Introspective Planning:

Aligning Robots' Uncertainty with Inherent Task Ambiguity

We propose a novel introspective planning scheme that prompts language-enabled agents to proactively assess their own confidence regarding task compliance and safety for multiple candidate plans.


@inproceedings{liangintrospective,
      title={Introspective Planning: Aligning Robots' Uncertainty with Inherent Task Ambiguity},
      author={Liang, Kaiqu and Zhang, Zixu and Fisac, Jaime Fern{\'a}ndez},
      booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}
    }

Citation

Authors

Acknowledgement

Next
Next

Who Plays First? Optimizing the Order of Play in Stackelberg Games with Many Robots