Progress: where we are, with evidence

What we are building: An open-source dual-arm mobile manipulator: our own mobile base with a DOLD Hexalift V4 lift and two OpenArm 2.0 arms, ROS 2 Jazzy software, released as a package a second site can build. The 2.0 target application is Use Case 1.
Where we are: Base navigation works in simulation only with fixtures; the safe command path, docking acceptance, 2.0 base firmware and the combined robot model are not on main; no physical acceptance is recorded. Since 19 Oct: command arbitration, the recovery behaviour tree and the STM32 CubeMX project are in review; only the Stop (software) label reached main. 111 PRs from 15 authors since 1 Oct (41 open).
What is next: Gate A and the electrical and safety release on 30 Oct, the I8 freeze and STM32 go/no-go on 6 Nov (P-02 rev19.2), the 13 Nov issue dues and Gate B regression 13 to 20 Nov, then v2.0.0-rc.1 on 20 Nov and physical integration from 23 Nov.

runtime: works in simulation with fixture runtime evidence (rounded) · physical: none recorded physical acceptance (box)

Attention now

Navigation

On main the simulated robot did not move: 0 of 3 goals, 0.000 m travel. With the fixture, 3 of 3 goals. Consequence: no navigation evidence on main is valid until the LiDAR body filter is fixed.

owner @skunal3318 (navigation lead) · scenarios S1 · code · closes when: Rerun S1 on main: 3 of 3 goals

Operator UI

UI teleop drove the simulated robot into a wall at 0.2 m/s. After the UI connection was lost the robot continued 2.25 m. Recording replay moved it 0.672 m with no operator command. Consequence: every UI motion path bypasses the reactive layer.

owner @parthkeshari (UI lead); reviewers @panthera-momagdii (tech lead, UI reviewer) and UI reviewer (GitHub handle to be confirmed); simulation checks @nikgolobo (simulation checks) · scenarios S4, S5, S7 · code · closes when: Rerun S4, S5 and S7a: no UI or replay publisher on /cmd_vel; stop before the wall; stop within 0.5 s after the connection drops

Docking

0 of 5 docked; 5 of 8 planned attempts ran. The navigation stack reset during docking because docking switched the collision monitor off. Consequence: docking has no reactive layer and no successful run in simulation.

owner @fridayali (docking and robot description) · scenarios S3 · code · closes when: Rerun S3: 5 of 5 docked; collision monitor the sole /cmd_vel publisher; zero lifecycle transitions

OA-04-P2 · scenario S3 · code openamr-platform-sw/ros2/src/openamrobot_docking/scripts/dock_trigger.py:833 · owner @skunal3318 (navigation lead) · closes when: Rerun S3: 5 of 5 docked; collision_monitor the sole /cmd_vel publisher; zero lifecycle transitionsOA-68 · scenario S3 · code openamr-platform-sw/ros2/src/openamrobot_docking/scripts/dock_trigger.py:833 · owner @fridayali (docking and robot description) · closes when: Rerun S3: reported monitor state equals the lifecycle state in every phaseOA-53 · scenario S3 · code openamr-platform-sw/ros2/src/openamrobot_docking/scripts/dock_trigger.py:942 · owner @fridayali (docking and robot description) · closes when: Undock when not docked is refused; cancel stops within 0.5 s; no automatic undock after a failed dockOA-54 · scenario S3 · code openamr-platform-sw/ros2/src/openamrobot_docking/scripts/dock_trigger.py:1800 · owner @fridayali (docking and robot description) · closes when: Rerun S3 at 0.1 real-time factor with the same parameters as at 1.0: no wall-clock timeout

Stop and RESET (AR-05)

Motion commands are not held at zero across an E-stop and RESET: no host node reads the safety-relay state, motion sources keep publishing while actuator power is off, and the UI stop does not latch. Consequence: after RESET, the last command of any direct source can reach the base.

owner @skunal3318 (navigation lead) · scenarios S6, S7 · architecture review · closes when: After RESET, no motion from any source until an explicit operator release; verified in simulation and on the bench

NR-03 · owner @skunal3318 (navigation lead) · closes when: After RESET, no motion from any source until an explicit operator release; verified in simulation and on the bench

Obstacle tests

No contact with sudden obstacles, but the collision monitor never engaged; the planner avoided them with 0.117 to 0.128 m clearance. Consequence: the stop is not demonstrated.

owner @skunal3318 (navigation lead) · scenarios S2 · code · closes when: Rerun S2: clearance at least 0.25 m and the collision monitor engages in the barrier test

OA-35-P2 · scenario S2 · code openamr-platform-sw/ros2/src/openamrobot_nav2/config/nav2_params.yaml:177 · owner @skunal3318 (navigation lead) · closes when: Rerun S2: clearance at least 0.25 m; collision monitor engages in the barrier testOA-69 · scenario S10, S12 · code openamr-platform-sw/ros2/src/openamrobot_nav2/config/nav2_params.yaml · owner @skunal3318 (navigation lead) · closes when: Collision monitor and local costmap list the ToF and ultrasonic sources, or the register records their role

Manipulation

48 of 48 plans on mock hardware (branch, not main). One park pose sits exactly at a joint limit, and the lift is not a planned joint. Consequence: lift height is not part of planning.

owner @Tushargoel230 (upper-body model) · scenarios S8 · code · closes when: Rerun S8 with the lift in the planning group: 48 of 48 and every margin above 0.05 rad

OA-64 · scenario S8 · code PR #14 (branch evidence) · owner @Tushargoel230 (upper-body model) · closes when: Rerun S8 with the lift in the planning group: 48 of 48 and every margin above 0.05 rad

Coverage

85 requirement rows: 26 need hardware, 20 missing, 16 partial, 6 implemented untested, 5 documented only, 5 out of scope, 4 tested in simulation, 3 contradicted.

owner per row · traceability panel · closes when: See the traceability panel

v2.0.0-rc.1 readiness

19 items

Audit pass 2 recommends not cutting v2.0.0-rc.1 until the items below land. This page states the dependencies; it does not decide the release. Target 20 Nov 2026, 32 days after this snapshot.

rc.1 depends on	Delivery	Findings	Owner
UI teleop, blocks and voice actions enter an arbitrated input ahead of the smoother; the browser never publishes /cmd_vel	IN REVIEW platform-sw #90 (UI TwistStamped on /ui/cmd_vel), changes requested, sign-off missing	OA-02-P2 OA-45-P2 OA-52 OA-62	@parthkeshari (UI lead)
#35 option (b): docking commands through the chain, collision monitor never switched off, truthful monitor status	IN PROGRESS	OA-04-P2 OA-68	@skunal3318 (navigation lead); @fridayali (docking and robot description)
UI no longer starts its own navigation or mapping stack	NOT STARTED	OA-03-P2	@parthkeshari (UI lead)
Recording replay restricted to non-command topics	NOT STARTED	OA-48	@parthkeshari (UI lead)
UI command path hardening	IN PROGRESS ui issues #31 to #37 open: security hardening in progress	OA-50	@parthkeshari (UI lead)
Simulation body filter fixed on main with a URDF-derived test	NOT STARTED	OA-47	@skunal3318 (navigation lead)
UI command path hardening: access control	NOT STARTED	OA-49	@parthkeshari (UI lead)
Collision monitor and smoother under their own lifecycle manager	NOT STARTED	OA-56	@skunal3318 (navigation lead)
Bounded startup with retries	NOT STARTED	OA-57	@skunal3318 (navigation lead)
Docking deadlines on the node clock	NOT STARTED	OA-54	@fridayali (docking and robot description)
Undock only from docked, cancellable, with rear coverage	NOT STARTED	OA-53	@fridayali (docking and robot description)
Command timeout for the simulated base	NOT STARTED	OA-45-P2	@skunal3318 (navigation lead)
Stop and slowdown polygons; inflation at or above the inscribed radius	NOT STARTED	OA-35-P2	@skunal3318 (navigation lead)
Motion commands held at zero across E-stop and RESET until an explicit operator release (AR-05)	NOT STARTED	NR-03	@skunal3318 (navigation lead)
Latched software stop with its own label (not E-STOP)	IN REVIEW stop lock in platform-sw #90 (changes requested); Stop (software) label merged in ui #38	OA-16-P2	@parthkeshari (UI lead)
Lift as a planning joint; both-arms group; poses with limit margin	IN REVIEW	OA-64	@Tushargoel230 (upper-body model)
Simulated docking camera on base_camera_link (Gemini 336L) and docking retuned	IN REVIEW	OA-59	@fridayali (docking and robot description)
Release reproducible from a manifest with exact SHAs; drivers installable (release #12, manifest #17 and #18, drafts)	IN REVIEW	OA-05 OA-06 OA-39	@KARTHIKEYAN124 (release lead)
Decision: does rc.1 run the Teensy Gate A controller or the STM32 (P-02 rev19.2: STM32 go/no-go 6 Nov, fallback Teensy; input: CubeMX project platform-fw #22 in review)	NOT STARTED	OA-14	to be confirmed

Action plan: what to do, in order

Proposed sequence, ordered by dependency and the critical path; owners and dates only where a source names them.

35 steps; 16 on the critical path (steps 1, 2, 3, 6, 8, 9, 10, 11, 15, 17, 18, 30, 32, 33, 34, 35). Due kinds: confirmed (P-02 rev19.2, the register, a work package or an issue body; no repository uses GitHub milestones), proposed (audit pass 2), or none in any source. Each step shows its state on 24 Oct with the PR or issue it rests on.

35 items

Phase dependency diagram

What this shows: the six phases of the action plan in order, with how many steps each holds and how many are on the critical path.

How to read it: each phase depends on the one before; select a phase to jump to its steps; red arrows carry the critical path.

Source: this action plan (proposed sequence), built from audit pass 2, P-02 rev19.2, P-03 rev19.7 item 12, the register and the Team document · 24 Oct 2026

Phase 0: start now (from 19 Oct)

1
Fix the simulation LiDAR body filter on main with a URDF-derived test so navigation evidence on main becomes valid. critical path
Owner
@skunal3318 (navigation lead)
Status now
NOT STARTED
State 24 Oct
Open as platform-sw #74 (due 6 Nov), no PR yet. The simulation filter assumes an unrotated LiDAR; with the URDF yaw of pi it removes the robot front and keeps the body.
Depends on
none
Due
6 Nov 2026 confirmed (platform-sw #74 issue body)
Done when
S1 on main: 3 of 3 goals; the test fails when the mask covers the forward sector
Links
finding OA-47 · issue #74
2
Get a firm quote with delivery date for the GX11CAC K1/K2 contactors, or approve an alternative, and close the BUY-32 datasheet checks. critical path
Owner
@Evans-Kwashie (electrical and firmware lead)
Status now
BLOCKED
State 24 Oct
BUY-32 is HOLD in the BOM Issue 7.10 purchase view; blocked on quotes. No package is recorded as ordered.
Depends on
none
Due
none in any source
Done when
BUY-32 released: order placed with a delivery date, or an approved alternative recorded in the BOM and the register
Links
BUY-32 risk · finding OA-01
3
Resolve the open shared-base differences (second E-stop, battery pack, encoder supply, release 001 sensor list) so Sheet A, Sheet B and the wire list describe one shared base. critical path
Owner
@Evans-Kwashie (electrical and firmware lead); decisions @BotshareAI (project lead)
Status now
IN PROGRESS
State 24 Oct
WCM-500-24 on both robots is now recorded (P-03 rev19.7 item 23, BOM Issue 7.10 update of 20 Oct).
Depends on
none
Due
none in any source
Done when
Every open difference on the hardware page is closed or moved to an intended controller or sensor variant; Sheet A, Sheet B and the wire list are reissued against E-01 rev1.4
Links
open differences
4
Record the shared-base direction in decisions.yaml through a contract change request.
Owner
@BotshareAI (project lead)
Status now
MERGED AND TESTED (was NOT STARTED on 19 Oct; .github #55 (1c3f516, 22 Oct) records SHARED-BASE)
State 24 Oct
Done by .github #55 (1c3f516, 22 Oct). Open follow-up: the register document label still says rev19.6 (drift 1).
Depends on
3
Due
none in any source
Done when
A register entry for the shared base is merged (done: SHARED-BASE at 1c3f516); the document label still says P-03 rev19.6
Links
register at 1c3f516

Phase 1: Gate A (30 Oct (P-02 F5))

5
Ship the Gate A UI scope: Stop (software) label, software stop lock, basic command routing and stop tests.
Owner
@parthkeshari (UI lead); reviewers @panthera-momagdii (tech lead, UI reviewer) and UI reviewer (GitHub handle to be confirmed); simulation checks @nikgolobo (simulation checks)
Status now
IN PROGRESS
State 24 Oct
"Stop (software)" label landed in ui #38 (22 Oct). The software stop lock and the /ui/cmd_vel relay are not on main; they sit in platform-sw #90 (changes requested 23 Oct, sign-off missing).
Depends on
none
Due
30 Oct 2026 confirmed (P-02 rev19.2 Gate A date; UI scope agreed in coordination, not yet in a plan)
Done when
Simulation check by the simulation reviewer: UI motion cannot bypass an active stop
Links
finding OA-16-P2
6
Define the twist_mux input topics, priorities and timeouts for #35 option (b) in a contract change request. critical path
Owner
@skunal3318 (navigation lead)
Status now
IN PROGRESS
State 24 Oct
Topics and priorities are proposed in platform-sw #90 (stop lock priority 255, UI TwistStamped on /ui/cmd_vel); no contract change request yet.
Depends on
none
Due
30 Oct 2026 confirmed (platform-sw #35 issue body)
Done when
Contract change request accepted by the software lead with topic names, priorities and timeouts
Links
issue #35
7
Pass Gate A on the Jetson with the existing Teensy and drivetrain.
Owner
@Evans-Kwashie (electrical and firmware lead)
Status now
MERGED UNTESTED
State 24 Oct
pid.h is not in the repository, so Gate A cannot be rebuilt from platform-fw main; host tests for the Gate A PID are in draft platform-fw #27.
Depends on
none
Due
30 Oct 2026 confirmed (P-02 rev19.2 Gate A, Jetson and Teensy)
Done when
Command, odometry, LiDAR, Nav2, the functional watchdog, recovery and safety-chain status telemetry (functional) pass the host-migration suite
Links
platform-fw #13 · platform-fw #27
8
Implement the LOCKED/RUN safety state machine on the Teensy and clear its verify-before-sending list. critical path
Owner
@Evans-Kwashie (electrical and firmware lead)
Status now
NOT STARTED
State 24 Oct
Issue platform-fw #17 not started; it still names LEV100H contactors (drift 8) and needs an /estop contract change request that does not exist (drift 11).
Depends on
3
Due
none in any source
Done when
Bench tests 1 to 6 of the state machine draft pass and the AR-05 closure test holds on the Teensy path
Links
issue #17 · state machine · finding NR-03
9
Build the release 001 shared base where parts allow and bench test it. critical path
Owner
@Evans-Kwashie (electrical and firmware lead); @ArkaStark (bench and integration)
Status now
NOT STARTED
Depends on
2, 3
Due
none in any source
Done when
Bench record of tests 1 to 6, including the tested PNOZ reset dial setting; K1/K2 tests wait for step 2
Links
equivalence table

