s back to working 50-60 hours a week to make sure these goals are accomplished (and therefore ensure that my advisor is happy).

One of the biggest lessons I’ve learned in grad school so far is how important it is to take time to unwind. So this summer I’m also working hard to take time away from working hard. I am trying to fully enjoy the sun while it’s here, spend time with my family, see my friends, and celebrate my sister’s wedding. Here are a few pictures from my adventures so far this summer: