Automatic Generation of Robot Applications Using a Knowledge Integration Framework

Konferenz: ISR/ROBOTIK 2010 - ISR 2010 (41st International Symposium on Robotics) and ROBOTIK 2010 (6th German Conference on Robotics)
07.06.2010 - 09.06.2010 in Munich, Germany

Tagungsband: ISR/ROBOTIK 2010

Seiten: 8Sprache: EnglischTyp: PDF

Persönliche VDE-Mitglieder erhalten auf diesen Artikel 10% Rabatt

Naumann, Martin; Bengel, Matthias; Verl, Alexander (Fraunhofer-Institut für Produktionstechnik und Automatisierung IPA, Stuttgart, Germany)

This article describes an integrated approach for the handling and modeling of knowledge for assembly processes in an automated production environment. High-level information on the process provided by the user needs to be transferred into executable code. This is informal information and might be even available in natural language. With the help of several information sources like CAD or sensor data, device and skill descriptions, planning algorithms, process knowledge and finally domain knowledge the Knowledge Integration Framework derives a formal application description. Therefore, the user input is parsed and analyzed with all the relevant knowledge made available in a formal representation beforehand. With the help of computer-based reasoning and inference algorithms the input is evaluated and enhancements for missing information are requested from the user. The core part of the implementation is to capture and to enable access to the available knowledge. Without a proper representation of all process and operation details the Knowledge Integration Framework cannot perform this challenging work. The formal application description is fed into the code generator where device-specific code together with the execution sequence is created. On runtime these executable programs - after deployment to the physical devices - are iteratively readapted taking strategies for safety and error recovery into account. This work promises a strong potential for realizing future robot installations in a robust and efficient way, taking into account a dynamic environment and human robot cooperation with respect to safe task execution and human injury prevention.