MOSAIC: Automated Model Transfer in Simulator Development

Similar documents
Real-Time AOCS EGSE Using EuroSim and SMP2-Compliant Building Blocks

A Methodology for Effective Reuse of Design Simulators in Operational Contexts: Lessons Learned in European Space Programmes

Software Tools for Modeling Space Systems Equipment Command-and-Software Control. Ludmila F. NOZHENKOVA, Olga S. ISAEVA and Alexander A.

MAPLESIM: Tailored solutions for networking legacy flight simulators an HLA based approach

SCOE SIMULATION. Pascal CONRATH (1), Christian ABEL (1)

ARTES Competitiveness & Growth Full Proposal. Requirements for the Content of the Technical Proposal. Part 3B Product Development Plan

Mission Applications for Space A&R - G.Visentin 1. Automation and Robotics Section (TEC-MMA)

Flight Control Law Development for the F-35 Joint Strike Fighter

Designing a WebGIS architecture for aviation impact assessment

ESA Human Spaceflight Capability Development and Future Perspectives International Lunar Conference September Toronto, Canada

Embedded Training and LVC

Match filtering approach for signal acquisition in radio-pulsar navigation

Unmanned on-orbit servicing (OOS), ROKVISS and the TECSAS mission

GNC/AOCS DEVELOPMENT SYSTEM FOR RENDEZ-VOUS AND DOCKING MISSIONS AT SENER, AND ASSOCIATED TEST FACILITIES

estec PROSPECT Project Objectives & Requirements Document

General Support Technology Programme (GSTP) Period 6 Element 3: Technology Flight Opportunities (TFO)

CPE/CSC 580: Intelligent Agents

COMPASS: Future trends and developments

Human Spaceflight Programmes and Possible Greek Participation

REMOTE OPERATION WITH SUPERVISED AUTONOMY (ROSA)

European Raw Additive Manufacturing Material Procurement For Space Applications

Electrical and Automation Engineering, Fall 2018 Spring 2019, modules and courses inside modules.

Advances in Columbus internal and external payload simulation for EDR and EuTEF

A Simulink/SMASH co-simulation interface Version October 2003

Honeywell Experion System:

Model-Based Design as an Enabler for Supply Chain Collaboration

Modeling and Simulation Made Easy with Simulink Carlos Osorio Principal Application Engineer MathWorks Natick, MA

Getting to Smart Paul Barnard Design Automation

A method to calculate ambient aircraft background noise

CAPE-OPEN and Simulis Thermodynamics enable you to use rigorous thermodynamics in MATLAB. Dr Michel PONS (1), Alain VACHER (2)

ARTES Competitiveness & Growth Full Proposal. Requirements for the Content of the Technical Proposal

Model Based AOCS Design and Automatic Flight Code Generation: Experience and Future Development

Prototyping interactive cockpit applications

1. Introduction. defining and producing new materials with advanced properties, or optimizing industrial processes.

Enabling Model-Based Design for DO-254 Compliance with MathWorks and Mentor Graphics Tools

Volume 2 - Telesat's Solution Ka-band Application APPENDIX 4. Corporate Profiles of COM DEV and EMS Technologies

AIV Platform for the Galileo Precise Timing Facility

Canadian Activities in Intelligent Robotic Systems - An Overview

IABG Space Centre We give you space on earth SPACE

Pathbreaking robots for pathbreaking research. Introducing. KINOVA Gen3 Ultra lightweight robot. kinovarobotics.com 1

- Software Engineer con Laurea Magistrale in Informatica, Telecomunicazioni o Elettronica

ATPE Simulator: Simulation Tool for Onboard GNC Development and Validation

Keywords: Aircraft Systems Integration, Real-Time Simulation, Hardware-In-The-Loop Testing

Real-Time Testing Made Easy with Simulink Real-Time

Pragmatic Strategies for Adopting Model-Based Design for Embedded Applications. The MathWorks, Inc.

Systems Engineering Overview. Axel Claudio Alex Gonzalez

Training and Verification Facilities CGS User Workshop. Columbus Training Facility Team

Digital Systems Design

Aeronautics Supply Chain in Greece Public Private partnership. The missing link Professor Vassilis KOSTOPOULOS

Model-based Design of Coordinated Traffic Controllers

NRC Aerospace and Efforts to Promote Technology Development Partnerships

KPI is one of the oldest and biggest technical universities in Ukraine. It was founded in 1898.

Internet Based HMI in Low-cost Simulators

GIDAS GNSS Interference Detection & Analysis System

Type Approval JANUARY The electronic pdf version of this document found through is the officially binding version

Improving Airport Planning & Development and Operations & Maintenance via Skyline 3D Software

Force Life Management of the F-16 of the Royal Netherlands Air Force. Maj Jacob van der Laan, DMO Marcel Bos, NLR

A Generic Simulink Model Template for Simulation of Small Satellites

Policy-Based RTL Design

The Test and Launch Control Technology for Launch Vehicles

The Virtual Spacecraft Reference Facility

DTP4700 Next Generation Software Defined Radio Platform

Introducing the European Space Agency Architectural Framework for Space-based Systems of Systems Engineering

From Antenna to Bits:

Keysight Technologies MATLAB Data Analysis Software Packages

What is New in Wireless System Design

PREFERRED RELIABILITY PRACTICES. Practice:

Marco Polo: The European contribution

Department of Operations Telecommunications Division

Technology and Manufacturing Readiness Levels [Draft]

A simple embedded stereoscopic vision system for an autonomous rover

OPAL Reactor Training Simulator

STM RH-ASIC capability

How to build an autonomous anything

Systems for Green Operations ITD

Rapid FPGA Modem Design Techniques For SDRs Using Altera DSP Builder

Low-Cost hardware connectivity with Simulink MATLAB-Day RWTH Aachen Sebastian Groß October 24th, 2013

SPACE-BASED SOLUTIONS & ANALYTICS

FOSS in Military Computing

Constellation Scheduling Under Uncertainty: Models and Benefits

DSpace Platform for Speed Estimation AC Slip-Ring Motor in Crane Mechatronic System

TELEMETRY RE-RADIATION SYSTEM

A Power-Efficient Design Approach to Radiation Hardened Digital Circuitry using Dynamically Selectable Triple Modulo Redundancy

Metodo ed Esperienza della CDF in ESTEC

RADAR ANALYST WORKSTATION MODERN, USER-FRIENDLY RADAR TECHNOLOGY IN ERDAS IMAGINE

2015 The MathWorks, Inc. 1

Managing Multipurpose Models in Aerospace M&S Challenges and Experiences

FAA Research and Development Efforts in SHM

Space in the next MFF Commision proposals

STEM Teacher Roundtable

Rheology Solutions Pty Ltd. Focused on providing our customers with materials characterisation solutions through knowledge, experience and support.

Horizon Work Programme Leadership in enabling and industrial technologies - Introduction

Automation & Robotics (A&R) for Space Applications in the German Space Program

UK MOD Policy and ATML

Subject Description Form. Industrial Centre Training I for EIE. Upon completion of the subject, students will be able to:

Best practices in product development: Design Studies & Trade-Off Analyses

Aerospace Software* Cost and Timescale Reduction *and complex electronic hardware

UNIT-III LIFE-CYCLE PHASES

Physics-Based Modeling In Design & Development for U.S. Defense Virtual Prototyping & Product Development. Jennifer Batson Ab Hashemi

Commission proposal for Horizon Europe. #HorizonEU THE NEXT EU RESEARCH & INNOVATION PROGRAMME ( )

Transcription:

MOSAIC: Automated Model Transfer in Simulator Development W.F. Lammen, A.H.W. Nelisse and A.A. ten Dam

Nationaal Lucht- en Ruimtevaartlaboratorium National Aerospace Laboratory NLR MOSAIC: Automated Model Transfer in Simulator Development W.F. Lammen, A.H.W. Nelisse and A.A. ten Dam This report is based on a presentation held at SESP 2002, ESTEC, Noordwijk, The Netherlands, 12-14 November 2002. This report may be cited on condition that full credit is given to NLR and the authors. Customer: National Aerospace Laboratory NLRNational Aerospace Laboratory NLR Working Plan number: 1.2.A.8b Owner: National Aerospace Laboratory NLR Division: Information and Communication Technology Distribution: Unlimited Classification title: Unclassified Issued November 2002

-2- Summary In large (space) programmes the road from feasibility study to operations and/or training may be long and demanding. During a typical programme many application models will be developed and usually different development environments are used during different phases of a programme. The Model-Oriented Software Automatic Interface Converter (MOSAIC) automates transfer of simulation models from MATLAB /Simulink /Stateflow to both EuroSim and to ESA s Simulation Model Portability (SMP) standard. Updates of Commercial Off-The-Shelf (COTS) tools and model standard definitions result in an upgrade of MOSAIC. Developments of MOSAIC are driven by customer requests. The MOSAIC product is developed and maintained by the National Aerospace Laboratory NLR. MOSAIC already is an important ingredient of various space development projects like ATV, SMART-I and GSSF. It has been experienced that the use of MOSAIC reduces costs, time, and effort in simulator development. In this paper the wide applicability and the use of MOSAIC is discussed.

-3- Contents 1 Introduction 4 2 MOSAIC philosophy 5 3 MOSAIC and EuroSim 5 4 MOSAIC and ESA s Simulation Model Portability (SMP) Standard 7 5 MOSAIC in Applications 8 6 MOSAIC Test Bed 9 7 MOSAIC Future Plans & Policy 10 8 References 11 Appendix A List of Abbreviations 12 (12 pages in total)

-4-1 Introduction Within the aerospace industry, simulation models are often developed using Commercial Off- The-Shelf (COTS) tools like MATLAB /Simulink /Stateflow, or, for short, the MATLAB Toolbox [1]. Without tooling, transfer to another simulation environment may require time-consuming adjustment of the simulation model by the model developer. On the other hand, a typical requirement for simulation models is compliance with interface standards because they are software application components that interact with other simulation components. The National Aerospace Laboratory NLR develops and maintains the Model-Oriented Software Automatic Interface Converter (MOSAIC) tool [2]. MOSAIC automatically transfers simulation models from the MATLAB Toolbox to the real-time simulation tool EuroSim [3] and ESA s Simulation Model Portability (SMP) standard [4]. It has been experienced that automatic model transfer with MOSAIC significantly reduces time and cost. The remainder of this paper is as follows. First the MOSAIC philosophy is presented. Second, focus is on automated model transfer to the (hard) real-time EuroSim environment. Third, the generation of models that are compliant to ESA s SMP Standard are detailed. Some applications, the raison d être of MOSAIC, are listed. Finally, the MOSAIC test bed and future plans are given. Fig. 1. MOSAIC place in the simulation model development process.

-5-2 MOSAIC philosophy MOSAIC fits within the simulator development life cycle. The purpose of MOSAIC is to let the model developer take advantage of dedicated model and simulation tools whenever possible. This requires an automatic model transfer tool between simulation environments that avoids the need to change low-level model source code. MOSAIC is such a tool. The design of MOSAIC is based on the philosophy that models are the products of complex design cycles and have been verified and validated with system requirements in mind. Therefore, MOSAIC does not in any way affect the mathematical logic of a model. MOSAIC adds the required interface code to a model, while leaving the internal model logic as a black box. Simulation results before and after a transfer using MOSAIC are identical up to machine level. When applying MOSAIC the powerful MATLAB Toolbox can be used for feasibility studies, model decomposition and verification of model interactions. Subsequently, application model parameter tuning in the target environment, e.g. for optimisation purposes, can be carried out in the real-time simulation environment using e.g. EuroSim or other SMP compliant environments. Changes in model design by the developer should take place in the originating environment, e.g. the MATLAB Toolbox. 3 MOSAIC and EuroSim Since the first official release of MOSAIC in 1999 it is possible to transfer MATLAB/Simulink models to EuroSim. MOSAIC automatically converts model source code that has been generated with the Real Time Workshop (RTW) of MATLAB into model source that can be incorporated in EuroSim (see Fig. 2). Additional EuroSim specific files are also created. Within EuroSim extra steps can be performed to further customise the model and the simulation specific information such as schedules to the user s needs. For example, one can add additional API information, change the simulation frequency, or change tuneable model parameters during simulation.