Phase 2: batch 1 (6 Nov)

10
Route UI teleop, blocks and voice actions into an arbitrated twist_mux input; the browser never publishes /cmd_vel. critical path
Owner
@parthkeshari (UI lead)
Status now
IN REVIEW (was IN PROGRESS on 19 Oct; platform-sw #90 opened (UI TwistStamped /ui/cmd_vel relay))
State 24 Oct
In platform-sw #90: twist_mux and stop gate with the UI relay; changes requested on 23 Oct; DCO sign-off missing. On main the UI still publishes /cmd_vel (drift 14).
Depends on
6
Due
6 Nov 2026 proposed (audit pass 2 action 1)
Done when
S4 and S7a rerun: no UI publisher on /cmd_vel; stop before the wall; stop within 0.5 s after the connection drops
Links
finding OA-02-P2 · finding OA-45-P2 · platform-sw #90
11
Land #35 option (b): docking commands through the chain without switching the collision monitor off. critical path
Owner
@skunal3318 (navigation lead); @fridayali (docking and robot description)
Status now
IN PROGRESS
State 24 Oct
Issue #35 open, due 30 Oct. On main docking still publishes /cmd_vel directly.
Depends on
6
Due
30 Oct 2026 confirmed (platform-sw #35 issue body); audit pass 2 action 2 proposed 6 Nov
Done when
S3: 5 of 5 docked; collision monitor the sole /cmd_vel publisher; zero lifecycle transitions
Links
finding OA-04-P2 · finding OA-68 · issue #35
12
Stop the UI from starting its own navigation or mapping stack.
Owner
@parthkeshari (UI lead)
Status now
NOT STARTED
Depends on
none
Due
6 Nov 2026 proposed (audit pass 2 action 3)
Done when
S5: building a map leaves one controller_server and the platform map server and localisation unchanged
Links
finding OA-03-P2
13
Restrict recording replay to non-command topics.
Owner
@parthkeshari (UI lead)
Status now
NOT STARTED
Depends on
none
Due
6 Nov 2026 proposed (audit pass 2 action 4)
Done when
Replay of a recording that contains command topics: robot travel 0
Links
finding OA-48
14
Harden the UI command path.
Owner
@parthkeshari (UI lead)
Status now
IN PROGRESS (was NOT STARTED on 19 Oct; ui issues #31 to #37 opened (security hardening in progress))
State 24 Oct
Security hardening in progress (ui #31 to #37 open).
Depends on
none
Due
6 Nov 2026 proposed (audit pass 2 action 5)
Done when
Security review of the UI backend passes
Links
finding OA-50
15
Decide which base controller rc.1 runs (STM32 go/no-go, Teensy Gate A as fallback). critical path
Owner
to be confirmed
Status now
NOT STARTED
State 24 Oct
Input: CubeMX project platform-fw #22 is ready for review (changes asked in comments: Ethernet DMA in DTCM, HAL edit, unused UART TX pins); boards/stm32 is empty on main.
Depends on
7
Due
6 Nov 2026 confirmed (P-02 rev19.2 STM32 go/no-go; fallback Teensy)
Done when
A recorded decision names the rc.1 controller
Links
open decision · finding OA-14
16
Freeze the I8 base-controller contract.
Owner
@Evans-Kwashie (electrical and firmware lead)
Status now
DOCUMENTED ONLY
Depends on
7
Due
6 Nov 2026 confirmed (P-02 rev19.2, I8 freeze)
Done when
I8 accepted by the software lead with its consumers listed
Links
finding OA-15

Phase 3: batch 2 (13 Nov (proposed))

17
Run the reactive motion layer (collision monitor and smoother) under its own lifecycle manager with respawn. critical path
Owner
@skunal3318 (navigation lead)
Status now
NOT STARTED
Depends on
11
Due
13 Nov 2026 proposed (audit pass 2 action 8)
Done when
S7b: after killing controller_server the monitor stays active and zero velocity holds
Links
finding OA-56
18
Hold motion commands at zero across E-stop and RESET until an explicit operator release, on the host and in firmware. critical path
Owner
@skunal3318 (navigation lead)
Status now
NOT STARTED
State 24 Oct
The stop gate in platform-sw #90 is a host input only; the firmware side is not started.
Depends on
8, 10, 11
Due
none in any source
Done when
After RESET, no motion from any source until an explicit operator release; verified in simulation and on the bench
Links
AR-05 · finding NR-03
19
Add bounded startup with retries and a startup watchdog.
Owner
@skunal3318 (navigation lead)
Status now
NOT STARTED
Depends on
17
Due
13 Nov 2026 proposed (audit pass 2 action 9)
Done when
S11 extended to 10 starts: all reach active within 120 s simulated
Links
finding OA-57
20
Add a command timeout to the simulated base.
Owner
@skunal3318 (navigation lead)
Status now
NOT STARTED
Depends on
none
Due
13 Nov 2026 proposed (audit pass 2 action 12)
Done when
S7a and S7c: the base stops within the timeout after publisher loss
Links
finding OA-45-P2
21
Add stop and slowdown zones and set inflation at or above the inscribed radius.
Owner
@skunal3318 (navigation lead)
Status now
NOT STARTED
Depends on
1
Due
13 Nov 2026 proposed (audit pass 2 action 13)
Done when
S2: clearance at least 0.25 m and the collision monitor engages in the barrier test
Links
finding OA-35-P2
22
Move docking deadlines to the ROS clock and make undock a cancellable action allowed only from docked.
Owner
@fridayali (docking and robot description)
Status now
NOT STARTED
Depends on
11
Due
13 Nov 2026 proposed (audit pass 2 actions 10 and 11)
Done when
S3 at real-time factor 0.1 with no wall-clock timeout; undock when not docked is refused; cancel stops within 0.5 s
Links
finding OA-54 · finding OA-53
23
Move the simulated docking camera to base_camera_link (Gemini 336L) and retune docking.
Owner
@fridayali (docking and robot description)
Status now
IN REVIEW
State 24 Oct
platform-sw #53 is a draft with changes requested.
Depends on
11
Due
13 Nov 2026 proposed (audit pass 2 action 16)
Done when
S10: image frame base_camera_link; S3 passes with it
Links
finding OA-59 · platform-sw #53
24
Add UI access control.
Owner
@parthkeshari (UI lead)
Status now
NOT STARTED
Depends on
14
Due
13 Nov 2026 proposed (audit pass 2 action 7)
Done when
Security review of the UI backend passes
Links
finding OA-49
25
Latch the software stop with its own label until an explicit release.
Owner
@parthkeshari (UI lead)
Status now
IN REVIEW (was NOT STARTED on 19 Oct; platform-sw #90 opened (stop lock))
State 24 Oct
Stop lock in platform-sw #90 (changes requested); not on main.
Depends on
5, 10
Due
13 Nov 2026 proposed (audit pass 2 action 14)
Done when
S6 plus: after stop, UI teleop and docking produce no motion until release
Links
finding OA-16-P2
26
Make the lift a planned joint with a controller, add a both-arms group and keep poses off the joint limits.
Owner
@Tushargoel230 (upper-body model)
Status now
IN REVIEW
State 24 Oct
Issues upperbody-sw #17 and #18 are open.
Depends on
none
Due
13 Nov 2026 confirmed (upperbody-sw #17 issue body); joint clamp default upperbody-sw #18 also due 13 Nov
Done when
S8 with the lift in the group: 48 of 48 and every margin above 0.05 rad
Links
finding OA-64
27
Investigate the intermittent parameter_bridge crash seen in PR #14 and record the cause.
Owner
@Tushargoel230 (upper-body model)
Status now
NOT STARTED
Depends on
none
Due
none in any source
Done when
Cause recorded in PR #14; A7 reruns on the merge candidate show no SIGSEGV
Links
PR #14
28
Address the review of PR #14 and merge the combined model.
Owner
@Tushargoel230 (upper-body model)
Status now
IN REVIEW
State 24 Oct
upperbody-sw #14 is a draft of 405 files with changes requested: NaN joint samples are accepted and the quality check must be restored.
Depends on
26, 27
Due
none in any source
Done when
PR #14 merged with A6 and A7 passing on main
Links
PR #14

Phase 4: rc.1 (by 20 Nov (register))

29
Run the STM32 Gate B regression if step 15 chose the STM32.
Owner
@Evans-Kwashie (electrical and firmware lead)
Status now
NOT STARTED
Depends on
15, 16
Due
13 to 20 Nov 2026 confirmed (P-02 rev19.2, Gate B regression)
Done when
Gate A host acceptance suite passes on the STM32 and new-drive configuration; bench and physical results reported separately
Links
-
30
Rerun audit scenarios S1 to S12 on main as the rc.1 simulation check. critical path
Owner
@nikgolobo (simulation checks)
Status now
NOT STARTED
Depends on
1, 10, 11, 17, 18, 21, 22
Due
20 Nov 2026 confirmed (rc.1 date, register RELEASE-MILESTONES; the rerun must finish before it)
Done when
Every scenario meets its pass criterion on main; results published with the tested SHAs
Links
scenarios
31
Merge release #12 and manifest #17 and #18 with one agreed pin set.
Owner
@KARTHIKEYAN124 (release lead)
Status now
IN REVIEW
State 24 Oct
release #8 and #9 merged 24 Oct. release #12 (draft) pins 14 sources, manifest #17 pins 11 (drift 12). manifest #13 is approved twice and not merged; manifest #11 is stale.
Depends on
none
Due
20 Nov 2026 confirmed (rc.1 date, register RELEASE-MILESTONES)
Done when
Two builds from the rc.1 manifest give identical checksums; drivers install from the manifest plus rosdep
Links
finding OA-05 · finding OA-39 · finding OA-06 · release #12
32
Cut v2.0.0-rc.1 from pinned SHAs as a GitHub pre-release. critical path
Owner
@KARTHIKEYAN124 (release lead)
Status now
NOT STARTED
Depends on
15, 30, 31
Due
20 Nov 2026 confirmed (register RELEASE-MILESTONES)
Done when
Pre-release published with the pinned manifest, checksums and the simulation check results
Links
-

Phase 5: v2.0.0 (23 Nov to 18 Dec (P-03 item 12))

33
Run supervised bench tests per subsystem, safety chain first (E-stop, contactors, brakes, RESET, charge interlock). critical path
Owner
@Evans-Kwashie (electrical and firmware lead); @ArkaStark (bench and integration)
Status now
NOT STARTED
Depends on
2, 9, 32
Due
23 Nov 2026 confirmed (P-03 item 12, window 23 Nov to 18 Dec)
Done when
Bench record per subsystem; release 001 evidence counted for 2.0 only where the equivalence table says yes
Links
equivalence table
34
Run robot-level physical acceptance, including the 26 rows that need hardware and the stopping tests that set the operating speed. critical path
Owner
to be confirmed
Status now
NOT STARTED
Depends on
33
Due
18 Dec 2026 confirmed (P-03 item 12)
Done when
Each hardware row has a physical test record; physical acceptance recorded
Links
hardware rows
35
Release v2.0.0. critical path
Owner
@KARTHIKEYAN124 (release lead)
Status now
NOT STARTED
Depends on
34
Due
18 Dec 2026 confirmed (register RELEASE-MILESTONES)
Done when
v2.0.0 published from pinned SHAs with the acceptance records
Links
-

Governance and organisation

How decisions are made and who owns what, as recorded in openAMRobot/.github on main (68833f2) and in each repository on 24 Oct 2026. Mismatches are listed under Known drift 20 and 21.

Decision flow

GOVERNANCE.md: an issue, then the source document is updated, then one PR, then owner approval. Botshare LTD has final authority. The decisions register (decisions.yaml, last changed 1c3f516 by .github #55 on 22 Oct) mirrors the documents in force; P-03 rev19.7 is in force by project-lead decision of 24 Oct.

Contributions

DCO sign-off on every commit and the individual contributor agreement ICLA v1.1 (or the corporate agreement). Rulesets on all public repositories: 1 approval, code-owner review, DCO and license/cla checks. openamrobot-embodied-ai has 0 required approvals and no CLA check.

Watchdog

A weekly scan on Thursday at 14:00 Berlin time writes one dashboard issue; every repository calls the shared harness at @main.

Agent rules and state files

Shared agent rules v2 are on main in .github, platform-sw, ui, manipulation, interfaces and manifest; drafts are open for platform-fw (#24) and platform-hw (#35). STATE.md exists only in platform-sw and interfaces, and both are stale.

