Simulation as a Service for Cooperative Vehicles

Simulating connected vehicles in realistic environments is a computationally expensive task, particularly when large numbers of traffic participants are involved. To cope with exploding hardware requirements it can become indispensable to tackle such simulations in a divide and conquer manner. A pro...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:2019 ACM/IEEE 22nd International Conference on Model Driven Engineering Languages and Systems Companion (MODELS-C) S. 28 - 37
Hauptverfasser: Kirchhof, Jorg Christian, Kusmenko, Evgeny, Rumpe, Bernhard, Zhang, Hengwen
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 01.09.2019
Schlagworte:
Computational modeling
Continuous integration
cooperative vehicles
Hardware
Memory management
Model driven engineering
Research and development
Resource management
Scalability
Shape
simulation as a service
Testing
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Simulating connected vehicles in realistic environments is a computationally expensive task, particularly when large numbers of traffic participants are involved. To cope with exploding hardware requirements it can become indispensable to tackle such simulations in a divide and conquer manner. A promising approach breaking down a simulation scenario in a set of small tasks is the spatial subdivision of the simulated area. Thereby, each spatial sector is simulated by a dedicated sub-simulator enabling efficient distribution of the overall simulation across multiple machines. However, a distributable and easy-to-use simulation solution providing automated scalability and hiding the distribution complexity from the simulation engineer requires a service-based simulator architecture equipped with interfaces for simulation control, integration, and resource management. In this work, building upon previous results we propose a cloud-ready simulation infrastructure providing automotive engineers with the ability to analyze large scale connected traffic scenarios without caring about the execution environment. As a consequence, the proposed solution can be integrated seamlessly into existing continuous integration environments supporting agile research and development processes for cooperatively interacting vehicles.
DOI:10.1109/MODELS-C.2019.00011