Collaborative Simulation and Integration Environments | Next Generation Aerospace Series - MATLAB & Simulink
Video Player is loading.
Current Time 0:00
Duration 23:45
Loaded: 0.70%
Stream Type LIVE
Remaining Time 23:45
 
1x
  • Chapters
  • descriptions off, selected
  • captions off, selected
  • en (Main), selected
    Video length is 23:45

    Collaborative Simulation and Integration Environments | Next Generation Aerospace Series

    From the series: Next Generation Aerospace Series

    Overview

    How are you making sure that your systems are matching functional and performance requirements?

    New systems are more complex and software defined. Development benefits from an integrated model-based systems engineering workflow, and continuous system testing at all stages.

    New programs define environments where collaboration is key for early proof-of-concept simulations for demonstration but also to reduce risk, cost, and late error detection - simulation and process digitalization is a solution.

    When you are designing complex systems requiring the involvement of many teams, you can leverage digital twins and test embedded hardware with Speedgoat HIL simulators. Automated and continuous testing of your controllers and controls systems enables you to deliver high-quality systems quickly, and cost-effectively.

    Highlights

    This 1hr presentation will show you the unified testing solution by MathWorks and Speedgoat and how it can support your system engineering and testing workflows:

    • Anticipate constraints, validate assumptions and make informed design choices in early stages
    • Turn bug fixing into early V&V and best practices adoptions
    • Accelerate the readiness of your requirements, models and testing environment
    • Enable different levels of fidelity to consolidate Architecture and Functional Objectives
    • Enable early integration and testing of critical or extreme conditions

    About the Presenter

    Alexandra Beaudouin is the Aerospace and Defence Industry Manager for the EMEA region at MathWorks. Her technical background is on software development for certified critical systems in aerospace and defense.

    Prior to MathWorks, Alexandra was Senior Engineering Manager in Embedded Software at Thales in France, and an Embedded Software Department Manager with a focus on DO-178B/C development projects and certification activities at SOLENT, as French company specialized in System and Software engineering in Aeronautics.

    Pablo Romero, Application Engineer at MathWorks, specialized in real-time simulation and testing. Holds a master in aeronautical engineering from Univ Politécnica de Madrid.

    Prior to MathWorks, Pablo worked at BMW and Airbus. At BMW, he has gained experience in modelling, simulating and implementing real-time systems e.g. for HIL systems. At Airbus, he worked as a flight dynamic engineer for the Eurofighter Program.

    Yves Gerster is the Aerospace Industry Manager for Speedgoat willing to further push the boundaries of innovation in Aerospace though design and test methods. Prior to Speedgoat, Yves was an Aerospace Engineer who worked for multiple OEM’s such as Pilatus Aircraft in Switzerland. He took on several engineering roles in mechanical systems engineering, testing and certification

    Recorded: 7 Apr 2022

    Hello, everyone, and welcome. My name is Alexandra Beaudouin, and I'm an aerospace and defense industry manager for MathWorks works the EMEA region. In today's presentation, we'll talk about the challenges of the next generation programs in aerospace and defense, and the solution that MathWorks can bring to answer those challenges. I will be jointly presenting with my colleague Pablo Romero Cumbreras, application engineer in MathWorks specialized in real-time simulation and testing. Hello, Pablo, and welcome.

    Hi, Alexandra, and good day everyone. Thanks for having me today to talk about such an exciting topic. I'm looking forward to the discussion. I hope that my experience in the aerospace industry as developer and years of experience in the tools on future matters for simulation, collaboration, and integration of different software and hardware components, will come on a matter of interest and usefulness for all our attendees. Thank you.

    The reason we will talk about simulation today is because we see that systems are becoming more and more complex and technologically ambitious. In order to put the best chances of success in a short time frame, simulation is an asset to anticipate and reduce the risk of several challenges. When we talk about simulation, we need to assess the representative of the simulation either for the system behavior or the simulated environment.

    This is why we ask if Yves Gerster from Speedgoat, our partner company, to join us today. Speedboat offers the real-time hardware which integrates seamlessly in our simulation solution. Hi, Yves, and welcome.