-6- Fig. 2. MOSAIC multiple model transfer process On special request of the Automated Transfer Vehicle (ATV) project [5] new features have been added to the MOSAIC product. Since MOSAIC Release 4.0 [6] in 2001 it is possible to choose between two ways of model transfer using MOSAIC from the MATLAB Toolbox to EuroSim: Atomic transfer, in this case a Simulink model is transferred as one entity. Multiple transfer, in this case a Simulink model can be divided into subsystems that will appear as submodels within the complete EuroSim model. MOSAIC Release 4.0 can transfer multiple model sources in one step, see Fig. 2. In EuroSim these models can be integrated with each other and with models originating from other sources, e.g. hand-written code. Using multiple model transfer allows models to be instantiated automatically by MOSAIC. Release 4.0 also allows user defined functions, e.g. S-functions, to be either shared by all transferred models or to adhere to a specific model.

-7-4 MOSAIC and ESA s Simulation Model Portability (SMP) Standard The international simulation community is in the process of defining standards with respect to model portability. One of the main objectives is to enable simulation models (source code) to run in any simulation tool, independent of which tool was used for model creation. As a rationalisation of simulation developments ESA has defined the SMP Standard [4]. The main purpose of the SMP standard is to promote portability and re-use of simulation models in different simulations with different simulation environments. NLR has developed MOSAIC Release 5.0 [7] to facilitate atomic model transfer from the MATLAB Toolbox to the SMP standard. Models that have been transferred by MOSAIC Release 5.0 can run in any SMP compliant simulation environment. It features the generation of SMP compliant model source code, including SMP specific files such as a Model Manager, and automatically generated model documentation. The Model Manager initiates publication of the models to the environment and scheduling of the model services within the simulation. As a special feature MOSAIC generates Model Manager routines that perform the SMP compliance test [4] of all transferred models, see Fig. 3. Fig. 3. MOSAIC to SMP transfer

-8- There exists a trend that simulation environments adhere to the SMP standard. For example, EuroSim is capable of running models that are SMP compliant. In this way full advantage can be taken of both the SMP standard and the (real-time) capabilities of EuroSim. 5 MOSAIC in Applications In the last two years the MOSAIC user community has increased significantly both inside and outside the space community. One can conclude that MOSAIC has become an important ingredient of various space development projects. Some examples of its use are: For the Small Missions for Advanced Research in Technology (SMART) I [8] development project ESTEC used MOSAIC to transfer MATLAB Toolbox models to EuroSim. For the Automated Transfer Vehicle (ATV) project EADS-LV [5] uses MOSAIC for multiple model transfer from the MATLAB Toolbox to EuroSim, see Fig.4. VEGA has successfully evaluated MOSAIC for use in the Galileo System Simulation Facility (GSSF) project [9], because it facilitates the generation of SMP compliant models. Evidently, there are more (proprietary) applications. Fig. 4. MOSAIC application in simulator development NLR itself also uses MOSAIC in a wide variety of projects. A typical example is model development in the Affordable Digital fly-by-wire Flight Control System (ADFCS) project [10], executed in NLR's Flight Simulators. The goal of this project is to develop an architecture for affordable fly-by-wire digital flight control systems for small commercial aircraft. ADFCS

-9- partners have created models of flight control systems all over Europe using the MATLAB Toolbox. NLR transfers the MATLAB Toolbox models with MOSAIC for use in the real-time simulation environment ProSim. (ProSim is an instantiation of EuroSim that has been adapted to the NLR flight simulator.) Other MOSAIC applications with the NLR flight simulator are the development of Joint Strike Fighter Simulation Architecture (JSA) project and the GARTEUR- AG 12 project for aerospace vehicle dynamics. Another challenging and rewarding application of MOSAIC is its usage in the development of on-board software for NLR s research aircraft. The wide use of MOSAIC shows it reduces time, effort, and error-prone source code editing. Therefore it is a crucial step within the simulator development process, see Fig. 4. 6 MOSAIC Test Bed Maintaining the quality of the MOSAIC product is one of our goals. This holds for the product itself, as well as all the user documentation, software quality and configuration management. Extensive use of MOSAIC within NLR ensures ongoing extensive testing of MOSAIC capabilities with ever more complex models in an increasing number of applications. At NLR, MOSAIC is also part of a larger model development and test environment for Computer Aided Control Engineering (CACE). Besides the actual simulation software, the CACE environment also facilitates documentation and tools for management and education. Development processes are structured by means of tool chaining in so-called workflows (see Fig. 5), allowing easy re-use of knowledge. The construction of the CACE environment is based on the SPINEware product [11]. SPINEware, developed by NLR, facilitates multiple and remote access to applications over a network through a single graphical user interface.

