Home // International Journal On Advances in Software, volume 10, numbers 1 and 2, 2017 // View article

Managing Semantic World Models for eRobotics Applications Two Approaches Based on Object-Relational Mapping and on a Graph Database

Authors:
Martin Hoppen
Juergen Rossmann
Ann-Marie Stapelbroek
Sebastian Hiester

Keywords: eRobotics; Semantic World Models; Data Management; Object Relational Mapping; Graph Database

Abstract:
The main objective of eRobotics is to cope with the complexity in the development and lifecycle of current robotic and mechatronic systems. A key technology for eRobotics applications are Virtual Testbeds, a more holistic approach to 3D simulation that considers the entire system under study within its operational environment modeled in so-called Semantic World Models. Ideally, this complex data should be managed using database technology instead of flat file formats. However, existing related approaches fail to fulfill all the identified requirements. Thus, we present a new database synchronization concept whose basic idea is to use local in-memory simulation databases (SimDB) and synchronize them to a central, external database (ExtDB). In this paper, two new realizations of this concept are presented using two different choices for ExtDB: The first is based on an object-relational mapping approach, the second uses the graph database Neo4j. Both approaches are described in detail, evaluated, and finally compared to each other. In conclusion, both approaches are very well-suited but the actual choice depends the concrete application scenario.

Pages: 79 to 95

Copyright: Copyright (c) to authors, 2017. Used with permission.

Publication date: June 30, 2017

Published in: journal

ISSN: 1942-2628