    Thank you, Alexandra, and hello, everyone. I am happy to talk about the hardware part, and therefore, the link to the real world of the simulation environment in aerospace.

    Today, we will focus on aerospace and defense industry and the challenges raised by some programs or initiatives for innovation such as human landing program in space, alternative propulsion solution in aeronautics, all the next generation defense programs. The complexity of the technology use has increased significantly. Programs are multi-company programs with multirole system design where faster response are required with more and more data.

    Integration through simulation becomes crucial to the program's success. Systems are more software defined and use data to improve the system's performance that implies to have a bigger picture of all the systems. Technology is enabling better communication channels in order to be more collaborative in operation, but also during the development phases. All this combined, having a validation environment, including simulated items and also a representative hardware in the loop can increase drastically the readiness of the systems in an efficient integration.

    With those larger programs, we have to see beyond individual system design. For example, an aircraft is not only a standalone aircraft system but is a contributor in the larger system of systems. System needs to be considered with more and more real-time capabilities, each system needs to be connected, acquiring, processing, and exchanging data in a secured manner with the other systems.

    The other systems involved are not only the ground stations but they are unmanned systems for research or observation purposes, or they are manned systems like fighter aircraft or a helicopter. We need to be able to process the increased number of software contents requiring more processing capabilities and involving more complexity, longer design, and certification timings still in the loop.

    Programs gather and get to rely on more technologies. And they need to be able to combine those technologies with proof of concepts or demonstrators directly in a new environment built for collaboration and integration. Indeed, more and more demonstrators are built in a virtualized and simulated environment to accelerate the feasibility demonstration. A program lifecycle involves a set of phases that are defined and each contributing to the success of those programs.

    Considering the new challenges and the next generation programs, collaboration from development to operation phases becomes crucial and necessary in the program development. If we have a look to the different levels of engineering, from the machine to the components development, let's try to see the interaction and the connection between them.

    The levels of the pyramids become concentric circles and the engineering involving different levels of perspectives. From the outside circles to the inner circle, they all have their own cycle of development. The definition of the first activities at the machine level will enable to define the global requirements. Then, the high level perspective enabled us to define the system of system architecture and establish the role of each system to achieve the machine purpose. Then, defining more in detail the system objectives and their own architecture into components down to the component requirements implementation and verification.

    Once the top-down definition and implementation is done, you enter the integration and validation phases. The defined requirements acts as an executable specification and become the functional reference for the integration and the implemented solution. The verification activity needs to comply with the requirements definition of the upper level. It helps, also, to assess the architectural choices and evaluate the reusability and the maintainability of the implementation.

    Lastly, it will enable an early validation reducing risk and expensive tests on large benches or even on the field. When we look at each circle one by one, that means each level of perspective and granularity. We can identify and refine the different objectives and challenges. At the machine level, it's the global purpose of the program that is at stake. That means also identifying the protagonists involved and the needs in terms of responsibility of each of them without forgetting the performance aspects for the response timing and the real-time objective so that the machine makes sense to do the operational requirements.

    For example, let's think about a rescue mission. The aircraft needs images to be able to perform the search phases of a rescue. The images come from the different UAVs through the satellite. Defining the timing constraint within which the images are required and provided to the aircraft becomes crucial so that the aircraft can search in the right place and the right time.

    After establishing those mission requirements and the required protagonists, let's have a look at how they need to behave, one, with each other. that is establishing the construct and the responsibilities of each of them, but also their interfaces and how they interact. And then, check if there are no conflicting requirements between them. This is where simulation can help you investigate and check through hundreds of scenarios if your assumptions, allocation, and responsibilities can be confirmed are need rework.

    This is also where you can evaluate how your system of systems needs to allocate and balance the requirements, particularly in terms of performance and timing so that the global objective can be met. For this level, down to the lower ones, the individual responsibilities usually also align with the contractual and industrial organization of the program. So clarifying those boundaries and interfaces clarifies, also, the constructional scopes.

