Software Supports MIL/Aero Applications

by Rick Nelson, Executive Editor

Software tools are critical in supporting the design, development, manufacturing test, deployment, and maintenance of mission-critical military and aerospace systems. In addition, software is invaluable in adopting new test hardware—often commercial off-the-shelf (COTS) hardware based on the PXI architecture—to support legacy systems. And, software tools can help extend the life of legacy test systems.

Several vendors offer a variety of tools that support these various aspects of a successful MIL/Aero application. MathWorks, for example, provides technical-computing, design, and simulation software tools that can run on the desktop or work with a selection of modular hardware platforms, including hardware from its partner Speedgoat as well as industry-standard PXI platforms and benchtop LXI or GPIB instruments.

Agilent Technologies, Geotest-Marvin Test Systems, and National Instruments make software tools as well as the modular hardware platforms that can be integrated with the software. The tools from these companies have been used by organizations including Astronics DME, BAE Systems, Edwards Air Force Base, Embraer, Lockheed Martin, and NATO in MIL/Aero applications.

Technical Computing

Jon Friedman, manager of aerospace and defense and automotive industry marketing at MathWorks, described how tools from MathWorks support different engineering workflows. One workflow is what he called the technical computing flow, which generally is served by MATLAB. The other flow involves simulation and design, often called model-based design. Customers following that path typically use Simulink although Friedman cautioned that there is not a boundary between the two tools, and engineers often model in MATLAB and use Simulink to create data processing algorithms.

Addressing technical computing, Friedman said engineers have data that they want to analyze and share. He described MATLAB as a fourth-generation language that allows engineers who are not programming experts to smooth, filter, and otherwise process the data they have collected. The data, he said, can be imported from many file formats including spreadsheets, or, using the company’s Instrument Control Box, it can be acquired directly from instruments.

Figure 1. Block 40 Global Hawk at Edwards Air Force Base
Courtesy of Northrop Grumman

Friedman described a MIL/Aero application for MATLAB.1 Engineers at Edwards Air Force Base, he said, needed to conduct extensive performance flight tests on the Global Hawk unmanned aerial reconnaissance system (Figure 1). After test flights lasting as long as 30 hours, the engineers must analyze up to 20 GB of data before the test craft can be cleared for its next flight. Using a 30-year-old Fortran-based application called Uniform Flight Test Analysis System (UFTAS), the process could take days. Further, the legacy UFTAS was difficult to configure for new aircraft programs.

As a solution, the engineers developed MUFTAS, a MATLAB-based version of UFTAS. Using MathWorks’ Parallel Computing Toolbox and MATLAB Distributed Computing Server, they deployed the MATLAB-based version on a computer cluster with 16 dual-core, 2.61-GHz AMD Opteron processors, allowing them to analyze flight data overnight and promptly resume flight tests.

Model-Based Design

Addressing the simulation and design path, Friedman said model-based design enables early verification and even validation. Traditionally, he continued, system engineering involved pursuing better documentation and interface specs, but real systems integration testing could not take place until the components became available. He noted facetiously, “We used to say beautiful systems engineering always got spoiled when the parts arrived.”

That’s no longer the case, he said. With model-based design, engineers can develop requirements models and design models and even create test vectors to drive those models. “You can start simulating and testing long before any of the parts show up.”

When parts do begin to arrive, engineers can move on to hardware-in-the-loop (HIL) test or, as an intermediate stage, software-in-the-loop (SIL) test. “Customers will generate the embedded code they plan to deploy. They then can put a wrapper around that code and test it on the host or use cosimulation technology in the case of HDL. They can then take that same code and deploy it to hardware in an HIL system and reuse their test vectors.”

Friedman said that MathWorks works with many hardware vendors; the company’s tools generate ANSI C that can be ported to most platforms. For example, Speedgoat, a MathWorks partner, makes hardware that runs MathWorks-generated code in real time.

Friedman described a Simulink application. The Royal Navy’s Type 45 Anti-Air Warfare Destroyer requires an on-board trainer (OBT) to train crew to control, reconfigure, and recover the ship’s main systems in a variety of fault and damage scenarios. The trainer simulates, in real time, the Type 45’s electric propulsion, generation, and auxiliary machinery, which can be simultaneously controlled by up to 16 crew members via the Platform Management System (PMS).

