The Robotics Institute

Carnegie Mellon Robotics Institute

Copyright notice: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. These works may not be reposted without the explicit permission of the copyright holder.

Abstract

In our research, we view human behavior as a structured se- quence of context-sensitive decisions. We develop a con- ditional probabilistic model for predicting human decisions given the contextual situation. Our approach employs the principle of maximum entropy within the Markov Decision Process framework. Modeling human behavior is reduced to recovering a context-sensitive utility function that explains demonstrated behavior within the probabilistic model. In this work, we review the development of our probabilis- tic model (Ziebart et al. 2008a) and the results of its appli- cation to modeling the context-sensitive route preferences of drivers (Ziebart et al. 2008b). We additionally expand the approach's applicability to domains with stochastic dynam- ics, present preliminary experiments on modeling time-usage, and discuss remaining challenges for applying our approach to other human behavior modeling problems.

Notes

Text Reference

Brian D. Ziebart, Andrew L. Maas, J. Andrew (Drew) Bagnell, and Anind Dey, "Human Behavior Modeling with Maximum Entropy Inverse Optimal Control," April, 2008.

BibTeX Reference

@inproceedings{Ziebart_2008_7032,
   author = "Brian D. Ziebart and Andrew L. Maas and J. Andrew (Drew) Bagnell and Anind Dey",
   title = "Human Behavior Modeling with Maximum Entropy Inverse Optimal Control",
   booktitle = "",
   month = "April",
   year = "2008",
   number= "CMU-RI-TR-",
}

The Robotics Institute

Menu