Relatively shorter blog this week as we begin to wrap up the semester! However, these final few pool tests are still vital. So, let us see what happened this past weekend.
04/11/2026
Oogway
This past Saturday, Oogway focused on gathering acoustics data. Specifically, previously gathered acoustics data was centered at 32 kHz. However, during competition, pings can be anywhere from 25-40kHz. Therefore, at this pool test, we collected samples at 25kHz and 40kHz. We hope to see how our models trained on 32kHz data will generalize to the new samples, and see what adjustments we might need to make.
Crush
For this test, Crush worked on tuning static power once more. After some tuning, we found that Crush could reliably stay underwater while no specific movement input was given, meaning that the applied downwards thrust was sufficient for it to keep depth. After this, we worked to tune X, Y, Z, and Yaw PID.
04/12/2026
Oogway
On Sunday, Oogway worked to test our new move to CV object algorithm. In a previous pool test, we detailed how calculating the bounding box size of an object to determine relative distance is unreliable due to clipping on the frame of the camera. This week, we tested a modified algorithm that adjusts movement when the object is hitting the edge of the frame.
When running tests with the robot moving 4 meters at a time towards the object, while we observed improvement, the robot still yaws when the object hits the side of the tarp. We suspect that this yaw is due to the fact that when the robot tries to move 4 meters at a time, the thrusters run at full power, not giving them enough control to fix the yaw. However, this may also be a PID issue, as Oogway PID has not been tuned in some time. We will continue to investigate this week and at our final set of tests.
Crush
For this test, Crush tested our current coin flip algorithm code and movement, which worked as expected. Unfortunately, while trying to test our barrel roll, water leaked and spilled over the electrical multiplexer. While it did not break the electrical stack, we paused testing to investigate out of an abundance of caution.
Post-Pool Test Thoughts
Unfortunately, we only have one more set of tests this semester. With a lot of code needed to be tested still, next weekend will definitely be busy. We are also looking into some summer pool testing, so if that happens, be sure to check here for updates! That is all for this week.