GR1 · PnPCanToDrawerClose¶
Agent runs
Each mode of this task runs once per run. Est. cost is tokens at list price (data/prices.yml), never a bill; see the RoboCasa-GR1 runs for every task of a run.
Codex + GPT-6 Luna, reasoning xhigh (ChatGPT login)default runclosed
| Trial | Started | Agent time | Model requests | Tokens in / out | Est. cost | Grade |
|---|---|---|---|---|---|---|
| U✓ 26m | 10-01 21:04 | 26m | — | 4.0M / 51k | $0.087 | deterministic 1, grader_error 0, missing_trajectory 0, n_actions 985, replay_success 1, video_rendered 1 |
| L✗ 49m | 10-01 20:09 | 49m | — | 15.8M / 120k | $0.246 | deterministic 1, grader_error 0, live_success 0, missing_log 0, replay_success 0, video_rendered 1 |
Task instruction (upstream)
Pick up the can, place it into the drawer and close the drawer.
Playback speed
Recorded by us: our reference solution, replayed by the verifier (the scene camera).
What RoboCasa-GR1 states about this task
| Success criteria | 1. the can's centre (its body origin) is inside the drawer's interior 2. the drawer is shut: upstream's opening fraction is at most 0.005 (0 = shut, 1 = fully open). 3. unlimited (privileged): both fresh-process replays of the handed-in trajectory end in the same state, and the check holds on it 4. limited (standard): the one episode (no reset) is recorded by the service and replays to the same state; the check holds on it live and in the replay |
| Family | robocasa-gr1/PnPCanToDrawerClose |
| Robot | Fourier GR1 humanoid: two arms, waist and two six-command Fourier hands; fixed base |
| Category | Pick and place, then close |
| Instance | the initial state of official demonstration episode 0 (demo_1, HDF5/PnPCanToDrawerClose.hdf5); MuJoCo state, model arrays, task state and RNG frozen in instance.npz (SHA-256 6c8f166549ca…, checked on load) |
| Deliverable | (T, 24) native controller commands, 1 ≤ T ≤ 3000, 20 Hz |
| Reference Solution | The demonstration's own actions (original: 288 actions), executed from the frozen scene by the robot's own controller: 288 actions in solution/oracle.npz. The demonstration is a reference solution: it is not in the image and the agent cannot reach it. |
| Limited Mode | Standard mode (robot as a service, eai-standard/2.1) of robocasa-gr1-can-to-drawer-close-i00-privileged: the same frozen instance and success check, served by the sim sidecar (images/robocasa-gr1/standard/server_casa.py, tool module casa_tool), which records the episode, replays it in a fresh simulator and judges it; the verifier grades the sidecar's record in a container of its own. The agent sees the robot's cameras (RGB-D, calibrated), its own joints and end effector, and the instruction. |
| Oracle | full |
| Base Image | ghcr.io/mll-lab-nu/eai-robocasa-gr1:0.1.0 |
| Agent Budget | 3600 s of wall clock per mode |
| Environment Source | https://github.com/robocasa/robocasa-gr1-tabletop-tasks/blob/4840e671596f93ca03651524b9f72ffb1aadfeff/README.md#L104 |
| Task Dirs | robocasa-gr1-can-to-drawer-close-i00-privileged, robocasa-gr1-can-to-drawer-close-i00-standard |
From https://github.com/robocasa/robocasa-gr1-tabletop-tasks @ 4840e67, as defined in our task definitions @ 8c5594a43.
Tags¶
Task DomainManipulation
Why this task is interesting¶
Not yet written.
Capability notes¶
Not yet written.
Oracle demo review¶
Not yet reviewed.