On Saturday, Max generated a fully synthetic dataset and trained a CV model to detect the buoy. In addition to the four buoy glyphs, it also detected the entire buoy – something which I asked Max to add on Saturday after realizing it would be useful for sonar testing. On Sunday, we tested the new …
Prior to the pool test, we replaced the bottom camera with poker chips of equivalent weight to prevent the camera from potentially being damaged. We also added quite a bit of tape around that end today as a temporary measure to prevent too much water from getting into the bottom camera capsule and changing the …
Before going to the pool today, I tested the capsule that was repaired yesterday in the sink, and it worked well – no leaks. We got to the pool on time. The issue with sonar appearing as busy to CV was fixed when it could not find the object. We tested the fix on land, …
We got to the pool about 20 minutes late due to the acrylic separating from the capsule we had been using. We instead used a non-ideal capsule. We tested the sonar. The new sonar algorithm using scikit-learn worked well, returning accurate distances to the buoy. If sonar can’t find the buoy, CV would think that …
We were delayed in getting to the pool by two hours due to the time needed to wire the bottom camera and resolve other electrical issues. We only had about 45 minutes at the pool. We tested recent changes to controls and task planning. We resolved some bugs related to tf2 in task planning, but …
We got to the pool with only a small delay (we cut and drilled buoyancy blocks at the Foundry before leaving). We adjusted the buoyancy, making the robot almost level. We tuned x, y, z, and yaw PID. The robot can hold depth and rotate (yaw) with great accuracy. X and Y movement is also …
We were delayed by an hour in getting to the pool due to issues with the pressure sensor, camera capsule, and tether. We moved the robot with the joystick; it behaved as expected given the updated nonlinear thrusters and controls code. We then tried to begin tuning Z PID, but couldn’t make much progress given …
We left for the pool having ensured that all sensors worked, but didn’t test the tether. At the pool, we couldn’t connect to the robot via the tether. We narrowed it down to an issue with the ethernet head plugged into the robot’s switch. New ethernet heads were crimped onto two ethernet cables plugged into …
Last friday we executed our first fully-integrated test. This involved carefully sealings all the pressure hulls, loading up the golf cart with the robot, the support frame, the ethernet reel, buckets of supplies, laptops, and four people. It took us two trips to actually get everything. I think we can actually reduce it to one in the future. In the …
Copyright © 2025 Duke Robotics Club. All Rights Reserved.