You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Proposal: dock-drive source-time freshness and odometry lease
#82
Proposal to harden the OOMWOO dock-drive node against stale detector poses and odometry, while making its state stream observable to a late-joining command relay.
The implementation snapshot is preserved on contributor branch contrib/reactive-planner-dock-preserved-Valdi320-20261101, commit aafdf547. It is intentionally not included in the KeepoutFilter PR #1.
Proposed changes
Accept detector poses only in the configured scan frame, with a positive, recent, strictly increasing source timestamp.
Reject non-finite pose values and malformed/non-normalized quaternions before updating the dock estimate.
Require odometry in the documented odom -> base_footprint frame pair, with fresh monotonic stamps and finite linear/angular velocity.
Propagate the carried dock estimate using the source-time delta rather than callback receipt time; discard it when the unobserved interval exceeds the bounded propagation window.
Publish the current dock state periodically, including IDLE, so a late-joining relay can observe a live state rather than relying on a one-shot transition.
Hold with zero velocity when the pose/odometry lease is invalid; the reverse-seating budget is a kinematic lab bound, not measured physical braking.
Evidence and limits
The preserved commit adds focused dock tests for stale/replayed source stamps, frame and quaternion validation, odometry freshness, estimate propagation and state publication. These are software/ROS behavior checks only. They do not qualify physical braking distance, bumper/cliff stops, docking contact, charging, or hardware motor arbitration.
Questions for upstream
Is base_scan the canonical detector pose frame, or should the detector publish directly in base_footprint?
Is odom -> base_footprint the required odometry frame contract across supported robot configurations?
Should the periodic state stream be the canonical contract for the NAV/SWEEP/DOCK owner, or is there an existing status/action interface to use?
No upstream runtime files are changed by this discussion.
Hi @Valdi320, thank you for splitting these out. #81 is perhaps the direction I'd prefer. Let's land the keep-out PR first, then a small PR for the reactive executor, opt-in and off by default. #82 looks sensible as a follow-up PR on its own.
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
Uh oh!
There was an error while loading. Please reload this page.
Summary
Proposal to harden the OOMWOO dock-drive node against stale detector poses and odometry, while making its state stream observable to a late-joining command relay.
The implementation snapshot is preserved on contributor branch
contrib/reactive-planner-dock-preserved-Valdi320-20261101, commit aafdf547. It is intentionally not included in the KeepoutFilter PR #1.Proposed changes
odom -> base_footprintframe pair, with fresh monotonic stamps and finite linear/angular velocity.Evidence and limits
The preserved commit adds focused dock tests for stale/replayed source stamps, frame and quaternion validation, odometry freshness, estimate propagation and state publication. These are software/ROS behavior checks only. They do not qualify physical braking distance, bumper/cliff stops, docking contact, charging, or hardware motor arbitration.
Questions for upstream
base_scanthe canonical detector pose frame, or should the detector publish directly inbase_footprint?odom -> base_footprintthe required odometry frame contract across supported robot configurations?No upstream runtime files are changed by this discussion.
All reactions