MLDT: Multi-Level Decomposition for Complex Long-Horizon Robotic Task Planning with Open-Source Large Language Model
About
In the realm of data-driven AI technology, the application of open-source large language models (LLMs) in robotic task planning represents a significant milestone. Recent robotic task planning methods based on open-source LLMs typically leverage vast task planning datasets to enhance models' planning abilities. While these methods show promise, they struggle with complex long-horizon tasks, which require comprehending more context and generating longer action sequences. This paper addresses this limitation by proposing MLDT, theMulti-Level Decomposition Task planning method. This method innovatively decomposes tasks at the goal-level, task-level, and action-level to mitigate the challenge of complex long-horizon tasks. In order to enhance open-source LLMs' planning abilities, we introduce a goal-sensitive corpus generation method to create high-quality training data and conduct instruction tuning on the generated corpus. Since the complexity of the existing datasets is not high enough, we construct a more challenging dataset, LongTasks, to specifically evaluate planning ability on complex long-horizon tasks. We evaluate our method using various LLMs on four datasets in VirtualHome. Our results demonstrate a significant performance enhancement in robotic task planning, showcasing MLDT's effectiveness in overcoming the limitations of existing methods based on open-source LLMs as well as its practicality in complex, real-world scenarios.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Monkey-banana task execution | Dual Bananas Scene 4 | Success Rate99 | 6 | |
| Robotic Planning | Comprehensive Scene 13 | SR98 | 6 | |
| Robotic Task Planning | Classic (Scene 2) | Success Rate85 | 6 | |
| Monkey-banana task execution | Dual Bananas Scene 5 | Success Rate65 | 6 | |
| Robotic Planning | Comprehensive (Scene 14) | SR12 | 6 | |
| Robotic Task Planning | Classic (Scene 1) | Success Rate (SR)98 | 6 | |
| Task Planning | Shortsighted Monkey (Scene 7) | SR96 | 6 | |
| Monkey-banana task execution | Dual Bananas Scene 6 | Success Rate90 | 6 | |
| Robotic Planning | Scene 15 Comprehensive | SR0.00e+0 | 6 | |
| Robotic Task Planning | Classic (Scene 3) | Success Rate (SR)0.00e+0 | 6 |