Publications
From Knowledgewiki
The following publications are describing aspects of the KnowRob system. This documentation targets more at the installation, usage and extension of the system, while the papers mainly describe the concepts and contributions.
KnowRob - Knowledge Processing for Autonomous Personal Robots
The KnowRob overview paper describes the overall system and presents some applications it can be used for. Recommended to be read first.
Abstract
Mobile household robots need much knowledge about objects, places and actions when performing more and more complex tasks. They must be able to recognize objects, know what they are and how they can be used. We present a practical approach to robot knowledge representation that combines description logics knowledge bases with a rich environment model, data mining and (self-) observation modules. The robot observes itself and humans while executing actions and uses the collected experiences to learn models of action-related concepts grounded in its perception and action system. We demonstrate our approach by learning places that are involved in mobile robot manipulation actions, by locating objects based on their function and by supplying knowledge required for understanding underspecified task descriptions as commonly given by humans.
@InProceedings{tenorth09knowledge,
author = {Moritz Tenorth and Michael Beetz},
title = {KnowRob --- Knowledge Processing for Autonomous Personal Robots},
booktitle = {IEEE/RSJ International Conference on Intelligent RObots and Systems.},
year = {2009}
}
Knowledge Representation for Cognitive Robots
An overview paper discussing general aspects of robot knowledge processing and the approaches we take in the KnowRob system.
Abstract
Knowledge processing methods are an important resource for robots that perform challenging tasks in complex, dynamic environments. When applied to robot control, such methods allow to write more general and flexible control programs and enable reasoning about the robot's observations, the actions involved in a task, action parameters and the reasons why an action was performed. However, the application of knowledge representation and reasoning techniques to autonomous robots creates several hard research challenges. In this article, we discuss some of these challenges and our approaches to solving them.
@article{tenorth10kr,
title={Knowledge Representation for Cognitive Robots},
author={Moritz Tenorth and Dominik Jain and Michael Beetz},
journal = {K{\"u}nstliche Intelligenz},
publisher = {Springer},
year = {2010},
volume = {24},
number = {3},
pages = {233--240}
}
Understanding and Executing Instructions for Everyday Manipulation Tasks from the World Wide Web
This paper describes the import of knowledge from web sites like ehow.com and the transformation of the natural-language instructions into a formal representation.
Abstract
Service robots will have to accomplish more and more complex, open-ended tasks and regularly acquire new skills. In this work, we propose a new approach to generating plans for such household robots. Instead composing them from atomic actions - the common approach in robot planning - we propose to transform task descriptions on web sites like ehow.com into executable robot plans. We present methods for automatically converting the instructions from natural language into a formal, logic-based representation, for resolving the word senses using the WordNet database and the Cyc ontology, and for exporting the generated plans into the mobile robot's plan language RPL. We discuss the problems of inferring information missing in these descriptions, of grounding the abstract task descriptions in the perception and action system, and we propose techniques for solving them. The whole system works autonomously without human interaction. It has successfully been tested with a set of about 150 natural language directives, of which up to 80% could be correctly transformed.
@InProceedings{tenorth10webinstructions,
author = {Moritz Tenorth and Daniel Nyga and Michael Beetz},
title = {Understanding and Executing Instructions for Everyday Manipulation Tasks from the World Wide Web.},
booktitle = {IEEE International Conference on Robotics and Automation (ICRA)},
year = {2010}
}
KNOWROB-MAP -- Knowledge-Linked Semantic Object Maps
This paper describes how KnowRob can be used to represent and reason about environment-related information.
Abstract Autonomous household robots are supposed to accomplish complex tasks like cleaning the dishes which involve both navigation and manipulation within the environment. For navigation, spatial information is mostly sufficient, but manipulation tasks raise the demand for deeper knowledge about objects, such as their types, their functions, or the way how they can be used. We present KNOWROB-MAP, a system for building environment models for robots by combining spatial information about objects in the environment with encyclopedic knowledge about the types and properties of objects, with common-sense knowledge describing what the objects can be used for, and with knowledge derived from observations of human activities by learning statistical relational models. In this paper, we describe the concept and implementation of KNOWROB-MAP and present several examples demonstrating the range of information the system can provide to autonomous robots.
@InProceedings{tenorth10envmodel,
author = {Moritz Tenorth and Lars Kunze and Dominik Jain and Michael Beetz},
title = {KNOWROB-MAP -- Knowledge-Linked Semantic Object Maps},
booktitle = {Proceedings of 2010 IEEE-RAS International Conference on Humanoid Robots},
month = {December 6-8},
year = {2010},
address = {Nashville, TN, USA}
}
Putting People's Common Sense into Knowledge Bases of Household Robots
This paper describes the translation of knowledge from the OMICS database into a format that can be loaded into KnowRob to equip the robot with a large amount of common-sense knowledge.
Abstract
Unlike people, household robots cannot rely on commonsense knowledge when accomplishing everyday tasks. We believe that this is one of the reasons why they perform poorly in comparison to humans. By integrating extensive collections of commonsense knowledge into mobile robot's knowledge bases, the work proposed in this paper enables robots to flexibly infer control decisions under changing environmental conditions. We present a system that converts commonsense knowledge from the large Open Mind Indoor Common Sense database from natural language into a Description Logic representation that allows for automated reasoning and for relating it to other sources of knowledge.
@InProceedings{kunze10omics,
author = {Lars Kunze and Moritz Tenorth and Michael Beetz},
title = {Putting People's Common Sense into Knowledge Bases of Household Robots},
booktitle = {33rd Annual German Conference on Artificial Intelligence (KI 2010)},
month = {September 21-24},
year = {2010},
address = {Karlsruhe, Germany},
publisher = {Springer}
}