Roles (maintainers.yaml)

Role	Handle	Covers
`platform-lead`	@BotshareAI	platform, hardware, safety chain, decisions of record, final review
`software-lead`	@panthera-momagdii	robot software, AI, interfaces, agent rules
`ci-owner`	@wikki26	CI/CD, reusable workflows, quality gates
`release-owner`	@KARTHIKEYAN124	release, installation, manifest
`docs-owner`	@anandgawai123456-glitch	documentation site, public extracts
`electrical-owner`	@Evans-Kwashie	base controller firmware, electrical and power distribution, safety chain implementation
`robot-description-owner`	@fridayali	URDF, base robot model, sensor frames, chassis and panel CAD

Code owners (CODEOWNERS on main)

Repository	Code owners
`openamr-platform-sw`	@fridayali, @panthera-momagdii, @skunal3318
`openamr-platform-fw`	@BotshareAI, @evans-kwashie
`openamr-platform-hw`	@AnushkaAMuragod, @BotshareAI, @fridayali
`openamrobot-ui`	@BotshareAI, @panthera-momagdii
`openamrobot-interfaces`	@KARTHIKEYAN124, @panthera-momagdii, @wikki26
`openamrobot-manipulation`	@Brinex45, @panthera-momagdii
`openamr-upperbody-sw`	@Brinex45, @panthera-momagdii
`openamr-upperbody-hw`	@BotshareAI
`openamr-upperbody-fw`	@BotshareAI
`openamrobot-release`	@BotshareAI, @KARTHIKEYAN124
`openamrobot-manifest`	@KARTHIKEYAN124, @panthera-momagdii
`openamrobot-comm`	@BotshareAI, @panthera-momagdii
`openamrobot-embodied-ai`	@SatyaA27, @panthera-momagdii
`.github`	@BotshareAI, @panthera-momagdii, @wikki26
`openamrobot-docs`	@anandgawai123456-glitch, @panthera-momagdii

PRs missing DCO sign-off

"Blocked on sign-off" where the sign-off is the only thing left.

PR	State on 24 Oct
platform-sw #90	changes requested; sign-off missing
platform-sw #92	approved once (24 Oct); needs sign-off and a second approval
platform-sw #93	draft; sign-off missing
platform-sw #66	sign-off missing; needs an /estop contract change request
platform-fw #18	draft; sign-off missing
platform-hw #27	sign-off missing
platform-hw #29	sign-off missing
interfaces #28	approved; blocked on sign-off (the only blocker)
manipulation #20	99/99 tests; one controller-config defect left; sign-off missing

Activity and licences

111 PRs from 15 authors opened since 1 Oct (41 open on 24 Oct); 8 commit authors on main since 24 Sep. Licences: software MIT, hardware CERN-OHL-P-2.0, documentation CC BY 4.0. CITATION.cff is on openamrobot-release main (release #13) and FUNDING in .github (#56). The 2.0 work is announced in GitHub Discussions #27.

Source: openAMRobot/.github GOVERNANCE.md, maintainers.yaml, WATCHDOG.md and CONTRIBUTING.md at 68833f2; CODEOWNERS on main of each repository; PR lists of each repository · 24 Oct 2026

Known drift: 22 items

Definition: a value or rule on which two sources disagree. The roadmap shows the design value (P-03 rev19.7, then E-01 rev1.4, then BOM Issue 7.10) or the rule in force, marks the place with a drift marker and lists both values and both sources here. A drift row closes when the named fix is merged.

#	Item	Design value or rule (source)	Other value (source)	Fix
1	Register document label	P-03 rev19.7 is in force (project lead, 24 Oct); its decisions are already in decisions.yaml (.github #55, 1c3f516)	decisions.yaml still labels the document P-03-rev19.6 (.github main 68833f2)	Register label to rev19.7 (.github PR, platform lead)
2	ToF pitch	10 deg down (P-03 rev19.7 item 19 D6; register near-field entry)	20 deg on platform-sw main (sensor_frames.yaml, sensors.xacro) and in platform-hw near-field.md; draft platform-sw #93 sets 10 deg but keeps 20 deg in comments and docs	platform-sw #93 and platform-hw near-field.md
3	TF-Luna x position	+/-0.390 m, panel (P-03 rev19.7 item 19, BOM Issue 7.10)	+/-0.399 m (lens face) on platform-sw main, in draft #93 and in the register near-field entry (robot-description owner confirmation, 20 Oct)	platform-sw #93 sets +/-0.390, or P-03 and the BOM record the lens-face value; register entry to match
4	Ultrasonic x position	+/-0.390 m (register; P-03 rev19.7 item 19)	model: +0.3906 front, -0.3886 rear (platform-sw sensor_frames.yaml)	Robot description owner to align the model or record the measured values in P-03
5	IMU frame	imu_link (0.0544, 0, 0.1210), yaw -90 deg (platform-sw main and register; P-03 rev19.7 item 19 D5 update)	P-03 item 15 and D5 text still (0.195, 0, -0.044); BOM sensor sheet and P-02 say axes aligned; platform-hw imu.md z 0.133; platform-fw I/O doc and docs index.md x 0.195	P-03 item 15, BOM sensor sheet, P-02, platform-hw imu.md, platform-fw I/O doc and docs index.md
6	LiDAR mount	17.5 mm printed spacer, no standoffs, scan plane 222.05 mm (register, platform-sw, BOM Hardware BOM sheet MEC-008 after the 24 Oct correction)	P-03 item 13, platform-hw lidar.md and docs index.md: standoffs plus 7.1 mm plate; BOM Mechanical sheet: aluminium spacer plate; BOM Connections sheet: 222.0 mm	P-03 item 13, BOM Mechanical and Connections sheets, platform-hw lidar.md, docs index.md
7	Base camera pose	base_camera_link (0.387, 0, 0.046), 10 deg up (P-03 rev19.7 item 19 D3; platform-sw main)	docs index.md: (0.376, 0, 0.0445) at 142.5 mm	docs index.md (docs owner)
8	Contactors	2 x GX11CAC (register; E-01 rev1.4; BOM Issue 7.10 SAF-003)	P-03 item 5, platform-fw #17, platform-fw #18 and platform-hw #27 still LEV100H	P-03 item 5 and the three PRs or issues
9	Near-field sensor set	4 x MB7060 (UART) and 2 x TF-Luna (UART) (P-03 rev19.7 item 19; BOM Issue 7.10 SNS-001, SNS-002)	platform-fw #11 title and platform-hw #15 body: TF-Luna and MB7040 over I2C	Retitle platform-fw #11 and update platform-hw #15
10	Navigation contract version	CONTRACT_VERSION 1 on interfaces main (d048fb8)	interfaces #28 bumps it to 2; repositories pin three different interfaces commits (87ab8a9, fa7c438, d048fb8)	Merge or close interfaces #28, then one pin across repositories (release owner)
11	/estop contract	Contract changes go through a contract change request (GOVERNANCE.md)	platform-fw #17 and platform-sw #66 require an /estop contract change request; none exists	Open the /estop contract change request (software lead, electrical owner)
12	rc.1 pin count	One pinned source set for v2.0.0-rc.1 (register RELEASE-MILESTONES)	release #12 pins 14 sources; manifest #17 pins 11	Release owner to agree one list
13	Arm mount	Arm profile on a 10 mm adapter plate on the column top (register; BOM Issue 7.10 MEC-003, MEC-004)	ADR-010: profile overlaps the column; upperbody-sw #14 model: profile in front of the column	ADR-010 and upperbody-sw #14 to follow the register
14	Motion chain	collision_monitor is the only /cmd_vel publisher (P-03 rev19.7; register)	on platform-sw main the UI and docking publish /cmd_vel directly; twist_mux and the UI relay are only in platform-sw #90	platform-sw #90 and #35
15	Hub motor	ZLLG80ASM250-L-B, 120 kg per pair (BOM Issue 7.10 DRV-001; P-03 rev19.5 item 21, supplier answer 18 Oct)	platform-hw datasheets README drops the -B suffix; platform-hw #28 uses -B2; the drawing says 100 kg per pair	platform-hw datasheets README and #28
16	Charger	WCM-500-24 on release 001 and 2.0 (P-03 rev19.7 item 23; register SHARED-BASE and CHARGING; BOM Issue 7.10)	docs concepts.md names the WCM-300	docs concepts.md (docs owner)
17	Raspberry Pi 5 label	Jetson Orin NX 16 GB is the 2.0 compute; the Raspberry Pi 5 is release 001 only	The Pi 5 is named without a release 001 or legacy label in platform-hw, platform-sw, docs and the release README	Label each mention release 001 or legacy
18	Old revisions cited as current	P-03 rev19.7 and BOM Issue 7.10	rev19.0 and BOM 7.7 in docs hardware-architecture, platform-hw product-architecture and BOM files, platform-fw base-controller-2-0-io, platform-sw sensor_frames.yaml, upperbody-hw README and manipulation; rev18.9 in upperbody-sw; the release README title still says v0.0.1 above the 2.0 plan	Each repository owner; release README title (release owner)
19	.github ROADMAP.md and org profile	Two 7-DoF OpenArm 2.0 arms and a 2D RPLIDAR S3 (register; org profile); brake rating is an open supplier gate (BOM Issue 7.10)	ROADMAP.md says dual 6-DoF arms and a 3D LiDAR; the org profile says fail-safe holding brakes	.github ROADMAP.md and profile (platform lead)
20	Governance	GOVERNANCE.md: platform lead reviews safety paths; two approvals on safety changes; maintainers.yaml lists the roles	Code owners skunal3318, AnushkaAMuragod, Brinex45 and SatyaA27 are not in maintainers.yaml or MAINTAINERS.md; the platform lead is not a code owner in several repositories (platform-sw, interfaces, manifest, manipulation, upperbody-sw, embodied-ai, docs); rulesets enforce one approval; embodied-ai is missing from maintainers.yaml and the Watchdog list; TECHNICAL_STEERING.md is empty	.github maintainers.yaml, MAINTAINERS.md, rulesets and Watchdog list (platform lead, CI owner)
21	BOM owners	Team document: @Evans-Kwashie owns power and safety, @AnushkaAMuragod supports	BOM Issue 7.10 power and safety rows still list the supporting contributor as owner; mechanical rows list an inactive contributor	BOM owner columns (platform lead)
22	Docs site	The deploy job verifies the live custom domain	Step "Verify the live custom domain" fails with exit 127 since at least 11 Oct because the deploy job has no checkout; draft docs #49 fixes it. Drafts docs #46 and #47 cite older revisions (P-03 rev19.2 and rev19.4) than this roadmap	docs #49; refresh #46 and #47 to P-03 rev19.7

Source: P-03 rev19.7, E-01 rev1.4, BOM Issue 7.10 (corrected 24 Oct), P-02 rev19.2, decisions.yaml at 1c3f516, and main and open PRs of each repository at the heads in the footer · 24 Oct 2026

Data refresh, 24 Oct 2026

Every repository was re-read on 24 Oct 2026 (heads of main in the footer, open and merged PRs since 19 Oct, issues with their due dates from the issue body; no repository uses GitHub milestones). Review, CI and sign-off states are as recorded on 24 Oct.

Documents in force. P-03 rev19.7 (20 Oct) is the document in force by project-lead decision of 24 Oct: ToF pitch 10 deg, base_link 0.098 m above the floor, item 23 (one shared base for release 001 and 2.0, WCM-500-24 on both, embodied-AI repository layout, platform-sw pin dfd44a5). E-01 body rev1.4 (file title rev1.3), BOM Issue 7.10 (20 Oct, corrected 24 Oct: camera, IMU, LiDAR and ToF values aligned with P-03 rev19.7), P-02 rev19.2 (20 Oct).
Effect: Every design value on these pages now cites P-03 rev19.7; plan dates cite P-02 rev19.2; Gate A and the STM32 go/no-go are marked as P-02 dates.
Register. .github #55 (1c3f516, 22 Oct) records the rev19.7 decisions (SHARED-BASE, WCM-500-24 on both robots, ToF 10 deg); the document label still says rev19.6. .github main is 68833f2 (#61, 24 Oct). FUNDING added (#56).
Effect: Action plan step 4 is MERGED AND TESTED; the shared base is no longer pending a register entry. Label mismatch is drift 1.
platform-sw (main 8a7a5b1, unchanged since 19 Oct). New PRs: #90 twist_mux and stop gate (stop lock priority 255, UI TwistStamped relay on /ui/cmd_vel), changes requested 23 Oct, sign-off missing; #91 recovery behaviour tree, changes requested (retry counter stays NOT_READY, one failing test); #92 approved once on 24 Oct, needs sign-off and a second approval; #93 draft, ToF frames to 10 deg. Audit pass 2 issues #74 to #88 are open with no comments; #35 is due 30 Oct; #74 (simulation body filter) is due 6 Nov with no PR.
Effect: Steps 10 and 25 move to IN REVIEW; the reactive motion layer block moves to IN REVIEW. Implementation on main is unchanged: no twist_mux, and the UI and docking still publish /cmd_vel.
platform-fw (main f96014d, unchanged since 12 Oct). CubeMX project #22 is ready for review with change requests as comments (Ethernet DMA in DTCM, HAL edit, unused UART TX pins); CI PRs #23, #27, #28 and #29 are drafts. Issues #20 (lift interlock) and #21 (micro-ROS transport) are due 13 Nov; #17 (safety state machine) is not started. pid.h is missing, so Gate A cannot be rebuilt from the repository; boards/stm32 is empty.
Effect: The 2.0 base controller block moves from NOT STARTED to IN REVIEW (platform-fw #22).
platform-hw (main eaac115, 24 Oct). Hardware repository checks merged (#34). PRs #27 and #29 lack sign-off; #27 and the near-field docs still carry LEV100H and 20 deg; agent rules draft #35.
Effect: Drift 2, 8 and 15.
openamrobot-ui (main 0ca798b, 22 Oct). Only the Stop (software) rename landed (#38). The stop lock and /ui/cmd_vel are not on main. Hardening issues #31 to #37 are open.
Effect: UI hardening moves to IN PROGRESS (security hardening in progress).
interfaces (main d048fb8, 22 Oct). Shared agent rules adopted (#26). #28 (SiteConfigStatus, CONTRACT_VERSION 2) is approved and blocked only on sign-off. C9 schema: 26/26 metadata checks pass; acceptance issue #18 open.
Effect: Contract version on main stays 1 (drift 10).
Upper body and manipulation. upperbody-sw #14 is a 405-file draft with changes requested (NaN joint samples accepted; quality check to restore); issues #17 (lift as planned joint) and #18 (joint clamp default) are due 13 Nov. manipulation #20 is ready, 99/99 tests, one controller-config defect left, sign-off missing; #24 (base and arm exclusion) is due 13 Nov with no PR.
Effect: Statuses unchanged (IN REVIEW); dates added from issue bodies.
Embodied AI (main 56bbd36, 22 Oct). Scaffold on main: C10 dataset QA validator with tests, registry, recorder and policy-runner skeletons. No capture session and no trained policy.
Effect: Learning baseline stays NOT STARTED.
Release and manifest. release #8 and #9 merged 24 Oct (main c39eda7); release #12 (rc.1 from pinned SHAs) is a draft and pins 14 sources; manifest #17 pins 11; manifest #13 is approved twice and not merged; #11 is stale.
Effect: Pin-count mismatch is drift 12.
Docs. Deploy step "Verify the live custom domain" fails with exit 127 since at least 11 Oct (no checkout in the deploy job); draft docs #49 fixes it. Drafts #46 and #47 cite older revisions.
Effect: Drift 22.

