Author: Steph Long

Hey blog readers!

This week was one of the best yet – testing. With many weeks of work done on the vehicle by the highly skilled Sentry team, Sentry is almost ready to exit its overhaul period and get back to sea. The last step in the overhaul, as is traditionally the last step in any engineering implementation, is testing. Sentry can be tested in a number of ways, but for this testing period it was privy to all of them. 

First was deck testing, which is exactly what it sounds like. Out of water, the general functionality of Sentry is tested. This includes engaging thrusters, moving servos, checking battery diagnostics, flashing strobes, beeping acoustic transmitters, and more. This is the first all-vehicle check to ensure that things are working after the extensive take-apart of the vehicle. During this iteration of testing, the team discovered some issues with the response time of the multibeam’s transmitting transducer, or its TX unit. We troubleshooted by changing the TX unit (a heavy–40lb–ceramic red rectangle that houses the necessary components and is bolted to the bottom of the vehicle) out for a spare. To do this, we didn’t actually change anything about the vehicle–a technician simply propped up the spare unit on a wooden block, so that the cable which connected the TX to the rest of the vehicle could reach the spare instead. This is an important bit of experienced/insider knowledge, because often the problem isn’t where you “think” it is, and if you were to actually switch out the whole TX unit and then discover it wasn’t the problem, you would have wasted at least an hour and a lot of nerves! It was discovered that the spare didn’t work either–so likely the issue wasn’t the spare, but really another component. Because the multibeam is a non-critical sensor, although a very important one, the team decided to defer the issue until well testing when more data could be gathered which would help troubleshoot the cause of the issue.

After deck testing–numerous times–comes well testing. Well testing is a lot more explanatory than I initially thought–it is essentially, sticking the robot in a well, and seeing how it does in water. This serves the same purpose as deck testing, as well as checking that all parts have been sealed properly (to prevent leaking of oil, which is annoying at surface but fatal under pressure) and the ability to actually gather data from the sensors and check that they are truly operating properly. Because most of the sensors rely upon acoustic data, and some even rely upon flushing ocean water through themselves, it is important that they are submerged to really understand how they are functioning.

That’s all for this week–hope you enjoyed! I sure did :).

Write you later,

Steph

Hey blog readers!

I am so excited to share my adventures from this week with you! Let’s start with a routine update of the work I’ve been able to accomplish this week, and then I’ll tell you about my super-exciting mini oceanography cruise.

After my presentation last week, I’ve pivoted gears from simply researching data science solutions to really digging into generating a proof of concept. I say all of this, but of course, I spent the majority of this week setting up, installing, and configuring the necessary software for that proof of concept. It turns out that deep-sea robots are incredibly complex—and that’s after all the work has already been done to make them! Just “plugging in” to the system, so to speak, is an endeavor all by itself.

So, this week I spent a lot of time jumping down various rabbit holes in order to correctly set up ROS (Robotic Operating System) on a virtual machine, and I finally have it up and running. I was even able to run this really cool simulation of Sentry.

Separately, this week I had the opportunity to go on a short, few-hours-long intro-to-field-oceanography cruise! We boarded the ship—a small but decently sized fishing vessel owned by a community member who participates in a large amount of citizen science here at WHOI—and cruised out of the harbor.

We deployed a camera tow and were able to visualize and inspect eel grass forests. It was a bit choppy in the channel, so we weren’t able to see as much as we had hoped with the camera, unfortunately. But the best part was yet to come—we deployed a number of catchment vessels to pull up some critters to research! We dredged a small section of the bottom, used a floating plankton catcher, and deployed a lobster pot.

Here are some more pictures from that short but sweet adventure:

Hope you enjoyed the pictures—I hope they made up for last week’s lack!

Write to you later,
Steph

Hey blog readers,

This week was jam-packed with a variety of activities and opportunities to learn!

On Wednesday, the AOPE (Applied Ocean Physics and Engineering) staff held scientific presentations, where scientists and engineers gave lightning-round talks about their current research and past findings. It was really cool to hear about the huuuuuge variety of things people are researching—all within the same department.

For example, many scientists are investigating the role and dynamics of sea ice and glaciers. Others are studying the health of kelp forests in South Africa and creating an accessible online map of the world’s giant (and other) kelp forests. Through the website, you can go back—say, 50 years—and explore how the forests have grown or disappeared over time.

In contrast, some researchers are using machine learning and artificial intelligence to “train” robots to swim gracefully—or, more scientifically, to move hydrodynamically through the ocean. Instead of brute-forcing their way through the water, these robots use physics and optimized movement to their advantage.

These are just two extremes of the research happening within the AOPE department. When I compare that to the work I’m doing in the DSL, and the work of those around me, it really highlights the incredible breadth of AOPE’s mission. You can also see how all of these pieces come together. For example, I can envision the data collected by Sentry (the deep-sea robot my team works on) being used to analyze certain oceanic plumes. Those findings, in turn, could help explain why certain oceanic vegetation—like giant kelp—is thriving or not.

There’s a chain of research and a cycle of knowledge that flows not just from Sentry and our robotics teams to external scientists, but internally as well. That’s part of what makes WHOI function so exceptionally.

Back to my work—this week, I had my mid-internship design review and presented my findings on data management for the Sentry machine learning pipeline to members of the Sentry team. The presentation went great, and the feedback I received will serve as a strong foundation for bringing Sentry into the future.

I deeply apologize for the lack of pictures this week—although, that probably gives you an accurate sense of how focused I’ve been on preparing for my presentation!

That’s all for this week—write you later!
Steph

Hey blog readers!

I hope you had a great 4th of July last weekend–I certainly did, and I’ve got some updates to share!

