I started this semester by emailing every professor in the mechanical engineering department asking for something to work on. Nothing came through at first, but a couple weeks later Dr. Jensen emailed me asking if I wanted to work on an autonomous electric airplane tug. I didn't have to think much about my response. Who says no to that?
I'm now on a team of three working on the third major iteration of the "E-Tug," and this post is the intro. More write-ups coming as we actually build the thing.
The problem it's solving
Airliners taxi on their jet engines, and that burns a ton of fuel. Something like 70-80 gallons just going from the jetway to the runway at airports like SLC. In 2019 alone, electric tugs could have saved airlines at Salt Lake City International about $2.17 million in fuel, plus the noise and emissions equivalent of taking 2,400+ cars off the road. It also gets tarmac workers away from live jet engines, which creates safer work environments.
Where the project came from
The previous etug was a senior capstone project from 24-25. It was a proof of concept that displayed hands-free coupling and decoupling, remote control over a cellular datalink with a live video feed, and a redesigned drivetrain. It was demonstrated at the Provo Airport, won first place at the 2024 ACRP University Design Competition, and has patents pending.
The old etug v2.0 pulling a Diamond DA20
The old etug v2.0 ready for the dump after we ransacked it for parts. As you can see we managed to get the old F-22 mount on top with quite a struggle. You can tell by the brand new chip in the floor between the front tires.
The old mount from when the tug was pulling F-22s. This thing was like 600 lbs and not fun to move.
What we're doing in v3
My team is three people: Kaimen is in charge of sensors, Jesse is in charge of mechanical design, and I'm the bridge between the two since I study both mechanical and electrical engineering. Right now I'm in charge of integrating the LiDAR data and mounting all the LiDAR sensors to the machine.
For v3 we're upgrading a remote-controlled tug prototype donated by Oshcut Aerospace into a fully autonomous one. The goal is a working prototype by spring to prove commercial viability for airliners. Our end goal is a building a complete fleet of tugs that park and back out airplanes at an airport with zero human intervention. They're kind of like Tesla cybercabs, but for planes (How cool is that? How did I get here?)
The front view of the tug. We are working on installing the tire mount right now.
The back of the tug. As you can see, we are currently gutting all of the old electronics.
Current work on the mechanical side is the upgraded tire mount which is the piece that cradles the aircraft's nose wheel.
The upgraded tire mount that we're working on installing
The 3D model of the new tire mount
The LiDAR (my favorite part)
For the LiDAR we're using an LS CH32, a 32-line hybrid solid-state LiDAR as the proof-of-concept sensor. To summarize what that means, there's no big spinning barrel on top like the old autonomous car rigs. A small internal mirror steers the laser instead, so the whole unit is compact and solid. The "32-line" part means it sweeps 32 horizontal beams at once, which gives you a dense point cloud instead of a flat laser scan.
This sensor is way bigger and cooler than anything I've played with before, and I'll to be honest, I've really been nerding out about it. I used to own a Matterport Pro 2 that I took real estate scans with, and the CH32 is on another level.
A distance test from the LS CH32 32-line hybrid solid-state LiDAR. Over 200m during the day! This is way better than my old 3D camera.
A visual representation of a busy street from the CH32 mounted on a self-driving cab
Why this project
It combines my two big engineering interests mechatronics and airplanes in one machine. I've had a ton of fun so far and I can't wait to see where it is by spring. Who knows, maybe I'll end up an airplane tug designer someday.