Skip to content

Attempt to fix Odometry inaccuracies - #27

Open
jeffzakrzewski wants to merge 3 commits into
mainfrom
fix_odo
Open

Attempt to fix Odometry inaccuracies#27
jeffzakrzewski wants to merge 3 commits into
mainfrom
fix_odo

Conversation

@jeffzakrzewski

@jeffzakrzewski jeffzakrzewski commented Feb 11, 2026

Copy link
Copy Markdown
Contributor
  1. Don't use error prone relative encoders for angle for odometry or wheel spin cals, use the non slipping absolute encoders.
  2. Remove 4x threads by shifting sync encoder operation to updateModules thread
  3. Introduce optional gradual encoder sync - more accurate driving, at cost of ~5% canbus utilitzation
  4. Optimize Limelight tag acquisition, don't just accept all updates all the time. Put a distance scaled STDDEV, penalty for < 2 tags, jumpy distance cutoff.
  5. Must do wheel diameter measurements after this is applied.

…e the much more accurate absolute encoders. 2. Optimize Limelight tag aquisition, don't just accept all updates all the time. Put a distance scaled STDDEV, penalty for < 2 tags, jumpy distance cutoff
…spin calcs

- Remove separate 4 x threads for encoder sync, can happen via the updateModules() thread instead
- Introduce a mode where gradual correction is sent to relative encoders much more frequently. estimated canbus impact is 5% so measurement needed to see if this can be supported.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant