Creation Review

Type
Creation
State
Successful
End Date of the Review Period

Reviews run for a minimum of one week. The outcome of the review is decided on this date. This is the last day to make comments or ask questions about this review.

Proposal

Eclipse MOSAIC

Monday, August 5, 2019 - 10:08 by Robert Protzmann
This proposal is in the Project Proposal Phase (as defined in the Eclipse Development Process) and is written to declare its intent and scope. We solicit additional participation and input from the community. Please login and add your feedback in the comments section.
Parent Project
Proposal State
Created
Background

For over 10 years the Smart Mobility research team at Fraunhofer FOKUS worked in the field of developing, testing, and assessing new approaches for connected and automated driving and mobility. Jointly with car and communication equipment manufactures as well as suppliers and telecommunication operators, we developed an extensive co-simulation framework for virtual testing and synthetic data generation. Alongside our co-simulation framework, we developed various simulation scenarios ranging from large scale full day traffic on regional level of Berlin and Brandenburg to medium scale sub-microscopic vehicle dynamics.



Recently we joined the Eclipse community in the working group “openMobility” and we plan to join “OpenADx” these days. The next step for us is the contribution in form of the proposed project - Eclipse MOSAIC (Multiscale Simulation and Testing for mobility Applications of the Next Generation).

 

Scope

Eclipse MOSAIC provides a multi-domain/multi-scale co-simulation environment for virtual testing of connected and automated driving and mobility solutions. The co-simulation environment consists of

  • Run-time Infrastructure (RTI) for simulator coupling, providing all management tasks for simulation orchestration and simulator interaction including the respective internal model adaptations and transfer
  • A collection of pre-packaged simulators for domains of interest, i.e. ad hoc and mobile network communication simulation, vehicle dynamics, and applications

 

Description

New technologies in the fields of communication (e.g. V2X, 5G networks), computing (Mobile Egde Clouds), advanced sensors (LiDAR or cameras), and batteries enable novel connected and automated mobility solutions, which take effect on different levels of scale. For instance, automated driving and other vehicle safety applications usually concern a limited number of vehicles and their local interplay (communication, computing). Mobility as a Service (MaaS) solutions for sustainability and traffic efficiency always consider large scales on whole city or regional level of mobile entities, which are not limited to vehicles only, but also public transport, bicycles etc. The development of all these solutions exhibits similarities. Preferably, the process starts with simulation, again on different scales and even levels of detail, ranging from individual vehicle simulation to whole system simulation. However, the simulation of those novel mobility solutions not only requires the integration of various domains ranging from realistic and large-scale traffic demand patterns to exact simulation of mobile communication networks including cell capacity and latency, but also has to consider the interactive nature of connected mobility domain, where results of one simulation domain e.g. mobile network availability influencing traffic patterns or demands. The proposed multi-domain/multi-scale co-simulation environment is able to fulfil all the above requirements.

First of all, the co-simulation environment will include a runtime-environment for coupling discrete event simulators from multiple domains. According to the standardized coupling principles of IEEE High Level Architecture (HLA), simulators are embedded in a federate and the interface to the RTI is realized by an ambassador. The project integrates interfaces to the following simulators:

  • Eclipse SUMO for large scale traffic simulation (third-party development under EPL 2.0)
  • ns-3 for communication simulation (third-party development under GPL)
  • OMNeT++ for communication simulation (third-party development)
  • PHABMACS for sub-microscopic simulation with detailed models (see below)
  • CELL2 for simulation of cellular mobile networks including novel 5G features (part of this project, see below)

An advanced feature of dynamic co-simulation is the possibility to exchange simulators between simulation runs or even during run-time. This allows for quick setup of new simulation scenarios with a simple simulator and a later more detailed run with a more elaborate simulator. Additionally, reusing simulation scenarios but varying e.g. communication technologies such as ITS G5, LTE-V2X, and mobile networks is easily done with a co-simulation environment.

 

Why Here?

We already have a community of over 140 institutions utilizing this co-simulation environment for their work. We want to further strengthen our efforts by bringing MOSAIC under the Eclipse umbrella. A co-simulation environment is getting better with each new domain supported; we also hope that the Eclipse umbrella brings the drive and continuity for others to integrate their simulator into this project. Additional simulation domain could include electric grid, advanced driver models, pedestrian and bicycle simulation and many more.

Additionally, the Eclipse working groups “openMobility” and “OpenADx” are prime user of this proposed project.

Future Work

According to our ideas, the new feature sets could regard to

  1. new model details for existing domains – e.g.
    • 5G innovations like cloud computing in the cellular communication simulator
    • Sensor models or drive-train models in the vehicle simulator
  2. new simulation domains – e.g.
    • battery simulation in an electric mobility scenario
  3. new standardized simulation scenarios for testing and certification

 

Project Scheduling

The project schedule aims for a one step contribution from the already developed code base and regular later commits (meaning complete feature sets)

  1. Initial Contribution (see item before)
    • The initial contribution should happen in a time frame of 6 months after the project proposal was accepted by the Eclipse community
  2. Regularly later Commits
    • Still slightly open and dependent on strategic roadmap developed with our partners in the Eclipse WGs

 

Mentors
Interested Parties
  • Our long-term partners and users from current cooperations
  • Our partners from associated Eclipse working groups “openMobility”, “OpenADx”
Initial Contribution

The initial contribution includes the following parts of the already existing code base. The aim is to make all needed parts available for coupling and development of the main domains of smart mobility systems – traffic, vehicle, (wireless) communication, applications.

  • Simulation Libraries
  • Simulation Domains
    • SUMO Ambassador -> Eclipse SUMO
    • PHABMACS Ambassador -> PHABMACS Scenario SE
    • OMNeT++ Ambassador -> OMNeT++ / INET (and possibly other frameworks as VEINS)
    • ns-3 Ambassador -> ns-3
    • CELL2 Ambassador and CELL2 Simulator
    • ApplicationNT Ambassador and Simulator

 

Source Repository Type