March
1999
Learning Situation-Dependent Rules: Improving Task Planning for an Incompletely Modelled Domain
Authors:
Abstract:
Most real world environments are hard to model completely and correctly, especially to model the dynamics of the environment. In this paper we present our work to improve a domain model through learning from execution, thereby improving a task planner's performance. Our system collects execution traces from the robot, and automatically extracts relevant information to improve the domain model. We introduce the concept of {em situation-dependent rules}, where situational features are used to identify the conditions that affect action achievability. The system then converts this execution knowledge into a symbolic representation that the planner can use to generate plans appropriate for given situations.
copied = false, 2000);
">
@conference{Haigh-1999-14866,
author = {Karen Zita Haigh And Manuela Veloso},
title = {Learning Situation-Dependent Rules: Improving Task Planning for an Incompletely Modelled Domain},
booktitle = {Proceedings of AAAI '99 Spring Symposium on Search Techniques for Problem Solving under Uncertainty and Incomplete Information},
year = {1999},
month = {March},
pages = {30 - 35},
}
author = {Karen Zita Haigh And Manuela Veloso},
title = {Learning Situation-Dependent Rules: Improving Task Planning for an Incompletely Modelled Domain},
booktitle = {Proceedings of AAAI '99 Spring Symposium on Search Techniques for Problem Solving under Uncertainty and Incomplete Information},
year = {1999},
month = {March},
pages = {30 - 35},
}