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, and it worked.
We got the robot in the water, and immediately we could no longer connect to the sonar. We then pivoted to testing task planning’s move to CV object function. We resolved a few bugs without enabling controls, but just as we were about to try it with controls for the first time, we found a major leak on the robot.
One of the ports on the bottom wasn’t properly sealed and thus a lot of water had gotten into the main capsule. Additionally, the bottom camera capsule also had quite a bit of water in it. We ended the pool test and took the robot back to the Foundry.
At the Foundry, we discovered that the cause of the issue was the ethernet cable connecting the bottom camera to the electrical stack. It was not properly epoxied on both ends, thus allowing water in the main capsule and bottom camera capsule. Thus, it seems that the capsule was never the issue (even yesterday); the bottom camera port was the issue. Additional epoxy was applied to make it (hopefully) watertight.
The water getting into the capsule may also have contributed to our inability to connect to the sonar.
Today’s events further underscore how useful it would be to have a humidity/water level sensor to catch such issues early.