I’m finally done! At 26 years old I’ve finally finished my degree in Mechanical Engineering. The 8-year journey deserves its own post which I’ll likely get to in the coming month. For this post however, I’d like to instead debrief about my experience in senior design and compile some of my work from throughout the year.

At UCF our senior design project is split into two semesters with two main goals: Senior Design 1: Research and Concept Design and Senior Design 2: Manufacturing and Testing. During SD1 there were a lot of interesting challenges; from Monkey Tracking (sponsored by Northrop Grumman the aerospace defense company?!) to McDonalds Delivery Drones. I wanted to work on a more interdisciplinary project so I ended up choosing the Aerojet-sponsored Autonomous Rover project. Aerojet sponsored a few rocket and payload teams with the ultimate goal of having UCF participate in the FAR 51025 competition in the Mojave Desert. The rocket teams were working together on a single rocket while the individual payload teams were competing to be the chosen payload.

Our payload’s mission was to ride the rocket to 30,000 feet, parachute down, release a tiny rover and have the rover navigate back to the launch site. As the sole-mechanical engineer on the project, I was tasked with designing the rover and the payload canister it would ride in. Our computer engineer would help me with the electronics, power distribution, and controls engineering while the remaining four CS students would work on the autonomous navigation code.

The first iteration of our rover was the simplest approach just stick all of our necessary components together as close as they’ll go, add a simple frame and put some wheels on it.

While this was definitely an approach, it wasn’t a good one. For the wheels to fit radially in the rocket (which only had a diameter of about 5 inches) they had to be incredible thin, wheels with a small contact patch like this would just dig themselves into the sand in the desert. Also with the chassis and motors being so low to the ground, it would be extremely easy to get stuck on rocks and small holes. Before moving to our next design, I made a barebones robot file for the CS team for them to use in their simulations.

With our second iteration I decided to use our geometry constraints to my advantage:

While I didn’t have a lot of radial space to work with inside the rocket cylinder, I had an abundance of lengthwise space. Around this time, the rocket team told us that their simulations gave us an estimated travel distance back to base of about a mile. A mile is a really long distance for a rover the size of a rabbit so I decided to sacrifice simplicity a little here and give our rover legs. These legs would stow in front of and behind the rover, and upon deployment unfold outwards and downwards to give the rover a wider wheelbase, greater ground clearance, and above all allow us to fit some decent sized wheels!

I spent quite a lot of time working out the best way to kinematically achieve this motion with the fewest points of failure and modifying the design as well as the equations I derived to accommodate the quirks of this approach (i.e the wheels rotating undesirably during deployment). At this point I started printing our first prototype and our parts were on order. That is until we were told halfway through the semester that our budget had changed and instead of having quite a bit of spare change after all of our parts were purchased, we actually had only half of the parts we needed with no money left over at all. From here, I did a little begging to our procurement office and eventually got an order placed for two motors instead of our original four. I began salvaging parts from old RC cars at goodwill and started CADing everything that would be useful for our project

Since the first axles I modeled were too wide to accommodate large wheels I started to model my own 3D-printed axle assemblies that could replace the rigid one. The bonus of this approach is it gave us built-in suspension. It was around this point that the motor I ordered to dimension and test was due back to amazon and two should have been ready for me soon from procurement. It was at this point that I stopped hearing back from procurement over email. After a few unrequited panicked requests I decided to break a rule that my professor had set and show up to our procurement office in person when I knew they would be there. Our parts had been there for over a month but they had ordered the wrong number for each of our parts. Only one motor and two LiDAR sensors. Additionally the motor turned out to be DOA, so I had to order a new motor (I could only afford one as well) and once again redesign the rover once more to only use one motor. The main catch being I now had one week for design, manufacturing and testing before our final demo and presentation.

Needless to say that I had a really bad time this last semester in senior design. Week after week I would work all day long every single day on the project and as soon as I began to make progress, something outside of my control would swoop in and change everything. This project really took a toll on me and I don’t think I want to grace it with more time on my blog. Maybe one day I’ll go back and fill in some of the blanks and clean up the formatting here but I think for now I’ll just leave some of the highlights here and start working on something new.