Action plan status changes since 19 Oct

Step	19 Oct	24 Oct	Source
4	NOT STARTED	MERGED AND TESTED	.github #55 (1c3f516, 22 Oct) records SHARED-BASE
10	IN PROGRESS	IN REVIEW	platform-sw #90 opened (UI TwistStamped /ui/cmd_vel relay)
14	NOT STARTED	IN PROGRESS	ui issues #31 to #37 opened (security hardening in progress)
25	NOT STARTED	IN REVIEW	platform-sw #90 opened (stop lock)

Earlier refresh notes (19 Oct 2026)

Three draft PRs opened by @KARTHIKEYAN124 (release lead): release #12 (rc.1 built from pinned commit SHAs with verified checksums, OA-05), manifest #17 (SHA-pinned rc.1 release manifest, OA-39) and manifest #18 (pinned Jazzy driver sources and installation checks, OA-06). Effect: OA-05, OA-39 and OA-06 have fix work IN REVIEW (draft); the release item now lists these PRs; release owner @KARTHIKEYAN124 (release lead) (Team document: release and installation lead). openamrobot-release/pull/12 openamrobot-manifest/pull/17 openamrobot-manifest/pull/18
platform-sw #54 (behaviour server routed through the reactive motion layer) merged on 18 Oct after two change requests; it is the platform-sw main head 8a7a5b1. Effect: Already in the first snapshot; confirmed. openamr-platform-sw/pull/54
SiteConfigStatus: interfaces issue #27 and PR #28 (open, not draft) add SiteConfigStatus to NavigationStatus and bump CONTRACT_VERSION from 1 to 2; platform-sw #73 (open, not draft) produces it. Effect: Shown IN REVIEW in the interfaces and contracts map; contract version on main stays 1. openamrobot-interfaces/issues/27 openamrobot-interfaces/pull/28 openamr-platform-sw/pull/73
platform-fw #19 (2.0 controller I/O docs, open; @Evans-Kwashie (electrical and firmware lead) requested as reviewer), platform-fw issue #13 (pin the linorobot2 base commit for v0.0.1 and Gate A, open), platform-hw #30 (datasheets for the 2.0 pin map, open; one review comment from @Evans-Kwashie (electrical and firmware lead)). Effect: Statuses unchanged: IN REVIEW for #19 and #30; issue #13 open. openamr-platform-fw/pull/19 openamr-platform-fw/issues/13 openamr-platform-hw/pull/30
upperbody-sw #14 is still a draft at e5cf613 with two change requests from @panthera-momagdii (tech lead, UI reviewer). Effect: PR #14 stays IN REVIEW, not merged, not accepted. openamr-upperbody-sw/pull/14
Issue #35 (route all velocity sources through arbitration) is open; platform-fw issue #17 (E-01 safety state machine, Teensy 4.0 first, then STM32) is open. Effect: Arbitration stays IN PROGRESS (branch); the safety state machine is NOT STARTED in code (design draft only). openamr-platform-sw/issues/35 openamr-platform-fw/issues/17
Counts: 96 PRs merged 21 Sep to 19 Oct in 14 repositories (unchanged; two independent counts agree); 30 open PRs (was 27); 79 open issues in the 14 repositories (the first snapshot's 80 included one issue in a repository outside the 14). Effect: Merged-PR figure unchanged; open counts updated.
20 Oct: the charging receiver is decided by @BotshareAI (project lead): the TZBOT WCM-500-24 is used on both robots; the WCM-300 is only the unit on release 001 today. Review of openamrobot-docs #48 by @panthera-momagdii (tech lead, UI reviewer): reactive motion layer terminology and the parameter-check classification applied. Effect: Charging leaves the open shared-base differences; Sheet A is to be updated. S12 now counts 7 contradictions and 6 observations. openamrobot-docs/pull/48
Main heads unchanged since the first snapshot (newest: .github 9109d3e, 19 Oct). Effect: No code on main changed; implementation statuses stand.

Progress in pictures

Area by stage

What this shows: how far each area has come through design, merged code, simulation test, bench and physical acceptance; the lower rows are the shared base, with whether release 001 evidence will carry over to 2.0.

How to read it: green check: reached; orange half circle: partly; red bar: fails; white dashed circle: not reached; grey: not applicable. No percentages are used.

Scroll sideways inside the diagram.

Source: per-area deliverables and runtime layers on this page; equivalence table on the hardware page · 24 Oct 2026

rc.1 blocker board

What this shows: every rc.1 dependency grouped by proposed batch and owner, with counts.

How to read it: each box has a coloured left bar, a status chip with an icon and the status in words; batch dates are proposed targets from audit pass 2; items with no batch date have no date in any source.

19 items

Source: audit pass 2 top 20 actions and the rc.1 table on this page · 24 Oct 2026

Requirement coverage by area

What this shows: the 85 traceability rows of each area as one stacked bar, split by coverage class.

How to read it: each segment is one coverage class with its count; icons and the legend under the bars name the classes; the number after the bar is the area total.

Source: audit pass 2 traceability (85 rows) · 24 Oct 2026

Merged PR activity since 21 Sep

What this shows: how many pull requests were merged per week in each repository.

How to read it: darker cells mean more merged PRs; the number in each cell is the count; a merged PR is not a tested product change.

Scroll sideways inside the diagram.

Source: GitHub search and per-repository PR lists, 19 Oct 2026 · 24 Oct 2026

Risk matrix

What this shows: the main risks to rc.1 and v2.0.0 placed by likelihood and impact.

How to read it: top right is worst; each dot is a risk ID listed below the matrix; likelihood and impact are this review's assessment (opinion), the evidence is linked.

Source: this review's assessment, based on the linked evidence · 24 Oct 2026

R1: Command bypasses of the reactive motion layer likelihood high, impact high. Evidence: S3, S4, S5; OA-02-P2, OA-45-P2, OA-48. details
R2: Docking not working through the chain likelihood high, impact medium. Evidence: S3: 0 of 5 docked. details
R3: Main not navigating in simulation likelihood high, impact high. Evidence: S1: 0 of 3 goals on main (OA-47). details
R4: Contactor procurement (BUY-32) likelihood high, impact high. Evidence: BUY-32 blocked; out of stock; about 16 weeks lead time for other codes. details
R5: Base controller decision for rc.1 likelihood medium, impact high. Evidence: No approved decision; go/no-go 6 Nov. details
R6: Intermittent bridge crash in PR #14 likelihood medium, impact medium. Evidence: 2 of 9 A7 runs failed on 18 Oct; cause not investigated. details
R7: Little physical test time before integration likelihood high, impact high. Evidence: Nothing built for 2.0; window 23 Nov to 18 Dec. details
R8: Shared base not yet built likelihood high, impact high. Evidence: Design draft only; BUY-32 blocks K1/K2. details

Per area

Delivery status, runtime evidence (with the tested branch or commit) and physical acceptance are separate columns. Physical acceptance is none recorded everywhere.

Base and navigation owner @skunal3318 (navigation lead)

Nav2 output passes through the smoother and collision monitor; reactive stopping has not been demonstrated (S2). #54 routed recoveries through the chain. The simulation on main cannot navigate, the reactive layer covers only Nav2, and there is no 2.0 real-robot bringup.

Deliverable	Delivery	Runtime evidence (tested setup)	Physical	Findings · closes when
Nav2 chain (controller, smoother, collision monitor)	MERGED UNTESTED	runtime: fails in simulation main 8a7a5b1	physical: none recorded	OA-47 OA-35-P2 Rerun S1 and S2 on main
Recoveries through the chain (#54)	MERGED AND TESTED	runtime: works in simulation with fixture main 8a7a5b1 + fixture	physical: none recorded	-
Navigation status producer	MERGED AND TESTED	runtime: not run unit tests on main (113)	physical: none recorded	OA-58 Status flags a non-monitor /cmd_vel publisher at runtime
Motion-source arbitration (#35)	IN PROGRESS	runtime: works on branch or prototype review/issue35-arbitration c25ceef	physical: none recorded	OA-04-P2 OA-02-P2 S3, S4 and S7 reruns with a single protected /cmd_vel publisher
2.0 real-robot bringup	NOT STARTED	runtime: not run -	physical: none recorded	OA-17 Launch test: TF single-rooted on the 2.0 profile

Merged work

openamr-platform-sw #54 Behavior server routed through the smoother and collision monitor, with a routing test (#35, part) (merged 2026-10-18)
openamr-platform-sw #39 Navigation status producer node (merged 2026-10-02)
openamr-platform-sw #47 Navigation stack failed state after a timeout (merged 2026-10-08)
openamr-platform-sw #52 Navigation status mirrored on /diagnostics (merged 2026-10-09)
openamr-platform-sw #55 REP 105 base frames and 2.0 sensor frames (merged 2026-10-12)
openamr-platform-sw #70 2.0 sensor frames, simplified collision and Gazebo sensors (merged 2026-10-16)
openamr-platform-sw #71 Chassis collision box clears the caster wheels (merged 2026-10-16)
openamrobot-interfaces #5 Navigation health contract messages (merged 2026-09-23)
openamrobot-interfaces #22 Reason code MOTION_SOURCE_PARTIALLY_PROTECTED (merged 2026-10-07)
openamrobot-interfaces #23 Reason code COLLISION_APPROACH_ACTIVE (merged 2026-10-09)

Open work

openamr-platform-sw #73 SiteConfigStatus producer (open)
openamrobot-interfaces #28 SiteConfigStatus contract (open)

Runtime evidence

runtime: fails in simulation S1 main 8a7a5b1
runtime: works in simulation with fixture S1 main + F1 to F3
runtime: works in simulation with fixture S2 no contact; monitor never engaged
runtime: works in simulation with fixture S6 stop 0.79 s for a Nav2 goal; no latch

Simulation at about 0.1 real-time factor; not safety evidence.

Requirement coverage

partially implemented: UC1-02 UC1-10 UC1-42 UC1-90 UC1-93 P03-13 P03-15 P03-19
contradicted: UC1-07
missing: UC1-41 UC1-91

Open blockers

OA-47 · scenario S1 · code openamr-platform-sw/ros2/src/openamrobot_nav2/config/scan_body_filter.yaml:1 · owner @skunal3318 (navigation lead) · closes when: Rerun S1 on main: 3 of 3 goals; a test fails when the mask covers the forward sectorOA-56 · scenario S7 · code openamr-platform-sw/ros2/src/openamrobot_nav2/launch/navigation_launch.py · owner @skunal3318 (navigation lead) · closes when: Rerun S7b: after killing controller_server the monitor stays active and zero velocity holdsOA-57 · scenario S1, S11 · code openamr-platform-sw/ros2/src/openamrobot_bringup/launch/bringup.launch.py · owner @skunal3318 (navigation lead) · closes when: S11 extended to 10 starts: all reach active within 120 s sim timeOA-35-P2 · scenario S2 · code openamr-platform-sw/ros2/src/openamrobot_nav2/config/nav2_params.yaml:177 · owner @skunal3318 (navigation lead) · closes when: Rerun S2: clearance at least 0.25 m; collision monitor engages in the barrier testNR-03 · owner @skunal3318 (navigation lead) · closes when: After RESET, no motion from any source until an explicit operator release; verified in simulation and on the bench

Docking owner @fridayali (docking and robot description)

Tag detection is upstream apriltag_ros; the dock_trigger.py sequencer is custom, has no tests and no cancel. DockingStatus is issue #33 (no PR). No Use Case 1 site-marker alignment.

Deliverable	Delivery	Runtime evidence (tested setup)	Physical	Findings · closes when
dock_trigger sequencer	MERGED UNTESTED	runtime: fails in simulation main + fixture; prototype	physical: none recorded	OA-04-P2 OA-53 OA-54 OA-68 Rerun S3: 5 of 5 docked
Docking retuned for Gemini 336L	IN REVIEW	runtime: not run PR #53 draft	physical: none recorded	OA-59 S10 and S3 pass on base_camera_link
DockingStatus (issue #33)	NOT STARTED	runtime: not run -	physical: none recorded	Status at 2 Hz plus on change; docked and charging separate
Site ArUco alignment (Use Case 1)	NOT STARTED	runtime: not run -	physical: none recorded	OA-46 Align on site markers in simulation to +/-1 cm and +/-1 deg

Merged work

No product PR merged in this area since 21 Sep.

