Add queued execution speed and explicit pause/resume - #45
Draft
Jepson2k wants to merge 8 commits into
Draft
Conversation
execution_paused was only ever cleared by an explicit resume, so a pause followed by a program stop left every later queued command silently held by the segment player with no error and an empty-looking pipeline. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
…cution-speed-override # Conflicts: # parol6/server/controller.py
…cution-speed-override
…cution-speed-override
…nfirmed as 0 A queued delay counted nominal control ticks, so on a loaded controller -- where periods routinely exceed the nominal one -- every delay stretched by the loop's cumulative overrun: five seconds became six at 20% overrun. It measures real time now, with each tick's contribution capped, because the player stops ticking a paused delay altogether and that gap arrives as one enormous tick. The trajectory hot path ran the per-joint acceleration-window search and the admissibility bisection on every tick of every playback, including the common case where nobody has overridden the speed and the rate they compute is forced to zero. They run when the scale is actually changing. pause(), resume() and set_execution_speed() document 0 as "unconfirmed", but a readback whose reply was lost inside the confirmation window raised ConnectionError -- telling the caller the controller was unreachable when it had acked the command moments earlier. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
…cution-speed-override # Conflicts: # parol6/client/dry_run_client.py # parol6/commands/utility_commands.py
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Make queued execution speed a controller capability.
set_execution_speed(0.1..1.0)retimes the existing planned joint path;pause()decelerates to a hold with the queue retained, andresume()uses the selected positive speed. Changing speed while paused keeps the hold. Fresh readback reports selected, requested, and applied speed separately.The player interpolates the original joint path before step quantization. Rate changes are checked against acceleration derived from the emitted positions, with preallocated buffers in the control loop. Queued dwell time pauses; positive overrides leave dwell and independent actuator timing unchanged. Jog and servo-stream timing remains separate. Override transitions have their own rate and acceleration bounds; nominal motion-profile jerk ceilings are not guaranteed during a transition.
Standalone completion waits retain wall-clock deadlines. Unconfirmed blocking motion completion raises
TimeoutError; the queued motion remains until completed or stopped. Preview retimes trajectory durations and reports paused operations as unresolved.Depends on #44 and waldoctl #32. Coordinated branches use
feat/execution-speed-override.Validation: 25 motion-pipeline, hostile-wire-input, and simulated client tests passed; preview checks and lint/type/format checks passed. The acceleration regression reproduced a 20% excess under a fast transition before the emitted-position check fixed it. Validation used simulated execution; physical transition tuning remains unverified.
Companion PRs: waldoctl #32, PAR6 #75, Commander #58.