Combining Offline and Online Computation for Solving Partially Observable Markov

Abstract

Partially observable Markov decision process (POMDP) provides a general and mathematically elegant way of formulating planning and control problems under uncertainty. Unfortunately, POMDPs are computationally intractable to solve in the worst case, prompting the development of approximation algorithms. In this project, we explore the use online algorithms… (More)

7 Figures and Tables

Topics

  • Presentations referencing similar topics