BAE Systems Surface Ships engineers were charged with developing the OBT plant simulation before the systems it emulated were fully designed. To make the OBT realistic, the BAE team needed to activate approximately 4,000 inputs and outputs to PMS with many more internal signals to interconnect the 16 systems it emulated. In addition, the BAE team needed to deliver efficient code that would require an average of less than 20% CPU utilization on a 2-GHz processor.

To meet such requirements, the BAE engineers created Simulink models of the Type 45’s main physical systems, including shafts, propellers, gas turbines, and diesel generators as well as electric distribution, steering, bilge, high-pressure seawater, and fire-fighting systems. In addition, the BAE team was able to assist in the development of the PMS.

“In the first stages of development, our simulation results raised issues around how the PMS interacted with certain systems,” said Peter Worthington, engineering manager at BAE Systems Surface Ships.2 “The MATLAB and Simulink models helped us communicate what was happening to the Type 45 project team. That knowledge enabled them to review the PMS, identify and address specification issues early, and avoid costly rework later on.”

In addition to MATLAB and Simulink, MathWorks also offers a tool called Polyspace, which uses formal methods-based abstract interpretation techniques to verify the absence of conditions such as divide by 0 or overflow. Friedman said the NATO HAWK Management Office (NHMO) used Polyspace for Ada to improve the HAWK surface-to-air missile system by identifying and eliminating causes of run-time errors.3

Software and COTS Hardware

National Instruments focuses on both hardware and software. Carl Heide, market developments manager for aerospace/defense at National Instruments (NI), said NI’s approach is to build COTS products that include software-defined instrumentation in the PXI and NI CompactRIO modular platforms. “We make software development tool chains whose purpose is to integrate these modular platforms into complete solutions for measurement and control applications.”

Heide emphasized, “CompactRIO is especially important to keep in mind because it’s really suited for control and monitoring systems. For example, you can monitor a machine and perform condition-based maintenance. You do maintenance when appropriate for the condition of the machine—not because a certain amount of time has passed.”

Figure 2. CACI CBATS Model 201
Courtesy of National Instruments

NI hardware and software products have been used in a variety of military test applications, Heide said, such as the NI CACI CBATS (Common Benchtop Automatic Test System). The CACI Model 201 Test Set (Figure 2), for example, includes PXI instruments and a controller installed in a chassis with a common connector interface to the UUT.

Heide said one recent focus of NI has involved FPGA-based modules, for which users can write their own code to customize software-defined instruments. Such instruments, he said, can operate time-deterministically—a feature important in HIL test. Heide noted that LabVIEW FPGA makes the customization of an FPGA-based instrument easy. While you might need a hundred lines of HDL to implement a counter in an FPGA or a couple of thousand lines of HDL to implement a DMA operation, with LabVIEW FPGA you manipulate a few icons—just as you would if your target were a standard microprocessor.

Figure 3. NI VeriStand Real-Time Testing Software Tool
Courtesy of National Instruments

NI VeriStand (Figure 3) is another key tool. Heide said, “NI VeriStand was designed specifically for embedded software validation and closed-loop control applications. It is a core component in our toolset for HIL. It integrates with [several other tools]: LabVIEW so you can leverage the development environment; TestStand for test automation; and DIAdem software, which is useful for data post-processing.” He added, “The VeriStand software also is flexible so you can integrate a variety of modeling and simulation environments because it has the capability to run a number of compiled models.”

Iron Bird Simulator

Embraer S.A. is one company employing NI VeriStand. Guilherme Seelaender of Embraer said, “We selected NI VeriStand for the Legacy 500 Iron Bird because of the breadth of functionality the environment provides out of the box, which significantly reduced our development efforts.”4 The Iron Bird is a full aircraft simulator that consists of 21 real-time PXI systems networked together using reflective memory and Ethernet interfaces, Seelaender explained.

