How Robots Can Recognize Activities and Plans Using Topic Models

Abstract

The ability to identify what humans are doing in the environ-ment is a crucial element of responsive behavior in human-robot interaction. We examine new ways to perform plan recognition (PR) using natural language processing (NLP) techniques. PR often focuses on the structural relationships between consecutive observations and ordered activities that comprise plans. However, NLP commonly treats text as a bag-of-words, omitting such structural relationships and us-ing topic models to break down the distribution of concepts discussed in documents. In this paper, we examine an anal-ogous treatment of plans as distributions of activities. We explore the application of Latent Dirichlet Allocation topic models to plan execution traces obtained from human postu-ral data read by a RGB-D sensor. This investigation focuses on representing the data as text and interpreting learned ac-tivities as a form of activity recognition (AR). Additionally, we explain how the system may perform PR. The initial em-pirical results suggest that such NLP methods can be useful in complex PR and AR tasks. 1