Open work

openamr-platform-sw #53 Gemini 336L base camera simulation (draft)

Runtime evidence

runtime: fails in simulation S3 main, fixture
runtime: works on branch or prototype S3 #35 c25ceef: single publisher; 0 of 2 docked
runtime: fails in simulation S7 (c) drove into the dock after dock_trigger was killed

Simulation at about 0.1 real-time factor; not safety evidence.

Requirement coverage

contradicted: UC1-08
missing: UC1-47

Open blockers

Safety chain owner to be confirmed

Design is final in E-01 rev1.4 and the register (PNOZ s3, two E-stops, GX11CAC K1/K2, single RESET, charge inhibit). Contactor order BUY-32 is blocked on datasheet checks and supply. No repository on main carries the E-01 chain; drafts are open. No bench or physical record.

Deliverable	Delivery	Runtime evidence (tested setup)	Physical	Findings · closes when
E-01 rev1.4 safety chain design	DOCUMENTED ONLY	runtime: not run -	physical: none recorded	OA-23 E-01 chain on main of platform-hw with two human approvals
K1/K2 GX11CAC procurement (BUY-32)	BLOCKED	runtime: not run -	physical: none recorded	Datasheet checks pass and a firm quote exists
Supervised safety-chain bench test	NOT STARTED	runtime: not run -	physical: none recorded	OA-38 Bench record: each E-stop cuts the SAFE bus; reset with E-stop pressed does nothing

Merged work

.github #52 Register brought to P-03 rev19.6, E-01 rev1.3 and BOM Issue 7.10 (GX11CAC, item 22 I/O) (merged 2026-10-19)
.github #48 Register records P-03 item 20 (E-stop scope, single RESET, wireless charging) (merged 2026-10-16)

Open work

openamr-platform-hw #27 E-01 safety chain and power distribution baseline (draft)
openamr-platform-fw #18 Hardware chain is the primary stop (E-01) (draft)
openamr-platform-sw #66 Safety docs point to the E-01 chain (draft)
openamrobot-docs #46 E-01 safety and power page (draft)

Runtime evidence

runtime: not run - hardware chain; nothing to simulate

Requirement coverage

needs hardware, not checkable in simulation: UC1-03 UC1-04 UC1-14 UC1-70 UC1-71 UC1-72 UC1-73 UC1-75
out of scope this pass: P03-20 P03-21 P03-22

Open blockers

OA-23 · code openamr-platform-hw (main) · owner to be confirmed · closes when: Register check passes on each repository carrying the E-01 rev1.4 chain; two human approvalsOA-38 · owner to be confirmed · closes when: Evidence record location and procedure exist; supervised bench record attached

Operator UI owner @parthkeshari (UI lead); reviewers @panthera-momagdii (tech lead, UI reviewer) and UI reviewer (GitHub handle to be confirmed); simulation checks @nikgolobo (simulation checks)

The UI works as an operator tool, but its teleop, blocks, Maps page and replay are motion sources outside the platform chain. Web tests: 0. Reporting Mode: no code yet.

Deliverable	Delivery	Runtime evidence (tested setup)	Physical	Findings · closes when
Safe UI command path	IN PROGRESS	runtime: fails in simulation UI c5558ad5 + platform fixture	physical: none recorded	OA-02-P2 OA-03-P2 OA-48 OA-45-P2 S4, S5, S7a reruns pass
Stop (software) with lock	NOT STARTED	runtime: works in simulation with fixture S6 (Nav2 path only)	physical: none recorded	OA-16-P2 S6 plus: no motion from any source until release
UI access control and command path hardening	NOT STARTED	runtime: not run -	physical: none recorded	OA-49 OA-50 Access-control tests pass
Web tests	NOT STARTED	runtime: not run 0 tests	physical: none recorded	OA-24 More than 0 web tests run in CI
Reporting Mode	DOCUMENTED ONLY	runtime: not run -	physical: none recorded	Full-screen reporting session recorded

Merged work

openamrobot-ui #25 UI loads the robot description from platform-sw (merged 2026-10-12)

Open work

openamrobot-ui #21 Fixture-only run-state adapter (Proposed) (draft)

Runtime evidence

runtime: fails in simulation S4 UI c5558ad5
runtime: fails in simulation S5 replay and second controller
runtime: works in simulation with fixture S6 stop for a Nav2 goal
runtime: fails in simulation S7 (a) 2.25 m after UI loss

Simulation at about 0.1 real-time factor; not safety evidence.

Requirement coverage

contradicted: UC1-07
implemented, untested: UC1-54

Open blockers

OA-02-P2 · scenario S4 · code openamrobot-ui/web/src/shared/constants/index.js:7 · owner @parthkeshari (UI lead); reviewers @panthera-momagdii (tech lead, UI reviewer) and UI reviewer (GitHub handle to be confirmed); simulation checks @nikgolobo (simulation checks) · closes when: Rerun S4 and S7a: rosbridge is not a /cmd_vel publisher; robot stops before the wall; stops within 0.5 s after the UI connection dropsOA-03-P2 · scenario S5 · owner @parthkeshari (UI lead); reviewers @panthera-momagdii (tech lead, UI reviewer) and UI reviewer (GitHub handle to be confirmed); simulation checks @nikgolobo (simulation checks) · closes when: Rerun S5: build_map leaves one controller_server; platform map_server and AMCL states unchangedOA-48 · scenario S5 · code location withheld (security) · owner @parthkeshari (UI lead); reviewers @panthera-momagdii (tech lead, UI reviewer) and UI reviewer (GitHub handle to be confirmed); simulation checks @nikgolobo (simulation checks) · closes when: Replay of a recording that contains command topics: robot travel 0OA-49 · scenario S4, S5 · code location withheld (security) · owner @parthkeshari (UI lead); reviewers @panthera-momagdii (tech lead, UI reviewer) and UI reviewer (GitHub handle to be confirmed); simulation checks @nikgolobo (simulation checks) · closes when: Security review of the UI backend passesOA-50 · scenario S5 · code location withheld (security) · owner @parthkeshari (UI lead); reviewers @panthera-momagdii (tech lead, UI reviewer) and UI reviewer (GitHub handle to be confirmed); simulation checks @nikgolobo (simulation checks) · closes when: Security review of the UI backend passesOA-16-P2 · scenario S6 · code openamrobot-ui/web/src/components/StatusBar.jsx:63 · owner @parthkeshari (UI lead); reviewers @panthera-momagdii (tech lead, UI reviewer) and UI reviewer (GitHub handle to be confirmed); simulation checks @nikgolobo (simulation checks) · closes when: Rerun S6: after stop, UI teleop and docking produce no motion until release; button not labelled E-STOP

Upper body and manipulation owner @Tushargoel230 (upper-body model); other owners to be confirmed

PR #14 is a draft: IN REVIEW, not merged, not accepted. Head e5cf613 adds evidence and docs on top of code fe2e821; the PR description still names 49d2368. Remaining WARN: M07 (four upstream joint limits) and V01 (no human visual check). Risk: intermittent bridge crash on shutdown, cause not investigated.

PR #14 evidence (openamr-upperbody-sw) IN REVIEW

Draft: IN REVIEW, not merged, not accepted. Head e5cf613 (19 Oct 2026). The PR description still names 49d2368.

Latest batch on upper-body code fe2e821 with platform-sw dfd44a5: A6 25 PASS / 3 WARN / 0 FAIL and A7 14 PASS / 2 WARN / 0 FAIL, 3 of 3 runs each.

Cumulative on platform-sw dfd44a5: A7 9 runs, 2 failed (18 Oct: parameter_bridge base_joint_states_bridge SIGSEGV during stop, cause not investigated per the test index), the last 6 passed.

Risk: Intermittent bridge crash, cause not investigated. Remaining WARN: M07 (four upstream joint limits), V01 (no human visual check).

Separate from audit pass 2 S8 (48 of 48 plans on mock hardware, branch followups/a6-a7-49d2368). Owner @Tushargoel230 (upper-body model). Sources: PR #14, branch test index. Simulation at about 0.1 real-time factor; not safety evidence.

Deliverable	Delivery	Runtime evidence (tested setup)	Physical	Findings · closes when
Combined base, lift and arm model with MoveIt (PR #14)	IN REVIEW	runtime: works on branch or prototype PR #14 head e5cf613, code fe2e821, platform-sw dfd44a5 (see PR #14 evidence)	physical: none recorded	OA-12 A6 and A7 green on the exact head with M07 and V01 closed, bridge crash investigated; merged
Combined model with MoveIt on mock hardware (audit pass 2 S8, separate from PR #14 evidence)	IN REVIEW	runtime: works on mock hardware followups/a6-a7-49d2368 a490503	physical: none recorded	OA-64 Rerun S8 with the lift in the planning group: 48 of 48 and every margin above 0.05 rad
Lift as a planned joint	NOT STARTED	runtime: not run -	physical: none recorded	OA-64 OA-13 S8 with the lift in the group
Base-arm and lift interlocks	NOT STARTED	runtime: not run -	physical: none recorded	OA-37 Integration test both directions
A5/I2 device package (PR #20)	IN REVIEW	runtime: not run draft; hosted tests skipped	physical: none recorded	Schema and conformance tests run on the PR head

Merged work

openamr-upperbody-sw #11 Option B upper-body model (merged 2026-10-12)
openamr-upperbody-sw #12 Column axis at x +0.060 (merged 2026-10-12)
openamrobot-manipulation #9 OpenArm 2.0 fake-hardware baseline (development only) (merged 2026-09-30)
openamr-upperbody-hw #10 Option B upper-body mounting documentation (merged 2026-10-12)

Open work

openamr-upperbody-sw #14 Combined option B model and MoveIt configuration (draft)
openamrobot-manipulation #20 A5/I2 device package schemas and validator (draft)
openamrobot-manipulation #18 SO-101 capability discovery sketch (draft)

Runtime evidence

runtime: works on mock hardware S8 audit pass 2: followups/a6-a7-49d2368 a490503, 48 of 48 plans
runtime: works on branch or prototype A6/A7 PR #14 head e5cf613, upper-body code fe2e821 with platform-sw dfd44a5: latest batch A6 25/3/0 and A7 14/2/0 (3 of 3 each); cumulative A7 on dfd44a5 9 runs, 2 failed

Simulation at about 0.1 real-time factor; not safety evidence.

Requirement coverage

partially implemented: UC1-06 UC1-20 UC1-46 P03-16
implemented, untested: UC1-12
implemented and tested in simulation: UC1-13 UC1-15 P03-17 P03-18
missing: UC1-19 UC1-40

Open blockers

OA-64 · scenario S8 · code PR #14 (branch evidence) · owner @Tushargoel230 (upper-body model) · closes when: Rerun S8 with the lift in the planning group: 48 of 48 and every margin above 0.05 radOA-37 · code openamr-platform-sw (main) · owner to be confirmed · closes when: Integration test both directions: arm goal rejected while base moves; base goal rejected while arm movesOA-40 · code openamrobot-manipulation/integration/README.md · owner to be confirmed · closes when: Planning test passes without strict xfail at an approved acceleration limit set

Firmware and base controller owner @Evans-Kwashie (electrical and firmware lead)

Main has the experimental Teensy Gate A overlay with a 200 ms /cmd_vel watchdog. The STM32H723ZG port is not on main; the I/O allocation is documented and in review (#19).

Deliverable	Delivery	Runtime evidence (tested setup)	Physical	Findings · closes when
Teensy Gate A overlay (200 ms watchdog)	MERGED UNTESTED	runtime: not run -	physical: none recorded	Gate A host-migration suite passes on 30 Oct
STM32H723ZG port and I8 Gate B	NOT STARTED	runtime: not run -	physical: none recorded	OA-14 I8 Gate B evidence on the bench
2.0 controller I/O documentation (#19)	IN REVIEW	runtime: not run -	physical: none recorded	OA-19 Merged with the 16 Oct IMU pose

Merged work

openamr-platform-fw #14 Gate A labels and the decided 2.0 base controller I/O (documentation) (merged 2026-10-12)
openamr-upperbody-fw #11 Lift interface on CAN3 (documentation) (merged 2026-10-12)

Open work

openamr-platform-fw #19 2.0 controller I/O aligned with P-03 item 22 (documentation) (open)
openamr-platform-hw #30 Datasheet references for the 2.0 pin map (open)

Runtime evidence

runtime: not run - no firmware runtime evidence; Teensy watchdog is a code statement

Requirement coverage

contradicted: UC1-82
out of scope this pass: P03-14 P03-22

Open blockers

OA-14 · code openamr-platform-fw/boards/stm32/.gitkeep · owner @Evans-Kwashie (electrical and firmware lead) · closes when: I8 Gate B evidence on the NUCLEO-H723ZG bench (owner implements; agents do not write safety logic)

Simulation owner @nikgolobo (simulation checks)

Gazebo Harmonic runs the base and the combined model. The simulated model is rated NEEDS WORK: body filter, legacy docking camera, LiDAR range_min 0.40 m, no command timeout, near-field sensors without consumers.

Deliverable	Delivery	Runtime evidence (tested setup)	Physical	Findings · closes when
Simulated 2.0 robot model	MERGED UNTESTED	runtime: fails in simulation main 8a7a5b1	physical: none recorded	OA-47 OA-59 OA-69 OA-45-P2 S1, S7, S10 reruns pass on main

Merged work

openamr-platform-sw #70 Gazebo near-field sensors and IMU in simulation (merged 2026-10-16)

Open work

openamr-platform-sw #53 Gemini 336L RGB-D simulation (draft)

Runtime evidence

runtime: works in simulation with fixture S9 REP 105 tree
runtime: fails in simulation S10 camera on legacy frame; near-field unused
runtime: works in simulation with fixture S11 start and stop cycles
runtime: fails in simulation S7 simulated base keeps the last command

Simulation at about 0.1 real-time factor; not safety evidence.

