Localization of Objects in Cyber Physical Production Systems for Industry 4.0 via Heterogeneous Camera Environments

Conference: Mobilkommunikation – Technologien und Anwendungen - 21. ITG-Fachtagung
05/11/2016 - 05/12/2016 at Osnabrück, Deutschland

Proceedings: Mobilkommunikation – Technologien und Anwendungen

Pages: 5Language: englishTyp: PDF

Personal VDE Members are entitled to a 10% discount on this title

Authors:
Fraunholz, Daniel; Schneider, Joerg; Wendling, André; Schotten, Hans D. (German Research Center for Artificial Intelligence, Germany)

Abstract:
Recently, Cyber Physical Production Systems (CPPS) have become an exciting emerging research area that has drawn the attention of many researchers. The localization of objects and persons in production lines is an ambitious topic and related to the issues of indoor localization. One of the most known localization systems is the Global Positioning System (GPS), which is based on received signals from GPS satellites and therefore have only limited usability in indoor usage. Indoor positioning systems (IPS) have been developed for position detection of objects inside a building. The majority of these systems are based on detection of wireless signal features like RSSI. Nonradio localization systems are for example magnetic positioning and inertial measurements. All these proposed systems assume measurements on an object itself and need additional hardware. Especially in CPPS it seems to be difficult to fulfill this requirement. To overcome this weakness, we introduce a stereo camera based localization system, build on the open-source framework OpenCV and the Nvidia Compute Unified Device Architecture (CUDA) framework. Furthermore, to illustrate the relevance, an industrial use case has been identified. In contrast to available implementations and results, our system is based on real time video data captured from two Microsoft HD Webcams.