Authors:
Bernhard G. Humm
1
and
Guglielmo van der Meer
2
Affiliations:
1
Hochschule Darmstadt, University of Applied Sciences, Darmstadt and Germany
;
2
KUKA Roboter GmbH, Augsburg and Germany
Keyword(s):
Robots, Time Series, Complex Event Detection.
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence
;
Artificial Intelligence and Decision Support Systems
;
Business Analytics
;
Cardiovascular Technologies
;
Computational Intelligence
;
Computing and Telecommunications in Cardiology
;
Data Engineering
;
Decision Support Systems
;
Decision Support Systems, Remote Data Analysis
;
Enterprise Information Systems
;
Health Engineering and Technology Applications
;
Informatics in Control, Automation and Robotics
;
Intelligent Control Systems and Optimization
;
Knowledge-Based Systems
;
Knowledge-Based Systems Applications
;
Soft Computing
;
Symbolic Systems
Abstract:
This paper presents an approach for detecting domain-specific events based on robot sensor data. Events may be error situations as well as successfully executed manufacturing steps, depending on the application domain at hand. The approach includes segmenting streams of sensor data into meaningful intervals and subsequently matching patterns on those segments. Pattern matching is performed in near real-time allowing events to be detected continuously during the execution of a robotics application. The approach is demonstrated by means of a real-world manufacturing use case, namely the automated assembly of electrical components by a robot. The approach has been implemented prototypically had has been evaluated successfully.