Requirement coverage

partially implemented: UC1-52

Open blockers

OA-47 · scenario S1 · code openamr-platform-sw/ros2/src/openamrobot_nav2/config/scan_body_filter.yaml:1 · owner @skunal3318 (navigation lead) · closes when: Rerun S1 on main: 3 of 3 goals; a test fails when the mask covers the forward sectorOA-45-P2 · scenario S7 · code openamr-platform-sw/ros2/src/openamrobot_description/urdf/gazebo_control.xacro:5 · owner @skunal3318 (navigation lead) · closes when: Rerun S7a and S7c: simulated base stops within the command timeout after publisher lossOA-59 · scenario S10 · code openamr-platform-sw/ros2/src/openamrobot_description/urdf/robo_urdf.urdf.xacro:696 · owner @fridayali (docking and robot description) · closes when: Rerun S10: /rgb_image frame base_camera_link; S3 passes with itOA-69 · scenario S10, S12 · code openamr-platform-sw/ros2/src/openamrobot_nav2/config/nav2_params.yaml · owner @skunal3318 (navigation lead) · closes when: Collision monitor and local costmap list the ToF and ultrasonic sources, or the register records their role

Interfaces owner to be confirmed

Navigation health contract merged and tested. I1 (manipulation), I8 (base status) and DockingStatus contracts are not in the repository.

Deliverable	Delivery	Runtime evidence (tested setup)	Physical	Findings · closes when
Navigation health contract	MERGED AND TESTED	runtime: works in simulation on main interfaces 87ab8a9 CI (interface tests)	physical: none recorded	-
I1 manipulation and I8 base-status contracts	DOCUMENTED ONLY	runtime: not run -	physical: none recorded	OA-15 Proposed contracts merged with consumers listed

Merged work

openamrobot-interfaces #9 ROS 2 Jazzy interface build gate (merged 2026-09-23)
openamrobot-interfaces #11 Navigation message exchange test with a reverted-interface failure (merged 2026-09-25)
openamrobot-interfaces #15 Exchange test on Fast DDS and Cyclone DDS (merged 2026-09-28)

Open work

openamrobot-interfaces #28 SiteConfigStatus (open)

Runtime evidence

runtime: works in simulation on main - exchange tests in CI on Fast DDS and Cyclone DDS (interface test, not robot runtime)

Simulation at about 0.1 real-time factor; not safety evidence.

Requirement coverage

missing: UC1-43

Open blockers

OA-15 · code openamrobot-interfaces (main) · owner to be confirmed · closes when: I1 and I8 Proposed contracts build with consumers listedOA-08 · code openamrobot-ui/ros2/src/openamr_ui_msgs/package.xml:3 · owner @parthkeshari (UI lead); reviewers @panthera-momagdii (tech lead, UI reviewer) and UI reviewer (GitHub handle to be confirmed); simulation checks @nikgolobo (simulation checks) · closes when: colcon build at the workspace root exits 0

Embodied AI owner to be confirmed

Interfaces and the C9 capture schema exist (merged 2 Oct, still Proposed). No trained policy, validated dataset or imitation-learning task is demonstrated.

Exists

C9 whole-body capture schema (interfaces PR #16, merged 2 Oct; contract Proposed)
Capture station items in the BOM (offboard)

Not demonstrated

Recorder and calibrated capture session
Dataset QA (C10 draft only)
Imitation-learning baseline (C11, due 6 Nov)
One timebase across streams (UC1-61)

Deliverable	Delivery	Runtime evidence (tested setup)	Physical	Findings · closes when
C9 whole-body capture schema	MERGED AND TESTED	runtime: not run schema tests in interfaces CI	physical: none recorded	Contract accepted by the owner; a capture session recorded
C10 dataset quality gate	IN REVIEW	runtime: not run draft PR #25	physical: none recorded	Each check has a corrupted-fixture test
Imitation-learning baseline	NOT STARTED	runtime: not run -	physical: none recorded	Held-out evaluation recorded

Merged work

openamrobot-interfaces #16 C9 whole-body capture schema (LeRobotDataset), merged 2 Oct; contract still Proposed (merged 2026-10-02)

Open work

openamrobot-interfaces #25 C10 dataset quality gate (draft)

Runtime evidence

runtime: not run - no capture or policy run

Requirement coverage

partially implemented: UC1-60
missing: UC1-61 UC1-63
out of scope this pass: UC1-62
documented only: UC1-64

Open blockers

none recorded

Docs and release owner @KARTHIKEYAN124 (release lead) (release); documentation owner to be confirmed

The register content is current (P-03 rev19.7 decisions at 1c3f516; its label still says rev19.6, drift 1; BOM 7.10). The release builder cannot reproduce rc.1; the manifest tracks main; public docs still carry some superseded values.

Deliverable	Delivery	Runtime evidence (tested setup)	Physical	Findings · closes when
Reproducible rc.1 release	IN REVIEW	runtime: not run release #12, manifest #17 and #18 (all draft, 19 Oct); release #9 (navigation install evidence)	physical: none recorded	OA-05 OA-39 OA-06 Two builds from the rc.1 manifest give identical checksums
Decisions register with the P-03 rev19.7 decisions (1c3f516)	MERGED AND TESTED	runtime: not run .github 9109d3e CI	physical: none recorded	OA-01 OA-07 -

Merged work

openamrobot-docs #37 Option B upper body and 2.0 sensor layout (merged 2026-10-13)
openamrobot-docs #40 Reference use cases page (2.0 is Use Case 1; dark-store picking is 3.0) (merged 2026-10-13)
openamrobot-docs #39 Dark-store simulation source (merged 2026-10-11)
openamrobot-manifest #7 Manifest keys fixed and manifest validation added (merged 2026-09-27)

Open work

openamrobot-release #12 rc.1 built from pinned commit SHAs with verified checksums (OA-05) (draft)
openamrobot-manifest #17 SHA-pinned rc.1 release manifest (OA-39) (draft)
openamrobot-manifest #18 Pinned Jazzy driver sources and installation checks (OA-06) (draft)
openamrobot-release #9 Pinned navigation install and release manifest evidence (open)
openamrobot-manifest #13 Pin ROS 2 Jazzy image by digest (open)
openamrobot-manifest #11 RPLIDAR S3 driver in the manifest (open)
.github #41 Baseline ROS 2 CI image (open)
openamrobot-docs #47 Hardware reference to P-03 rev19.4 (draft)

Runtime evidence

runtime: not run - release builder not run for rc.1

Requirement coverage

implemented, untested: UC1-83
missing: UC1-84

Open blockers

OA-05 · code openamrobot-release/release-config.json:1 · owner to be confirmed · closes when: Two builds from the same rc.1 manifest give identical checksumsOA-06 · code openamr-platform-sw/ros2/src/openamrobot_drivers/package.xml · owner to be confirmed · closes when: Clean-machine rosdep check exits 0; root colcon build passesOA-39 · code openamrobot-manifest/openamrobot.repos:5 · owner to be confirmed · closes when: Release manifest with exact SHAs exists and feeds the builderOA-20 · code openamrobot-docs/docs/reference/openamrobot-2/index.md:39 · owner to be confirmed · closes when: Closure test to be defined by the ownerOA-21 · code openamrobot-docs/docs/assets/hardware/openamrobot2-general-arrangement.html:53 · owner to be confirmed · closes when: Closure test to be defined by the owner

Traceability: 85 rows

75 Use Case 1 requirements (baseline rev18.5) plus P-03 rev19.7 items 13 to 22 (item 23, the shared base, is tracked in section C of the overview and in the register), with the audit pass 2 classification. Requirement texts are short paraphrases; site-specific wording is neutralised (Use Case 1 site data is restricted).

needs hardware, not checkable in simulation: 26missing: 20partially implemented: 16implemented, untested: 6documented only: 5out of scope this pass: 5implemented and tested in simulation: 4contradicted: 3

StatusAreaSearch

ID	Requirement	Area	Status	Implementation	Evidence	Acceptance test	Owner	Needed for

What is next

Batch dates of 6, 13 and 20 Nov are proposed targets from audit pass 2 unless the plan or the register confirms the date.

2026-10-30
Gate A confirmed date
- Jetson with the existing Teensy and ZBLD/PWM drivetrain: command, odometry, RPLIDAR S3, Nav2, functional 200 ms /cmd_vel watchdog, recovery and safety-chain status telemetry (functional) pass the host-migration suite
- UI: Stop (software) replaces the E-STOP label, software stop lock, basic command routing and stop tests (@nikgolobo (simulation checks) checks in simulation that UI motion cannot bypass an active stop)
Source: P-02 rev19.2 (Gate A date and base scope; electrical and safety release the same day); UI scope agreed in coordination on 19 Oct; only the Stop (software) label is on main (ui #38)
2026-11-06
Batch 1 proposed target
- UI teleop, blocks and voice into an arbitrated input with a 0.25 s timeout
- #35 option (b) for docking
- UI-side navigation and mapping launches removed
- Replay restricted
- UI command path hardened
- Simulation body filter fixed on main
Source: Audit pass 2 top 20 actions 1 to 6 (proposed target). Confirmed separately: I8 freeze and STM32 go/no-go 6 Nov, fallback Teensy (both P-02 rev19.2, not the register); simulation body filter due 6 Nov in platform-sw #74
2026-11-13
Batch 2 proposed target
- UI access control
- Reactive motion layer under its own lifecycle manager
- Bounded startup
- Docking on the node clock
- Undock only from docked, cancellable
- Simulated base command timeout
- Stop and slowdown polygons
- Latched software stop
- Lift as a planning joint
- Simulated camera on base_camera_link
Source: Audit pass 2 top 20 actions 7 to 16 (proposed target). Confirmed on 13 Nov by issue bodies: lift as planned joint (upperbody-sw #17), joint clamp default (#18), lift interlock (platform-fw #20), micro-ROS transport (#21), base and arm exclusion (manipulation #24); Gate B regression 13 to 20 Nov (P-02 rev19.2)
2026-11-20
Batch 3 and v2.0.0-rc.1 confirmed date
- Navigation status from the live graph
- Reverse disabled until rear coverage; tight goal checker
- Voice plans bounded server-side
- Real route outcomes and stale pose shown in the UI
- v2.0.0-rc.1 published as a GitHub pre-release
Source: rc.1 date: register RELEASE-MILESTONES, P-03 item 12. Batch 3 items: audit pass 2 actions 17 to 20 (proposed target). STM32 Gate B regression 13 to 20 Nov (P-02 F9)
2026-11-23 to 2026-12-18
Physical integration, testing and acceptance confirmed date
- Supervised safety-chain tests (E-stop, contactors, brakes, reset, charge interlock)
- 26 traceability rows that need hardware
- Physical stopping tests that set the accepted operating speed
Source: P-03 item 12; Team document
2026-12-18
v2.0.0 confirmed date
- Final release
Source: Register RELEASE-MILESTONES; P-03 item 12

Findings register

Severity and confidence are separate fields. A verified finding is not a fix. Evidence kind says whether a scenario or test ran or whether the finding comes from reading code. Security findings are described at area level only.