    As each single system is developed by different actors in the industry, using model-based design enables to have a common ground between them. Either within the same company or even over different companies, the teams have something to discuss and to exchange. The models, whatever their level of fidelity, can be safely exchanged through protection mechanism so that you can get a good enough representation of your systemic system, and the exchange models provide the functional behavior without sharing any design or implementation information. That means, without sharing your intellectual property.

    The development of your own system using model-based design can be deployed in a continuous integration flow and combined with a shared model and add high-level value to your continuous integrated systems during the maturation and validation. At the component level, you can have the same kind of environment refined at the architecture and implementation level, and you still can leverage all the advantages of the model-based development and the simulation. That means having a representative maturation and validation environment integrated in a continuous integration workflow, enabling the functional evaluation and the performance assessment of your subsystem or component.

    Thank you, Alexandra, for the insightful information about how we can break up technical and collaboration challenges into our systems hierarchy that would allow us to address the problems in different ways based on every scenario and requirements. We can see how we, at MathWorks, have collaborated and supported our aerospace and defense customers all over the world to succeed at multiple levels of collaboration and integration as you have explained from high-level mission studies and planification down to the very detailed development and testing.

    This is only a small fraction of the projects we have supported, and I can say I feel very proud and fulfilled to have participated and helped customers in some of them. We can see how the collaboration, not only between companies that are in the aeropsace industry, but also external expert, suppliers, and service providers, are a key to succeed and something to consider very seriously for those upcoming challenges that the generation of aerospace projects may involve.

    Let's focus on the first two levels of the previous hierarchy, mission and system of systems. I will briefly explain how MathWorks tools and experience can be used at the early conceptual design phases to evaluate different concepts and enable better decision-making.

    Banking on a strong representation of our system of systems is a great method for mission analysis and other early studies such as defining interfaces, requirements, and certain attributes for the components within the main system that will enable to capture some details about the global system architecture and improve decision-making due to the execution of early trade-off studies and objective evaluation of the impact of different architectural options.

    In this example, we say how Airbus Defense and Space has used MATLAB and Simulink to model and simulate the diverse mission scenarios in a few minutes to help engineers gain a better understanding of their system and take better decisions about the system configuration in the preliminary design review of the Jupiter's moons mission. This provides evidence that modeling and simulation is not only useful during the detailed design phase, but can and should be used up front to accelerate the design and evaluation of different requirements and system configurations by orders of magnitude.

    Also, we can start from different simulation scenarios and look into the system architecture from another perspective. I use System Composer for system and architecture design and analysis. This will facilitate collaboration between different companies or teams by defining a common collaboration platform so that stakeholders agree on requirements and interfaces involving heterogeneous systems working on a single framework.

    This methodology reduces friction in later design phases and enhances collaboration and visibility of requirements and properties essential for the fulfillment of the machine and success of the global system. After an initial overview about the upper levels of collaboration and systems integration, we can dive into detailed systems definition, behavioral models, and simulation at different levels of fidelity.

    We consider, very often, that stimulation must be as accurate and closer to the reality as possible. However, the fact is that stimulation is useful in a broad range of use cases and levels of fidelity starting with low fidelity models, also call behavioral models. These models don't require much modeling effort, but for its relatively low engineering investment, behaviorial models can capture and represent requirements in a way that support system engineers to map requirements through solution neutral functions and those enable whatever studies and supporting in general architecture design decisions.

    These models can be evolved or be complemented by further models with increasing levels of fidelity, which may come from different sources, have different variance in terms of parameterization or complexity, and be implemented in different tools or languages. In any case, we can capture all levels of fidelity in a progressive and continuously evolving simulation environment to address the needs of different teams that may even work in other companies or countries and still need to exchange information and collaborate for the sake of the successful final design of the main system.

    To enable trust in collaboration in a heterogeneous development framework, we should consider Simulink not only as one development environment more, but also as the ultimate integration platform. Simulink is used worldwide by engineers to design and test algorithms controls and plot models with a level of detail that can be used right for production purposes.

    Although this may be the most well-known use case, Simulink can import models from third-party tools and source code in different languages such as C, C++, and HDL, so that engineers can, in addition to Simulink's traditional strengths, build an integration platform where they can pull all systems and components separately expected in larger space projects involving teams with different preferences tools and methodologies, but who require, at some point of the development, to convert to a common platform where they can integrate, simulate, and test all their work regardless of their source.

    This integration work in a virtual environment like Simulink has the great advantage of being extended towards real-time simulation, hardware, and interaction with physical devices and components. This way, we can leverage all our integration efforts as well as the continuous workflow we have created, starting from system definition and requirements to accelerate the time to market and reduce costs derived of undesired iteration due to issues discovered late in the development phase, expensive and complex prototypes untested in the field.

    As announced at the beginning, we are joined today Yves Gerster from Speedgoat, who will elaborate more about that advantage of real-time testing. Yves, why real-time testing is so important?

    Thank you, Pablo. At the end of the day, or at the end of the project, you want to have a real-world system which is running decoupled from any simulation. Therefore, it is important to have a seamless transition from desktop simulation to real-time testing. That's where Speedgoat comes into play. After successfully setting up your simulation environment, you can use the same models you used for simulation directly for real-time testing. You can connect your hardware controller to the digital world using the Speedgoat target machine.

    This allows you to set up a hardware-in-the-loop testing environment with minimal additional effort. The seamless transition between simulation and hardware testing allows you to test your controller earlier during the development phase. Now, let's look into how your development process can benefit from real-time simulation and testing. We like to use this simple diagram where you want to move from left to right from an early design development in production.

    As a simple example, let's take a flight controller for your new vertical takeoff and landing vehicle. If you design your controller in Simulink, you can easily perform desktop simulation. But to verify that your design works with the real hardware, you should support this with real-time testing for rapid controller prototyping, or RCP in short. From your same Simulink model, you can generate the real-time application and deploy it to a Speedgoat target machine directly from within Simulink.

    Our aerospace I/O modules let you connect the target machine to the real plant to test and tune your algorithm. To test the flight controller, you can then use the real-time target machine for hardware-in-the-loop testing. To safely test your flight controller, you interface it with a real-time digital twin model of your plane. The real-time simulation and testing platform using MATLAB and Speedgoat simplifies your workflow and allows you to design and test better controllers faster.

    You can innovate, and you are not constrained by embedded testing environments or with hassles of integrating solutions. Speedgoat is supporting all key protocols for the aerospace industry, such as ARINC 429, AFDX, MIL standard 1553. Overall, we support over 200 I/O modules and protocols. We also have a general purpose protocols such as SPI, I2C, ethernet, and the vast range of analog and digital protocols.

    You can leverage all the relevant I/O protocols at the requested voltage level and sampling rate and even perform fault insertion. Those I/O interfaces can seamlessly be integrated into the workflow from MATLAB and Simulink. I am sure that you can find the communication protocol which is relevant for you somewhere in these categories. That said, I would like to show you an example on how Gulfstream leveraged speed boat for hill testing.

    Gulfstream connected their actual controller, the full authority digital engine controller, short FADEC, to a Speedgoat hardware-in-the-loop system for testing. The value of field testing for Gulfstream is that it not only allows them to save costs during the expense of flight testing, but it enables them to do fold insertion and edge cases analysis without the concern of damaging their aircraft or injuring the test crews.

    Thanks a lot, Yves and Pablo, for this inspiring information. We have seen that MathWorks and Speedgoat solution can be very helpful, and even sometimes crucial, at every level of the program development. Positioning the model of your system at the center, we have seen, during this presentation, that you can mature, define, and validate the functional behavior, manage the model at different levels of fidelity depending on where your focus of interest is, include or connect your model with third-party tools, include all the languages or code libraries to complement your model, prototype or test your generated code on a hardware board, and evaluate your system with a real-time platform.

    All this in MATLAB and Simulink environment completed by Speedgoat solution for the real-time testing. As a conclusion, we have seen that MATLAB and Simulink enables a collaborative environment, allows an early and formal validation process, can give you a large-scale vision, and globally accelerate the readiness of your development.