ROV #1: MVR

The messings around with the 3D printer have proved fruitful! In this post I will share my experience from never having opened a CAD program to printing cool parts that fit my ROV beautifully. I ordered the A1 Mini printer from Bambu Lab upon a recommendation from my friend Jonathan who is a wizened 3D printer. It’s a great entry-level printer that is nonetheless very capable, especially for my needs which are fairly limited. Its bed is about a foot square, which is larger than any part I will design, for this iteration at least. ...

Turns out none of my predictions of what I would work on next were correct! I did spend a little time getting a simulation set up in Gazebo using an existing URDF file for the BlueROV2, but that was a one-off endeavor on a long plane ride. When I thought about the next logical thing, it ended up being the wired comms, since as we know, radio comms underwater are impractical at best. ...

I took a break from the project to get married and go honeymooning, but once I was back, I had a recharged drive to work on this project and a lot more bandwidth with wedding/honeymoon planning fully wrapped up. The way I left it back in early June was with a thruster that sort of worked but not really. It wasn’t responding as expected to the basic controls I was giving it. ...

One of the principles I’m adhering to for this project is to build momentum by getting stuff working quickly. I don’t want it to languish in design purgatory. Since I bought a thruster early, I set a goal to get it running as soon as possible. This meant I would need a power source, a control input, and a way to wire those together with the thruster. The basic structure is: ...

Before I started purchasing parts, I wanted to have a rough design in mind to ensure I didn’t waste money/time on parts that didn’t fit in to the MVR. So began a design process that was very educational, but at times became circular to the point that I experienced decision paralysis. I’ve been reading a textbook - Underwater Robotics by Moore, Bohm, and Jensen - and they call this the “design spiral.” ...

The idea to build a ROV originated at the confluence of two things: A desire to apply my software expertise to the physical world. Living in Berkeley right by the water. I have primarily worked on purely virtual software throughout my career, and it feels like something is missing. I love the physical world - being in nature, working with my hands, interacting with people and places - and want to have a positive impact on it. When you apply software to the physical world, you get robotics! ...