SeverityArea

ID	Finding	Severity, confidence	Evidence kind	Area	Scenario	Public code location	Owner	Closure test	State	rc.1	Source
OA-01	Decisions register recorded the withdrawn LEV100H contactors instead of the GX11CAC	blocker verified (source conflict)	code review, not run	Safety chain	-	.github/decisions.yaml	to be confirmed	check_decisions.py flags LEV100H as current; register validates	closed on main Closed on main by .github PR #52 (9109d3e, 19 Oct): SAFETY-PROCUREMENT now records GX11CAC with LEV100H superseded. Checked in this review by reading decisions.yaml.	yes	audit pass 1
OA-02	Operator UI publishes teleop and block commands on the final velocity topic, past the smoother and collision monitor	blocker verified	code review, not run	Operator UI	-	openamrobot-ui/web/src/shared/constants/index.js:7	@parthkeshari (UI lead); reviewers @panthera-momagdii (tech lead, UI reviewer) and UI reviewer (GitHub handle to be confirmed); simulation checks @nikgolobo (simulation checks)	Rerun S4: UI command limited by the smoother and stopped by the collision monitor	open	yes	audit pass 1
OA-02-P2	UI teleop drove the simulated robot into a wall at 0.2 m/s (S4)	blocker verified	runtime	Operator UI	S4	openamrobot-ui/web/src/shared/constants/index.js:7	@parthkeshari (UI lead); reviewers @panthera-momagdii (tech lead, UI reviewer) and UI reviewer (GitHub handle to be confirmed); simulation checks @nikgolobo (simulation checks)	Rerun S4 and S7a: rosbridge is not a /cmd_vel publisher; robot stops before the wall; stops within 0.5 s after the UI connection drops	open	yes	audit pass 2
OA-03	Operator UI Maps page starts its own navigation stack outside the platform command chain	blocker verified	code review, not run	Operator UI	-	openamrobot-ui/ros2/src/openamr_ui_package/launch/move_base_launch.py:18	@parthkeshari (UI lead); reviewers @panthera-momagdii (tech lead, UI reviewer) and UI reviewer (GitHub handle to be confirmed); simulation checks @nikgolobo (simulation checks)	Rerun S5: build_map leaves one controller_server; platform map_server and AMCL states unchanged	open	yes	audit pass 1
OA-03-P2	UI Maps page started a second navigation controller (S5)	blocker verified	runtime	Operator UI	S5	-	@parthkeshari (UI lead); reviewers @panthera-momagdii (tech lead, UI reviewer) and UI reviewer (GitHub handle to be confirmed); simulation checks @nikgolobo (simulation checks)	Rerun S5: build_map leaves one controller_server; platform map_server and AMCL states unchanged	open	yes	audit pass 2
OA-04	Docking publishes the final velocity command and switches the collision monitor off during the dock	blocker verified	code review, not run	Docking	-	openamr-platform-sw/ros2/src/openamrobot_docking/config/dock_trigger.yaml:170	@skunal3318 (navigation lead)	Rerun S3: 5 of 5 docked; collision_monitor the sole /cmd_vel publisher; zero lifecycle transitions	open	yes	audit pass 1
OA-04-P2	Docking switched the collision monitor off; the navigation stack reset mid-dock (S3)	blocker verified	runtime	Docking	S3	openamr-platform-sw/ros2/src/openamrobot_docking/scripts/dock_trigger.py:833	@skunal3318 (navigation lead)	Rerun S3: 5 of 5 docked; collision_monitor the sole /cmd_vel publisher; zero lifecycle transitions	open fix: IN PROGRESS #35 branch	yes	audit pass 2
OA-05	Release builder cannot reproduce v2.0.0-rc.1 (no pinned SHAs, hand-supplied archives)	blocker verified	code review, not run	Docs and release	-	openamrobot-release/release-config.json:1	to be confirmed	Two builds from the same rc.1 manifest give identical checksums	open fix: IN REVIEW release #12 (draft)	yes	audit pass 1
OA-06	LiDAR and micro-ROS agent drivers cannot be installed from the manifest plus rosdep	blocker verified	runtime (build, test or smoke run)	Docs and release	-	openamr-platform-sw/ros2/src/openamrobot_drivers/package.xml	to be confirmed	Clean-machine rosdep check exits 0; root colcon build passes	open fix: IN REVIEW manifest #18 (draft)	yes	audit pass 1
OA-48	Recording replay can drive the robot; replay output needs restricting (S5)	blocker verified	area-level (detail withheld for security)	Operator UI	S5	withheld (security)	@parthkeshari (UI lead); reviewers @panthera-momagdii (tech lead, UI reviewer) and UI reviewer (GitHub handle to be confirmed); simulation checks @nikgolobo (simulation checks)	Replay of a recording that contains command topics: robot travel 0	open	yes	audit pass 2
OA-49	UI command path hardening: access control	blocker verified	area-level (detail withheld for security)	Operator UI	S4, S5	withheld (security)	@parthkeshari (UI lead); reviewers @panthera-momagdii (tech lead, UI reviewer) and UI reviewer (GitHub handle to be confirmed); simulation checks @nikgolobo (simulation checks)	Security review of the UI backend passes	open	yes	audit pass 2
OA-50	UI command path hardening	blocker verified	area-level (detail withheld for security)	Operator UI	S5 (code path)	withheld (security)	@parthkeshari (UI lead); reviewers @panthera-momagdii (tech lead, UI reviewer) and UI reviewer (GitHub handle to be confirmed); simulation checks @nikgolobo (simulation checks)	Security review of the UI backend passes	open	yes	audit pass 2
OA-07	Decisions register was four P-03 revisions behind (rev19.2, BOM 7.8)	should-fix verified (source conflict)	code review, not run	Docs and release	-	.github/decisions.yaml	to be confirmed	Register in_force is P-03 rev19.6 and BOM Issue 7.10	closed on main Closed on main by .github PR #52 (9109d3e, 19 Oct): in_force P-03 rev19.6, BOM Issue 7.10. Checked in this review.	yes	audit pass 1
OA-08	Workspace does not build at its root: openamr_ui_msgs exists in two repositories	should-fix verified	runtime (build, test or smoke run)	Operator UI	-	openamrobot-ui/ros2/src/openamr_ui_msgs/package.xml:3	@parthkeshari (UI lead); reviewers @panthera-momagdii (tech lead, UI reviewer) and UI reviewer (GitHub handle to be confirmed); simulation checks @nikgolobo (simulation checks)	colcon build at the workspace root exits 0	open	yes	audit pass 1
OA-09	Three lint tests fail locally on main; CI does not lint those packages	should-fix verified	runtime (build, test or smoke run)	Base and navigation	-	openamr-platform-sw/ros2/src/openamrobot_nav2/launch/navigation_launch.py:228	@skunal3318 (navigation lead)	Closure test to be defined by the owner	open	no	audit pass 1
OA-10	ToF frames at x +/-0.399 m in the URDF; register and P-03 give +/-0.390 m	should-fix verified	code review, not run	Base and navigation	-	openamr-platform-sw/ros2/src/openamrobot_description/urdf/sensors.xacro:125	to be confirmed	Frame check in CI against the register passes	open	no	audit pass 1
OA-11	Ultrasonic frames at x 0.3906 / -0.3886 m against the P-03 nominal +/-0.390 m (open item)	should-fix verified	code review, not run	Base and navigation	-	openamr-platform-sw/ros2/src/openamrobot_description/urdf/sensors.xacro:116	to be confirmed	Owner decision recorded; frame check passes	open	no	audit pass 1
OA-12	Upper body on main is a kinematic skeleton: no arms, no ros2_control, not attached to the base	should-fix verified	code review, not run	Upper body and manipulation	-	openamr-upperbody-sw/ros2/src/openamrobot_upperbody_description/urdf/openamrobot_upperbody.urdf.xacro:1	@Tushargoel230 (upper-body model)	PR #14 merged with A6 and A7 green on its exact head	open	yes	audit pass 1
OA-14	No 2.0 base-controller firmware on main (STM32 port, safety-chain diagnostic I/O, BMS, traction are placeholders)	should-fix verified	code review, not run	Firmware and base controller	-	openamr-platform-fw/boards/stm32/.gitkeep	@Evans-Kwashie (electrical and firmware lead)	I8 Gate B evidence on the NUCLEO-H723ZG bench (owner implements; agents do not write safety logic)	open	yes	audit pass 1
OA-15	I1 manipulation and I8 base-status contracts are not in the interfaces repository	should-fix verified	code review, not run	Interfaces	-	openamrobot-interfaces (main)	to be confirmed	I1 and I8 Proposed contracts build with consumers listed	open	yes	audit pass 1
OA-16	UI software stop is labelled E-STOP and does not latch	should-fix verified	code review, not run	Operator UI	-	openamrobot-ui/web/src/components/StatusBar.jsx:63	@parthkeshari (UI lead); reviewers @panthera-momagdii (tech lead, UI reviewer) and UI reviewer (GitHub handle to be confirmed); simulation checks @nikgolobo (simulation checks)	UI shows Stop (software); stop latches against every source	open	yes	audit pass 1
OA-16-P2	UI stop works for navigation goals but does not latch; any direct source resumes (S6)	should-fix verified	runtime	Operator UI	S6	openamrobot-ui/web/src/components/StatusBar.jsx:63	@parthkeshari (UI lead); reviewers @panthera-momagdii (tech lead, UI reviewer) and UI reviewer (GitHub handle to be confirmed); simulation checks @nikgolobo (simulation checks)	Rerun S6: after stop, UI teleop and docking produce no motion until release; button not labelled E-STOP	open	yes	audit pass 2
OA-17	Only real-robot bringup is the Gate A path with hand-set static transforms	should-fix verified	code review, not run	Base and navigation	-	openamr-platform-sw/ros2/src/openamrobot_bringup/launch/real_bringup.launch.py:78	@skunal3318 (navigation lead)	Launch test: TF tree single-rooted at map, odom, base_footprint on the 2.0 real profile	open	yes	audit pass 1
OA-18	Hardware repository states superseded base camera and IMU positions	should-fix verified	code review, not run	Docs and release	-	openamr-platform-hw/product-architecture.md:31	to be confirmed	Closure test to be defined by the owner	open	no	audit pass 1
OA-19	Firmware I/O page states the superseded IMU position	should-fix verified	code review, not run	Firmware and base controller	-	openamr-platform-fw/docs/architecture/base-controller-2-0-io.md:30	@Evans-Kwashie (electrical and firmware lead)	Closure test to be defined by the owner	open	no	audit pass 1
OA-20	Public docs and the dark-store simulation source carry pre-16 Oct LiDAR and camera values	should-fix verified	code review, not run	Docs and release	-	openamrobot-docs/docs/reference/openamrobot-2/index.md:39	to be confirmed	Closure test to be defined by the owner	open	no	audit pass 1
OA-21	Public assets show the superseded fixed-mast arrangement and LEV100 contactors without a label	should-fix verified	code review, not run	Docs and release	-	openamrobot-docs/docs/assets/hardware/openamrobot2-general-arrangement.html:53	to be confirmed	Closure test to be defined by the owner	open	no	audit pass 1
OA-22	Repositories cite superseded P-03 and BOM revisions	should-fix verified	code review, not run	Docs and release	-	-	to be confirmed	Closure test to be defined by the owner	open	no	audit pass 1
OA-23	E-01 safety chain is not documented on main of any repository (drafts open)	should-fix verified	code review, not run	Safety chain	-	openamr-platform-hw (main)	to be confirmed	Register check passes on each repository carrying the E-01 rev1.4 chain; two human approvals	open	yes	audit pass 1
OA-24	UI web front end has zero tests; CI passes the empty suite	should-fix verified	code review, not run	Operator UI	-	openamrobot-ui/.github/workflows/ci.yml:30	@parthkeshari (UI lead); reviewers @panthera-momagdii (tech lead, UI reviewer) and UI reviewer (GitHub handle to be confirmed); simulation checks @nikgolobo (simulation checks)	Web test suite runs more than 0 tests; --passWithNoTests removed	open	no	audit pass 1
OA-26	Runtime dependencies used but not declared in package.xml (heuristic)	should-fix inferred	code review, not run	Docs and release	-	openamr-platform-sw/ros2/src/openamrobot_bringup/package.xml	to be confirmed	Closure test to be defined by the owner	open	no	audit pass 1
OA-27	Hardware repository code owner differs from the Team document	should-fix verified	code review, not run	Docs and release	-	openamr-platform-hw/.github/CODEOWNERS:1	to be confirmed	Closure test to be defined by the owner	open	no	audit pass 1
OA-28	Communication repository code owner is a contributor who has left	should-fix verified	code review, not run	Docs and release	-	openamrobot-comm (main)	to be confirmed	Closure test to be defined by the owner	open	no	audit pass 1
OA-29	Owner conflicts inside the plan set (contactor rows, F2S)	should-fix verified (source conflict)	code review, not run	Docs and release	-	-	to be confirmed	Closure test to be defined by the owner	open	no	audit pass 1
OA-31	Older plan text keeps superseded IMU, LiDAR spacer and camera height values	should-fix verified (source conflict)	code review, not run	Docs and release	-	-	to be confirmed	Closure test to be defined by the owner	open	no	audit pass 1
OA-34	IMU topic ownership differs from the register (/imu vs /imu/data_raw and /imu/data)	should-fix verified	code review, not run	Base and navigation	-	openamr-platform-sw/ros2/src/openamrobot_gazebo/config/gz_bridge.yaml:37	@skunal3318 (navigation lead)	/imu/data_raw and /imu/data present in simulation; register check passes	open	no	audit pass 1
OA-35	Local costmap inflation radius below the inscribed radius	should-fix verified	runtime (build, test or smoke run)	Base and navigation	-	openamr-platform-sw/ros2/src/openamrobot_nav2/config/nav2_params.yaml:177	@skunal3318 (navigation lead)	No inflation error at start; S2 clearance at least 0.25 m	open	no	audit pass 1
OA-35-P2	Sudden obstacles passed at 0.117 to 0.128 m; collision monitor never engaged (S2)	should-fix verified	runtime	Base and navigation	S2	openamr-platform-sw/ros2/src/openamrobot_nav2/config/nav2_params.yaml:177	@skunal3318 (navigation lead)	Rerun S2: clearance at least 0.25 m; collision monitor engages in the barrier test	open	yes	audit pass 2
OA-36	Docking is a custom sequencer while Nav2 opennav_docking is installed and unused	should-fix verified	code review, not run	Docking	-	openamr-platform-sw/ros2/src/openamrobot_docking/scripts/dock_trigger.py	@fridayali (docking and robot description)	Recorded decision: opennav_docking routed through arbitration, or the sequencer kept with tests	open	no	audit pass 1
OA-37	No base-arm interlock, hold-and-move mode or lift interlock implemented	should-fix verified	code review, not run	Upper body and manipulation	-	openamr-platform-sw (main)	to be confirmed	Integration test both directions: arm goal rejected while base moves; base goal rejected while arm moves	open	yes	audit pass 1