Starting off strong, I’ve continued to dig into the world of big data science, and I’ve been exploring some tools that other institutions and entities have made for robotics and similar projects. The good news is their solutions are offering guidance on what ours should look like, but the neutral news is that their solutions aren’t going to work for our solution!

I’ve been putting together a presentation to share my research and findings with the rest of the Sentry software team, and I’ll be presenting that this Thursday to receive feedback and input on my proposed architecture design. To give you an idea of the type of thing I am talking about, here is a quick mockup of a possible solution’s architecture.

The other really exciting thing that happened this week isn’t exactly to do with Sentry, but it was inspired by it. Being around big, awesome, deep sea vehicles really gets you inspired. So inspired, in fact, that you may even become the owner of a vehicle yourself–for your own tinkering and scientific experiments of course.

This past weekend I picked up a deep cherry red 1975 Honda CB200 motorcycle.

I drove a moving truck 100 miles, on a 6 hour drive that should have taken me 2 hours, through peak 4th of July traffic to retrieve such a treasure and I’ll say it was completely worth it. Here’s a picture of the motorcycle when I picked it up.

And here are some comparison pictures on my first maintenance project; de-rusting and revitalizing the front suspension forks.

This small beast definitely has a lot of work left, but no worries because I don’t even know how to drive a motorcycle yet. All in good time!

Chat you later,

Steph

Hey blog readers,

This week I got to get my hands a bit more dirty and really dive into the mechanics of data science and the data world of robotics.

I’ve been assigned a project which will help Sentry and other AUVs utilize machine learning, deep learning, and other advanced computer thinking techniques. 

The problem statement and background is this:

Sentry cruises are structured in that there are a certain number of dives each cruise, anywhere between 5-15 is typical. Between each dive, when the vehicle is pulled shipside there are a number of checks and maintenance tasks performed on the vehicle. During these checks, experts will look over the data that Sentry retains from the dive. There is so much data, but the type that these experts are after is the engineering data and the sensor data. Engineering data reports upon things like thruster performance and battery health during the dive, whereas sensor data is the juicy stuff: it’s what the scientists are after, and what the whole point of the dive is. The thing is, the sensor data is helpful for us engineers as well. Experts can comb over the sensor data and notice if sensors are calibrated incorrectly, or if something is going wrong on a hardware or software level. The engineers then adjust the hardware and software accordingly, and this process happens each dive.

The issue is, some errors are sneaky, and not detectable even to the trained eye until much later in the post processing process, after the cruise has ended. This ends up causing big problems, as it can even render scientific data unusable or “dirty”. So the long term goal of the project I am working on is to automate this process, and be able to catch any errors or data discrepancies in the field, and be able to adjust accordingly so valuable data isn’t lost.

The specific project I am working on to meet that goal is about assembling a machine learning pipeline to automate that process and to even learn new things about the data in the process. The part of the pipeline “under construction” right now is the data staging area. We have tens of terabytes of archived dive data, from just Sentry dives alone. The question is, how do we successfully manage and organize this data, and using what platforms and tools, so that it is optimized to make Sentry a smarter robot? So that the solution is optimized to be used for training and deploying models, that can greatly improve the performance of not just Sentry, but other vehicles in due time.

Anyways, that’s a bit of what I’m working on! Right now it is mostly in the design and planning stages, but like I said, I’m getting my hands “dirty” with a bit of code!

Catch you later,

Steph

Hey blog readers,

I’ve had a first week jam packed with learning and exploring, both intellectually and geographically! To start with the intellectual; I have been gaining familiarity with the procedures and structures of which the team at WHOI operates Sentry, as well as digging into databases and machine learning pipelines. A lot of hours of research here, so not much else to share but I have learned a lot!

Geographically, I’ve explored a ton of beaches – my favorite being Nobska Beach which features a light house right on the eastward bluff. I also got to tour a few of the robotics facilities and meet them face to face (the deep sea robots, that is). And, last but certainly not least, I’ve been participating in the summer softball league and lets just say that the Applied Ocean Physics and Engineering team is eaaaasily the best.

Until next week,

Steph

Hey blog readers! My name is Steph Long and I am extremely excited to be writing this entry. 

I am a rising junior double majoring in Electrical Engineering and Computer Science at Brown University, in Providence, RI.

If I had to describe the core of who I am in a byline, I would say that I am deeply fascinated by the unknown – whether that be the depths of the ocean, the vast horizons of space, the view from the next mountain ridge on a hike, or the number of peanut butter cups in my favorite pint of Ben & Jerrys ice cream. I would also say that I am most driven not only by what I do not know, but more specifically by what no human knows. So hopefully it comes as no surprise that I am really really interested in the utilization of robotics and engineering to explore inhospitable environments, such as the deep sea. 

In my free time, I like to take my aptitude for engineering things and apply it to the highest stake activity possible–crafting. I recently have gotten really into sewing my own jeans, carving linocut stamps, and any other type of creative outlet that satisfies my need to build (on a lower frequency than robotics). On my drive out to Woods Hole I stopped at an art thrift store, so with any luck hopefully I will soon also be able to say I am into jewelry making! I also love the outdoors and this past year have gotten really into backpacking, so I look forward to filling my weekends with riding my bike around the Cape and camping on any shore that will have me.

Tomorrow I begin my internship in the National Deep Submergence Facility at Woods Hole Oceanographic Institute, specifically working on the Sentry AUV– an autonomous robot capable of diving 6,000 meters under the ocean’s surface. So I invite you to come along with me, and submerge yourself into the world that is deep sea robotics. I’m super excited to learn as much as I can, and to share my experience with all of you!  

Write you later,

Steph