Pack Objects Into Box

RoboDojo simulation reference

SE3 Labs real setup

Task ID
SE3 language instruction	“Place the car, shoe, toothbrush, and dual-face hammer into the open cardboard box. Point the car nose, shoe toe, toothbrush head, and hammer's yellow striking face to the .”
Official language instruction	Place all the objects into the box with their front sides facing left.
RoboDojo category	Generalization
Materials	orange toy car, red shoe, blue electric toothbrush, dual-face hammer with black and yellow striking faces, and an open corrugated-cardboard box with attached lid.
Difference	object identity, mass, compliance, and grasp affordances are substantially different from simulation; the box is deformable and “front side” is otherwise ambiguous.
RoboDojo task page	Pack Objects Into Box

Progress score

Score the highest milestone whose condition holds at the end of the episode; the episode succeeds at 100. Also record the failure mode and the execution metrics below.

An object is packed when it is fully inside the box and settled, and oriented when it also faces : car nose, shoe toe, toothbrush head, hammer's yellow striking face. The box may stay open.

Progress	Condition
0	No milestone below is reached.
5	A task object is grasped and lifted clear of the table.
15	One object is packed and the gripper is open.
30	Two objects are packed.
45	Three objects are packed.
60	All four objects are packed.
80	All four objects are packed and at least two are oriented.
100	All four objects are packed and oriented, and the box is within its initial footprint. Both arms are back at origin.

Execution metrics

Record these alongside progress; they do not change it. Count over the whole episode, including events later corrected.

Metric	What to count	Answer
Failed grasps	Grasps that close on or around a task object without lifting it clear of the table.	0 · 1 · 2 · 3+
Drops	Held objects that fall before they are released at a target.	0 · 1 · 2 · 3+
Re-placements	Objects picked up or pushed again after being released in the box.	0 · 1 · 2 · 3+
Box displaced	Did the box leave its initial footprint at any time?	yes · no

Randomization

Each episode samples every variable below and records its values.

Variable	Domain
	left · right
Start layout	Box position near the back, yaw within ±20°; object positions and yaw anywhere else on the table, clear of the arms.

The hammer’s yellow striking face defines its front. Its 26 × 8 cm layout footprint is estimated from the setup photo.

Schematic start layout, not to scale. Hover a shape for details.

Python SDK

from se3labs.sdk.eval.tasks import TaskRegistry

definition = TaskRegistry.get_task("ROBODOJO-REAL-03")
# Sample one episode's parameters and initial layout.
task = definition.sample(seed=42)
# Sample 20 episodes with separate seeds; seed=42 reproduces the same set.
tasks = definition.sample_set(seed=42, episodes=20)

Create and run a job:

from se3labs.sdk import eval_client

job = TaskRegistry.get_job(
    "ROBODOJO-REAL-03", policy=policy, seed=42, episodes=20,
)
report = eval_client.serve_policy("your-station", "your-relay:7443", job)

task.layout.visualize("layout.png") saves the sampled layout.