These two pool tests focused simply on prequalifying our main robot, Oogway.
03/21/2026
On Saturday, we began by re-testing our movement underwater. We observed that the robot was slightly off balance compared to the previous pool test, while buoyant at the surface of the water. This may be due to 3D prints taking on water, or our battery capsule being looser than before (new clasps coming soon!).
After fixing these issues by shifting around buoyancy blocks, we made several runs of the length of the pool and observed varying results. Currently, our prequalification code runs off an old version of main, and legacy code does not have many of the new fixes we have implemented to common movement and detection tasks. We focused on making some small tweaks to get this old code into a usable state.
03/22/2026
Sunday was the big day, with a three-hour testing window and two lanes reserved, we could finally build out the full prequalification course. The first half of the pool test focused on tethered runs. We continued to observe issues when the robot attempted to move forward. We reasoned that, due to the placement of heavy and blocking elements on the robot, the robot drifts left while moving. To fix this, we tried the following quick fixes to a likely larger systemic issue:
- Decreasing the amount the robot moves forward at a time to a maximum of 1-meter increments.
- Continually using the gyro after each step to correct the robot’s yaw.
- Adding small strafing movements to re-align the robot onto course.
After this, we were able to get the robot to move down the pool length, and we proceeded with untethered runs during the final hour of the test. We were able to prequalify the robot on the second untethered run (the first run had some testing code still present, and thus the robot began moving backwards after submerging). It submerged, went under and through the gate, down the pool, and around the buoy.

We uploaded both a video taken from the land as well as a video taken underwater by our swimmer as proof of prequalification, and are awaiting a result.
Post-Pool Test Thoughts
After several weeks of lane availability issues, we were finally able to prequalify Oogway! This allows us to move on to the newly released and refined tasks for Robosub 2026. We were hoping to begin testing the task code from previous years at Sunday’s pool test, but ran out of time to debug some small issues. Look forward to this in future updates!
Meanwhile, Crush is having some electrical issues due to the addition of two thrusters, an ongoing project from last semester. Unfortunately, due to reliability issues, Crush has yet to get into the pool this semester. We hope that it can join Oogway in the pool soon, as Crush requires testing time. When Crush does get in the pool, we will be sure to write about it in a blog, so stay on the lookout. That is all for this week!