OA-38	No physical, bench or HIL safety evidence record or record location	should-fix verified	code review, not run	Safety chain	-	-	to be confirmed	Evidence record location and procedure exist; supervised bench record attached	open	no	audit pass 1
OA-39	Manifest tracks main for every repository; no release manifest with immutable refs	should-fix verified	code review, not run	Docs and release	-	openamrobot-manifest/openamrobot.repos:5	to be confirmed	Release manifest with exact SHAs exists and feeds the builder	open fix: IN REVIEW manifest #17 (draft)	yes	audit pass 1
OA-40	Arm planning fails at the pinned upstream (no acceleration limits); gripper controller mismatch	should-fix verified	code review, not run	Upper body and manipulation	-	openamrobot-manipulation/integration/README.md	to be confirmed	Planning test passes without strict xfail at an approved acceleration limit set	open	yes	audit pass 1
OA-45-P2	After the UI connection dropped, the simulated robot kept driving 2.25 m (S7)	should-fix verified	runtime	Simulation	S7	openamr-platform-sw/ros2/src/openamrobot_description/urdf/gazebo_control.xacro:5	@skunal3318 (navigation lead)	Rerun S7a and S7c: simulated base stops within the command timeout after publisher loss	open	yes	audit pass 2
OA-46	Final alignment on site ArUco markers not implemented	should-fix verified	code review, not run	Docking	-	openamr-platform-sw/README.md:3	@fridayali (docking and robot description)	Align on site markers in simulation to +/-1 cm and +/-1 deg	open	no	audit pass 1
OA-47	Simulation on main cannot navigate: the body filter masks the forward sector of the yaw-pi LiDAR	should-fix verified	runtime	Simulation	S1	openamr-platform-sw/ros2/src/openamrobot_nav2/config/scan_body_filter.yaml:1	@skunal3318 (navigation lead)	Rerun S1 on main: 3 of 3 goals; a test fails when the mask covers the forward sector	open	yes	audit pass 2
OA-51	Voice plan path needs server-side bounds and confirmation before any motion	should-fix verified	area-level (detail withheld for security)	Operator UI	-	withheld (security)	@parthkeshari (UI lead); reviewers @panthera-momagdii (tech lead, UI reviewer) and UI reviewer (GitHub handle to be confirmed); simulation checks @nikgolobo (simulation checks)	Out-of-range plan rejected server-side; no direct velocity publish from plans	open	no	audit pass 2
OA-52	Block actions rely on one-shot velocity messages and the base watchdog	should-fix verified	code review, not run	Operator UI	S7	openamrobot-ui/web/src/features/blocks/robotActions.js:168	@parthkeshari (UI lead); reviewers @panthera-momagdii (tech lead, UI reviewer) and UI reviewer (GitHub handle to be confirmed); simulation checks @nikgolobo (simulation checks)	Timed motion goes through bounded Nav2 behaviors; no one-shot velocity	open	no	audit pass 2
OA-53	Undock reverses 0.7 m with no rear sensing, also when not docked; no cancel	should-fix verified	runtime	Docking	S3	openamr-platform-sw/ros2/src/openamrobot_docking/scripts/dock_trigger.py:942	@fridayali (docking and robot description)	Undock when not docked is refused; cancel stops within 0.5 s; no automatic undock after a failed dock	open	yes	audit pass 2
OA-54	Docking mixes wall-clock and sim-time deadlines; times out at low real-time factor	should-fix verified	runtime	Docking	S3	openamr-platform-sw/ros2/src/openamrobot_docking/scripts/dock_trigger.py:1800	@fridayali (docking and robot description)	Rerun S3 at 0.1 real-time factor with the same parameters as at 1.0: no wall-clock timeout	open	yes	audit pass 2
OA-55	Detected dock pose is re-stamped as fresh after the tag is lost	should-fix inferred	code review, not run	Docking	-	openamr-platform-sw/ros2/src/openamrobot_docking/src/detected_dock_pose_publisher.cpp:47	@fridayali (docking and robot description)	Stale detections older than a timeout are dropped	open	no	audit pass 2
OA-56	One Nav2 server crash takes the reactive motion layer (collision monitor and smoother) down (S7)	should-fix verified	runtime	Base and navigation	S7	openamr-platform-sw/ros2/src/openamrobot_nav2/launch/navigation_launch.py	@skunal3318 (navigation lead)	Rerun S7b: after killing controller_server the monitor stays active and zero velocity holds	open	yes	audit pass 2
OA-57	Startup is not reliable: 1 of 16 simulation starts stalled in localization	should-fix verified	runtime	Base and navigation	S1, S11	openamr-platform-sw/ros2/src/openamrobot_bringup/launch/bringup.launch.py	@skunal3318 (navigation lead)	S11 extended to 10 starts: all reach active within 120 s sim time	open	yes	audit pass 2
OA-58	Navigation status reports motion-source protection from a static profile, not the live graph	should-fix verified	code review, not run	Base and navigation	-	openamr-platform-sw/ros2/src/openamrobot_nav2/config/navigation_status_profile.yaml	@skunal3318 (navigation lead)	Navigation status flags rosbridge as an unprotected /cmd_vel publisher in S4	open	no	audit pass 2
OA-59	Simulated docking camera is on the legacy camera_link, not base_camera_link (Gemini 336L)	should-fix verified	runtime	Simulation	S10	openamr-platform-sw/ros2/src/openamrobot_description/urdf/robo_urdf.urdf.xacro:696	@fridayali (docking and robot description)	Rerun S10: /rgb_image frame base_camera_link; S3 passes with it	open fix: IN REVIEW PR #53 (draft)	yes	audit pass 2
OA-60	Route following reports success after a failed waypoint or cancel	should-fix verified	code review, not run	Operator UI	-	openamrobot-ui/ros2/src/openamr_ui_package/openamr_ui_package/waypoint_nav.py:122	@parthkeshari (UI lead); reviewers @panthera-momagdii (tech lead, UI reviewer) and UI reviewer (GitHub handle to be confirmed); simulation checks @nikgolobo (simulation checks)	Failed waypoint shows failure in the UI	open	no	audit pass 2
OA-62	UI angular limit 2.0 rad/s is four times the navigation limit and reaches the base unsmoothed	should-fix verified	runtime	Operator UI	S12	openamrobot-ui/web/src/shared/constants/index.js:53	@parthkeshari (UI lead); reviewers @panthera-momagdii (tech lead, UI reviewer) and UI reviewer (GitHub handle to be confirmed); simulation checks @nikgolobo (simulation checks)	UI angular limit taken from the platform and enforced behind the arbiter	open	no	audit pass 2
OA-63	Controller may reverse with the rear unobserved; goal tolerance 0.35 m against a 5 cm stage	should-fix verified	runtime	Base and navigation	S12	openamr-platform-sw/ros2/src/openamrobot_nav2/config/nav2_params.yaml:111	@skunal3318 (navigation lead)	Parameter test plus S1 final position error under 5 cm for approach goals	open	no	audit pass 2
OA-64	48 of 48 plans on mock hardware, but the lift is not a planning joint and the park pose sits at a joint limit	should-fix verified	runtime	Upper body and manipulation	S8	PR #14 (branch evidence)	@Tushargoel230 (upper-body model)	Rerun S8 with the lift in the planning group: 48 of 48 and every margin above 0.05 rad	open fix: IN REVIEW PR #14 (draft)	yes	audit pass 2
OA-68	Docking reports the collision monitor as down while it is active (false monitor status)	should-fix verified	runtime	Docking	S3	openamr-platform-sw/ros2/src/openamrobot_docking/scripts/dock_trigger.py:833	@fridayali (docking and robot description)	Rerun S3: reported monitor state equals the lifecycle state in every phase	open fix: IN PROGRESS #35 branch	yes	audit pass 2
OA-69	ToF and ultrasonic sensors publish in simulation but nothing consumes them	should-fix verified	runtime	Base and navigation	S10, S12	openamr-platform-sw/ros2/src/openamrobot_nav2/config/nav2_params.yaml	@skunal3318 (navigation lead)	Collision monitor and local costmap list the ToF and ultrasonic sources, or the register records their role	open	no	audit pass 2
OA-13	Lift joint has zero effort and velocity limits	later verified	code review, not run	Upper body and manipulation	-	openamr-upperbody-sw/ros2/src/openamrobot_upperbody_description/urdf/openamrobot_upperbody.xacro:66	@Tushargoel230 (upper-body model)	Closure test to be defined by the owner	open	no	audit pass 1
OA-25	Skip markers name no tracking issue	later verified	code review, not run	Docs and release	-	openamr-platform-sw (main)	to be confirmed	Closure test to be defined by the owner	open	no	audit pass 1
OA-30	Inconsistent GitHub handle and robot-model ownership records	later verified	code review, not run	Docs and release	-	-	to be confirmed	Closure test to be defined by the owner	open	no	audit pass 1
OA-32	E-01 component list reuses a removed BOM ID for the charge connector	later verified (source conflict)	code review, not run	Safety chain	-	-	to be confirmed	Closure test to be defined by the owner	open	no	audit pass 1
OA-33	Shoulder spacing 307 mm and mount y +/-0.031 m not checked against upstream arm geometry	later inferred	code review, not run	Upper body and manipulation	-	-	@Tushargoel230 (upper-body model)	Closure test to be defined by the owner	open	no	audit pass 1
OA-41	apriltag_node and move_group crash on shutdown	later verified	runtime (build, test or smoke run)	Simulation	-	openamr-platform-sw (main)	@skunal3318 (navigation lead)	Closure test to be defined by the owner	open	no	audit pass 1
OA-41-P2	Shutdown crashes recur: apriltag_node on every platform stop, move_group in 2 of 3 stops	later verified	runtime	Simulation	S11	openamr-platform-sw (main)	@skunal3318 (navigation lead)	Launch tests stop cleanly, or a tracking issue records the upstream crash	open	no	audit pass 2
OA-42	Unused hand-maintained SDF copy of the robot model	later verified	code review, not run	Simulation	-	openamr-platform-sw/ros2/src/openamrobot_description/urdf/robot.sdf	@fridayali (docking and robot description)	Closure test to be defined by the owner	open	no	audit pass 1
OA-43	Broken relative links and TODOs without issues	later verified	code review, not run	Docs and release	-	openamr-platform-sw/docs/troubleshooting/diagnostics.md	to be confirmed	Closure test to be defined by the owner	open	no	audit pass 1
OA-44	CI builds the navigation producer against an older interfaces commit	later verified	code review, not run	Interfaces	-	openamr-platform-sw/.github/workflows/nav-status-tests.yml:22	to be confirmed	Closure test to be defined by the owner	open	no	audit pass 1
OA-45	No ros2_control base description: simulation and robot share no controller or timeout	later opinion	code review, not run	Simulation	-	openamr-platform-sw/ros2/src/openamrobot_description/urdf/gazebo_control.xacro:5	@skunal3318 (navigation lead)	Simulated base stops within the command timeout after publisher loss	open	no	audit pass 1
OA-61	UI pose relay republishes a stale pose with its old stamp every second	later verified	code review, not run	Operator UI	-	openamrobot-ui/ros2/src/openamr_ui_package/openamr_ui_package/nav_relays.py:60	@parthkeshari (UI lead); reviewers @panthera-momagdii (tech lead, UI reviewer) and UI reviewer (GitHub handle to be confirmed); simulation checks @nikgolobo (simulation checks)	Pose older than 2 s shown as stale	open	no	audit pass 2
OA-65	joint_state_clamp is on in every profile, including hardware	later opinion	code review, not run	Upper body and manipulation	-	PR #14 (branch evidence)	@Tushargoel230 (upper-body model)	Closure test to be defined by the owner	open	no	audit pass 2
OA-66	UI writes maps and routes into the installed package share directory	later verified	code review, not run	Operator UI	-	-	@parthkeshari (UI lead); reviewers @panthera-momagdii (tech lead, UI reviewer) and UI reviewer (GitHub handle to be confirmed); simulation checks @nikgolobo (simulation checks)	Closure test to be defined by the owner	open	no	audit pass 2
OA-67	Copied navigation launch file drifted from upstream; two simulation entry points	later verified	code review, not run	Base and navigation	-	openamr-platform-sw/ros2/src/openamrobot_nav2/launch/navigation_launch.py	@skunal3318 (navigation lead)	Navigation launch includes upstream nav2_bringup; one simulation entry point	open	no	audit pass 2
NR-03	Motion sources keep publishing while actuator power is off; nothing holds them at zero across a RESET	blocker inferred	code review, not run	Base and navigation	-	-	@skunal3318 (navigation lead)	After RESET, no motion from any source until an explicit operator release; verified in simulation and on the bench	open fix: NOT STARTED design draft for the Teensy path (platform-fw issue #17)	yes	this review (new)
NR-01	The final velocity command is an unstamped Twist on every path	should-fix verified	code review, not run	Base and navigation	-	openamr-platform-sw/ros2/src/openamrobot_nav2/config/nav2_params.yaml	@skunal3318 (navigation lead)	See the recommendation in the architecture review	open	no	this review (new)
NR-02	No SROS2 configuration found in the inspected repositories	should-fix verified	code review, not run	Operator UI	-	-	@parthkeshari (UI lead)	See the recommendation in the architecture review	open	no	this review (new)
NR-05	No risk assessment found in the inspected repositories	should-fix opinion	opinion, not run	Safety chain	-	-	to be confirmed	See the recommendation in the architecture review	open	no	this review (new)
NR-06	No speed limit or contact policy for the arms near people	should-fix opinion	opinion, not run	Upper body and manipulation	-	PR #14 (branch)	@Tushargoel230 (upper-body model)	See the recommendation in the architecture review	open	no	this review (new)
NR-04	Package names mix two prefixes and one name exists twice	later verified	code review, not run	Interfaces	-	-	to be confirmed	See the recommendation in the architecture review	open	no	this review (new)

Merged pull requests since 21 Sep 2026

96 PRs merged since 21 Sep across 14 repositories (definition: pull requests in the openAMRobot organisation with merge date 21 Sep to 19 Oct 2026; GitHub search org:openAMRobot is:pr is:merged merged:2026-09-21..2026-10-19, cross-checked against the per-repository PR lists (19 Oct 2026, second snapshot); recount, matches the progress audit). A merged PR is not a tested product change; many are documentation, register and governance updates.

By repository

Repository	Merged
openamr-platform-sw	16
.github	14
openamrobot-interfaces	13
openamrobot-docs	12
openamr-platform-hw	8
openamrobot-ui	7
openamrobot-manipulation	6
openamrobot-manifest	4
openamr-upperbody-fw	4
openamr-upperbody-sw	4
openamr-upperbody-hw	3
openamr-platform-fw	3
openamrobot-comm	1
openamrobot-release	1

On 24 Oct: 111 PRs from 15 authors opened since 1 Oct, 41 of them open (the 21 Sep to 19 Oct merged count above is unchanged).