Tool Integration and Execution on P3 Real-world Deployment (Tasks 1-6)
100Success RateP3
Evaluation Results
| Method | Links | |
|---|---|---|
| P3Backbone=GPT-o4-mini, Task No=2, Task Type=Passive, Active, Scheduled, Tool Usage=IoT, Scheduled task completion required=true2025.08 | 100 | |
| P3Backbone=Qwen-max, Task No=2, Task Type=Passive, Active, Scheduled, Tool Usage=IoT, Scheduled task completion required=true2025.08 | 100 | |
| RoboOSBackbone=GPT-o4-mini, Task No=4, Task Type=Passive, Scheduled, Tool Usage=None, Scheduled task completion required=false2025.08 | 100 | |
| P3Backbone=GPT-o4-mini, Task No=4, Task Type=Passive, Scheduled, Tool Usage=None, Scheduled task completion required=true2025.08 | 100 | |
| P3Backbone=Qwen-max, Task No=4, Task Type=Passive, Scheduled, Tool Usage=None, Scheduled task completion required=true2025.08 | 100 | |
| P3Backbone=GPT-o4-mini, Task No=5, Task Type=Passive, Active, Tool Usage=VLN, VLA, Scheduled task completion required=true2025.08 | 100 | |
| RoboOSBackbone=GPT-o4-mini, Task No=5, Task Type=Passive, Active, Tool Usage=VLN, VLA, Scheduled task completion required=false2025.08 | 92.86 | |
| P3Backbone=Qwen-max, Task No=5, Task Type=Passive, Active, Tool Usage=VLN, VLA, Scheduled task completion required=true2025.08 | 92.86 | |
| P3Backbone=GPT-o4-mini, Task No=1, Task Type=Active, Scheduled, Tool Usage=None, Scheduled task completion required=true2025.08 | 85.71 | |
| P3Backbone=Qwen-max, Task No=6, Task Type=Passive, Active, Tool Usage=VLN, VLA, Web, Scheduled task completion required=true2025.08 | 85.71 | |
| P3Backbone=Qwen-max, Task No=1, Task Type=Active, Scheduled, Tool Usage=None, Scheduled task completion required=true2025.08 | 78.57 | |
| P3Backbone=GPT-o4-mini, Task No=3, Task Type=Passive, Active, Scheduled, Tool Usage=VLA, Scheduled task completion required=true2025.08 | 78.57 | |
| RoboOSBackbone=GPT-o4-mini, Task No=6, Task Type=Passive, Active, Tool Usage=VLN, VLA, Web, Scheduled task completion required=false2025.08 | 78.57 | |
| P3Backbone=GPT-o4-mini, Task No=6, Task Type=Passive, Active, Tool Usage=VLN, VLA, Web, Scheduled task completion required=true2025.08 | 78.57 | |
| P3Backbone=Qwen-max, Task No=3, Task Type=Passive, Active, Scheduled, Tool Usage=VLA, Scheduled task completion required=true2025.08 | 71.43 |