Fast Planning for Dynamic Preferences - Robotics Institute Carnegie Mellon University

Fast Planning for Dynamic Preferences

Brian D. Ziebart, Anind Dey, and J. Andrew (Drew) Bagnell
Conference Paper, Proceedings of 18th International Conference on Automated Planning and Scheduling (ICAPS '08), pp. 412 - 419, September, 2008

Abstract

We present an algorithm that quickly finds optimal plans for unforeseen agent preferences within graph-based planning domains where actions have deterministic outcomes and action costs are linearly parameterized by preference weights. We focus on vehicle route planning for drivers with personal trade-offs for different road types, and specifically on settings where these preferences are not known until planning time. We employ novel bounds (based on the triangle inequality and the concavity of the the optimal plan costs in the space of preferences) to enable the reuse of previously computed optimal plans that are similar to the new plan preferences. The resulting lower bounds are employed to guide the search for the optimal plan up to 60 times more efficiently than previous methods.

BibTeX

@conference{Ziebart-2008-10086,
author = {Brian D. Ziebart and Anind Dey and J. Andrew (Drew) Bagnell},
title = {Fast Planning for Dynamic Preferences},
booktitle = {Proceedings of 18th International Conference on Automated Planning and Scheduling (ICAPS '08)},
year = {2008},
month = {September},
pages = {412 - 419},
keywords = {Inverse Optimal Control, Probabilistic Reasoning, Navigation, Machine Learning},
}