amr_handover_dynamic@v0 — AMR dynamic handover (spec-only candidate)
A mobile robot hands a part to a fixed arm without stopping, so the cooperation channel is arrival-timing and motion rather than contact force. The task would extend the benchmark's coupling vocabulary beyond force coupling. This is a spec-only placeholder: kinematic and contact-solver-free by design, it is a low-build-risk v1.x candidate.
Tier 3 — documented candidates and closures · Status: CANDIDATE · Suite: CHAMBER-Bench v1.0 (adr/ADR-027-chamber-bench-v1-protocol.md)
Spaces
- Action space. To be specified: a mobile robot hands a part to a fixed arm without stopping (kinematic, no contact solver).
- Observation. To be specified at Gate-0 pre-registration.
- Simulator backend.
unspecified (spec-only)· Agents. 2
Stress channel
None — no force coupling channel.
Axis validity
Per-cell status under the validity matrix (adr/ADR-027-chamber-bench-v1-protocol.md):
| AS | OM | CR | CM | PF | SA |
|---|---|---|---|---|---|
untested |
untested |
untested |
untested |
untested |
untested |
Industrial analogue
Non-stop transfer between an autonomous mobile robot and a fixed picking arm in intralogistics: the AMR streams past the station and the arm picks from it on the move, buying cycle time that a stop-and-transfer layout gives up.
Committed falsifier
If a stop-and-pick fallback matches non-stop performance under realistic cycle time, the task is deconfliction-plus-pick and is not admitted.
Evidence
How to run one episode
import chamber.tasks
# Spec-only placeholder — no env factory is wired yet:
chamber.tasks.make("amr_handover_dynamic") # raises NotImplementedError
Notes
Timing-coupled cooperation candidate: the coupling channel is arrival-timing/motion, not contact force, so the task would extend the benchmark's coupling vocabulary beyond force coupling (stress_channel is None by design). Committed falsifier: if a stop-and-pick fallback matches non-stop performance under realistic cycle time, the task is deconfliction-plus-pick and is not admitted. CPU-friendly to build (kinematic, no contact solver) — a low-build-risk v1.x candidate.