Why does the OpenDORV2 dataset only have target rotation labels but no target position labels?

[babynabeauty]

Hello author, thank you very much for your work.
I would like to ask why in the OpenDORV2 dataset there are only target rotation labels but no target position labels? For example, in a sample from the task-refine-6dof, where the goal is to move the apple in front of the hammer, there is a ground truth label for the quaternion but no ground truth label for the position.

{
"selected_obj_names": [
"hammer",
"apple"
],
"selected_urdfs": [
"objaverse_rescale/6d155e8051294915813e0c156e3cd6de/material_2.urdf",
"ycb_16k_backup/013_apple_google_16k/013_apple_google_16k.urdf"
],
"target_obj_name": "apple",
"instruction": "Place the apple in front of hammer on the table. We also need to specify the rotation of the object after placement: the object is placed upright on the table and corresponds with how humans usually place the object, bottom down and top up.",
"init_obj_pos": [
[
0.38011598587036133,
-0.16197378933429718,
0.3258112072944641,
0.08212229609489441,
-0.005923671647906303,
-0.07015541940927505,
0.9941322803497314,
0.0002099604462273419,
0.0052893501706421375,
-0.01015145517885685,
-0.1913883090019226,
-0.33170220255851746,
-0.06518398225307465
],
[
0.596611738204956,
0.010879392735660076,
0.3002125024795532,
0.005460700951516628,
-0.020027074962854385,
-0.05191946402192116,
0.9984354972839355,
-0.0027114644180983305,
0.0007762593450024724,
-0.0005421206587925553,
-0.004948922898620367,
-0.07674112915992737,
0.03726702556014061
]
],
"position_instruction": "Place the apple in front of hammer on the table. ",
"rotation_instruction": "We also need to specify the rotation of the object after placement: the object is placed upright on the table and corresponds with how humans usually place the object, bottom down and top up.",
"rotation_instruction_label": "upright",
"obj_codes": [
"6d155e8051294915813e0c156e3cd6de",
"013"
],
"target_obj_code": "013",
"anno_target": {
"category": "apple",
"annotation": {
" the object is placed upright on the table and corresponds with how humans usually place the object, bottom down and top up.": {
"quat": [
[
-0.018678417429327965,
-0.0,
0.0,
0.9998255372047424
]
],
"stage": 1,
"axis": "z"
}
}
},
"rot_tag_detail": "upright",
"rot_tag_level": 0,
"position_tag": "front",
"rotation_tag": "upright"
}