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: reactive row executor for coverage sweeps (Nav2 transit-only)
#81
Proposal to evolve the OOMWOO coverage planner with an opt-in reactive row executor while retaining Nav2 for transit goals between sweep segments.
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 behavior
Keep the existing Nav2 waypoint execution as the default.
Add an opt-in reactive mode that drives sweep rows through a dedicated velocity input and uses Nav2 for longer transits.
Require a single-owner command relay for NAV/SWEEP/DOCK handoffs; fail closed on stale or invalid keepout data and ownership evidence.
Validate map/keepout geometry and preserve forbidden-zone checks in the reactive path, since direct /cmd_vel motion bypasses Nav2 costmap filters.
Evidence and limits
The preserved milestone report records one native Gazebo test_room profile at 95.38% coverage and 80.47% efficiency at the first 90% crossing, with zero forbidden crossings in the recorded audit. This is a single configured scene/profile, not multi-room or repeated-seed reliability, physical cleaning, motor arbitration qualification, or MCU safety validation. The associated regression runner, coverage meter, lifecycle configuration and map-generator changes are in the same preserved snapshot and need a focused review/split before proposing code for merge.
Questions for upstream
Is the intended ownership boundary still: reactive executor owns row motion, Nav2 owns transit/docking/localization?
Which existing package/node should own the exclusive velocity arbitration contract?
Should the public planner API expose the execution mode as a parameter, action goal field, or separate behavior?
No upstream runtime files are changed by this discussion.
The preserved commit also changes the stock nav2_params.yaml. That change is deliberately not in KeepoutFilter PR #1; I am treating any simulator tuning as a separate opt-in profile proposal rather than silently changing the project's stock Nav2 behavior. Please advise whether those tuning changes should be omitted or proposed as a separate simulation-only PR.
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.
Uh oh!
There was an error while loading. Please reload this page.
Summary
Proposal to evolve the OOMWOO coverage planner with an opt-in reactive row executor while retaining Nav2 for transit goals between sweep segments.
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 behavior
reactivemode that drives sweep rows through a dedicated velocity input and uses Nav2 for longer transits./cmd_velmotion bypasses Nav2 costmap filters.Evidence and limits
The preserved milestone report records one native Gazebo
test_roomprofile at 95.38% coverage and 80.47% efficiency at the first 90% crossing, with zero forbidden crossings in the recorded audit. This is a single configured scene/profile, not multi-room or repeated-seed reliability, physical cleaning, motor arbitration qualification, or MCU safety validation. The associated regression runner, coverage meter, lifecycle configuration and map-generator changes are in the same preserved snapshot and need a focused review/split before proposing code for merge.Questions for upstream
No upstream runtime files are changed by this discussion.
The preserved commit also changes the stock
nav2_params.yaml. That change is deliberately not in KeepoutFilter PR #1; I am treating any simulator tuning as a separate opt-in profile proposal rather than silently changing the project's stock Nav2 behavior. Please advise whether those tuning changes should be omitted or proposed as a separate simulation-only PR.All reactions