-10- Fig. 5. MOSAIC embedded in an automated CACE workflow 7 MOSAIC Future Plans & Policy Developments of MOSAIC are driven by customer requests. Updates of COTS tools and model standard definitions usually result in an upgrade of the MOSAIC capabilities. NLR already upgraded MOSAIC twice to new versions of the MATLAB Toolbox. In the near future it will be upgraded to EuroSim Mk3. MOSAIC only requires the user manuals of the COTS tools or standards between which it facilitates model transfer. In 2003 MOSAIC will be enhanced to support a new version of the SMP standard on hierarchical model structures as expected to be released by ESA by the end of 2002. This includes automatic creation of a generic model manager for multiple models; integration of model-specific files into a hierarchical model-tree; optional re-use of an existing Model Manager during each transfer. MOSAIC is available on Linux and SGI. Current work is directed to the release of a Windows version of MOSAIC.

-11- MOSAIC is available as a licensed NLR product [12]. More information can be obtained from mosaic@nlr.nl. 8 References [1] Real-Time Workshop for Use with Simulink, User s Guide, version 4, The Mathworks inc., 2000. [2] W.F. Lammen, A.A. ten Dam, W.M.M. Heesbeen and R. Franco, MOSAIC: Model- Oriented Software Automatic Interface Converter, Proceedings 6th International Workshop on Simulation for European Space Programmes SESP 2000, 10-12 October 2000, ESTEC, Noordwijk, The Netherlands. Also NLR Technical Publication NLR-TP- 2000-571, 2001.(URL: http://www.nlr.nl/public/library/new/2000-571-dcs.html) [3] EuroSim product information: URL: http://www.eurosim.nl. [4] Simulation Model Portability Handbook, EWP-2080, issue 1, revision 0, 2000/10/16, European Space Research and Technology Centre ESTEC, Noordwijk, the Netherlands. [5] J. Pérennec, Automated Transfer Vehicle Simulation Software Development, Proceedings 6th International Workshop on Simulation for European Space Programmes SESP 2000, 10-12 October 2000, ESTEC, Noordwijk, The Netherlands. [6] W.F. Lammen, A.A. ten Dam and A.H.W. Nelisse, MOSAIC Release 4.0 User Manual, NLR-TR-2001-437. [7] W.F. Lammen, A.A. ten Dam and A.H.W. Nelisse, MOSAIC Release 5.0 User Manual, NLR-CR-2001-237. [8] The SMART 1 Project Test Bed, Proceedings 6th International Workshop on Simulation for European Space Programmes SESP 2000, 10-12 October 2000, ESTEC, Noordwijk, The Netherlands. [9] I. Shurmer, M-E Bégin, P. Fritzen, A. Pidgeon, C. Seynat and R. Franco, Galileo System Simulation Facility - Phase 2; ESA/ESTEC, Netherlands, unpublished. [10] BE97-4098 Annex 1 Project programme ADFCS, September 29, 1997. [11] SPINEware product information: URL: http://www.spineware.com. [12] MOSAIC product information: E-mail: mosaic@nlr.nl.

-12- Appendix A List of Abbreviations ADFCS ATV ESA ESTEC CACE COTS EuroSim GSSF JSA MOSAIC NLR RTW SMART SMP Affordable fly-by-wire Digital Flight Control Systems Automated Transfer Vehicle European Space Agency European Space Research and Technology Centre Computer Aided Control Engineering Commercial Off-The-Shelf European Real-Time Operations Simulator Galileo System Simulation Facility Joint Strike Fighter Simulation Architecture Model-Oriented Software Automatic Interface Converter National Aerospace Laboratory NLR Real Time Workshop Small Missions for Advanced Research in Technology Simulation Model Portability