Incremental Learning of Task Sequences with Information-Theoretic Metrics

Abstract

Learning tasks from human demonstration is a core feature for household service robots. Task knowledge should at the same time encode the constraints of a task while leaving as much flexibility for optimized reproduction at execution time as possible. This raises the question, which features of a task are the constraining or relevant ones both for execution of and reasoning over the task knowledge. In this paper, a system to record and interpret demonstrations of household tasks is presented. A measure for the assessment of information content of task features is introduced. This measure for the relevance of certain features relies both on general background knowledge as well as task-specific knowledge gathered from the user demonstrations. The results of the incremental growth of the task knowledge when more task demonstrations become available is demonstrated within the task of laying a table.

This work has been partially conducted within the german collaborative research center “SFB Humanoide Roboter” granted by Deutsche Forschungsgemeinschaft, and within the EU Integrated Project COGNIRON (“The cognitive robot companion”), funded by the European Commision Division FP6-IST Future and Emerging Technologies under Contract FP6-002020.