Multi-agent Planning using Visual Language Models

ECAI 2024

Examine a CD in the light (trial_T20190907_114323_767231)

Ground truth Single Agent with table Multi Agent with table Single Agent with image Multi Agent with image
STEP 1: Turn around, go straight, then turn left to go toward the door.
STEP 2: Take the CD from the trash can, under the mirror.
STEP 3: Turn around, then turn right to go to the desk with the lamp.
STEP 4: Turn on the lamp.
Turn around and walk to the GarbageCan at position (2.257, -0.008, 2.603).,
Take the CD from the GarbageCan.,
Turn around and walk to the DeskLamp at position (-0.669, 0.923, -0.062).,
Examine the CD under the DeskLamp.
Turn and walk to the CD.,
Pick up the CD.,
Turn and walk to the LightSwitch.,
Turn on the LightSwitch.,
Turn and walk to the DeskLamp.,
Adjust the DeskLamp to shine on the CD.,
Examine the CD in the light.
1. Walk to the desk located on the left side of the room near the window.
2. Pick up the CD that is on the desk.
3. Turn to face the window to the left of the desk.
4. Walk to the window.
5. Hold the CD up in the light coming through the window to examine it.
1. Walk to the small table in front of the mirror.
2. Pick up the CD on the small table, next to the stapler.
3. Walk to the window, where the desk is next to.
4. Stand at the desk, on which the lamp is placed.
5. Turn on the lamp on the desk.
6. Hold the CD under the lamp on the desk to examine it in the light.
Architecture