Today, we tested the fiber optic gyroscope for the second time in the pool.
Gyro Testing
- We were able to produce the same results as our pool test yesterday.
- The robot performed a depth hold for several minutes without any noticeable yaw drift.
- We started the robot in a corner of the pool, moved it by hand to the other end of the lane, and returned back to the same corner in the same orientation. We found that the yaw value in state drifted by 2-3 degrees.
- At the beginning of the pool test, we experienced issues where the gyro stopped publishing values intermittently. As we were able to connect to the gyro, we believe this issue is coming from the external trigger signal sent by the peripheral arduino.
- We added queues in sensor fusion to account for the difference in publish rates between the gyro and IMU. More testing should be conducted to determine whether these changes are actually beneficial.
Sonar Testing
We tested sonar and were able to get scans of the back pool wall. To start using the sonar in task-planning (e.g., in the octagon task), a sonar task-planning interface must be made. Additionally, it would be helpful to output a sonar polar image in addition to the cartesian image we already have.
Coin Flip Testing
With our new gyro tested, we attempted to use the remainder of our time in the pool to test the coin flip task. When we executed the coin flip task from RoboSub 2024, we noticed that Oogway would roll and pitch substantially during the level corrections. When we killed task-planning, we found that roll and pitch values have substantially drifted.
We attempted to reproduce by hand by rolling and pitching Oogway about ±45 degrees and returning Oogway back to center. We noticed that roll had drifted by 100 degrees and pitch had drifted by 30 degrees. While we expect drift after extreme movements such as a barrel roll, the amount of drift experienced here seems abnormal. The next pool test should attempt to reproduce this issue.
Action Items
- The broken battery capsule latch should be fixed. We were having problems with the battery cable again which were temporarily fixed with zip ties.