Other companies using NI hardware and software for military applications include Lockheed Martin STS (LM STS) and G Systems. LM STS chose NI tools for its LM-STAR test system for the Joint Strike Fighter/F-35 program. Robert Dixon of Lockheed Martin STS said, “The innovative LM-STAR approach to standardized test system development based on COTS software has yielded many cost-saving benefits for LM STS, harmonization suppliers, and the U.S. government.”5

The LM-STAR system is an open software architecture based largely on NI TestStand and LabWindows/CVI. It supports the seamless transition of test systems from the factory to the field.

For its part, G Systems was contracted by Lockheed Martin Aeronautics to construct an F-35 Vehicle Systems Integration Facility (VSIF) to monitor aircraft subsystems integration tests. The VSIF system was distributed across several servers to enable load balancing and achieve the required system performance. The distributed software architecture, which included six major custom applications, provided for future expansion of the system. Michael Fortenberry of G Systems said, “Through the use of advanced software architecture and NI hardware, G Systems was able to provide Lockheed Martin Aeronautics with a highly configurable, expandable system to meet current and future requirements of the F-35 VSIF.”6

Fortenberry added, “We performed analog and digital data acquisition using five PXI chassis populated with a variety of NI data acquisition boards to achieve a system total of 640 analog channels and 480 digital channels. The capability to mix-and-match different types of DAQ boards while maintaining time synchronization was important to control the overall hardware costs for the system. The system maintained the time synchronization through the use of an IRIG time signal provided by VSIF data acquisition or another source within the VSIF lab. The system used this time source to provide the start pulse and 10-MHz clock, which was routed through the PXI-6653 synchronization boards to each PXI chassis.”

Heide from NI also addressed migration issues. He said that users can help themselves in this regard by planning ahead and establishing abstraction layers. You might, for example, think in terms of “DMM,” where your abstraction layer is the type of instrument. Or you might think in terms of “volts” where the abstraction layer is the measurement, independent of the instrument (DMM or digitizer, for example). He noted that NI tools do not enforce this abstraction layer approach because it is up to the customer to define abstraction layers as appropriate. One customer in an energy-related field, he said, even defined what it called a “state-machine” abstraction layer.

As for specific tools that NI offers for legacy system upgrade, Heide cited NI TestStand test-management software, which supports multiple development environments. NI TestStand works with NI LabVIEW and NI LabWindows/CVI as well as legacy programming languages such as ATLAS.

“If you want to support multiple versions of these environments,” he said, TestStand does that for you. If you are using TestStand, it’s possible to upgrade a system with new TPSs [test program sets] using new versions of LabVIEW or CVI while still operating the legacy TPSs that were written using the older versions.”

From ATEasy to DtifEasy

Mike Dewey, senior product marketing manager at Geotest-Marvin Test Systems, explained Geotest’s software offerings for MIL/Aero applications: “For program development we offer ATEasy, a test executive and program development environment. ATEasy combines the ease of use associated with Visual Basic with the flexibility of Visual C++, creating a complete object-oriented, 32-bit Windows programming environment. In addition, the command structure of ATEasy is very similar to ATLAS, making it easy for users familiar with the ATLAS language to transition to ATEasy. ATEasy also imports ATML test descriptions and exports ATML test results.”

Figure 4. ATEasy Test Executive and Program Development Environment
Courtesy of Geotest-Marvin Test Systems


He noted that ATEasy (Figure 4) is control-bus-agnostic and can be used with virtually all instrumentation control buses including GPIB, VXI, PXI, LXI, and USB.

Geotest supports program migration as well as program development. “For digital test program migration from legacy tester platforms,” Dewey said, “we offer our DtifEasy package, which allows users to migrate LASAR-based test programs from legacy testers to a modern, PXI digital test platform.” DtifEasy, he explained, can convert .tap files (as defined in IEEE 14457) from a legacy test application or a resimulated test application for rehosting on a PXI digital platform. All .tap files are supported providing go/no-go, fault dictionary, and guided-probe capabilities. Geotest also offers a digital waveform display/edit tool called DIOEasy, which supports all of our digital instruments.”

Dewey said ATEasy software is used in conjunction with flight line, I-level, and depot-level test systems for testing various missile systems including AMRAAM, Maverick, Sidewinder, Paveway, Hellfire, and TOW. ATEasy also supports testing of a variety of systems on platforms such as F-16, T-50, FA-50, AH-64, AH-1, SH-60, Tigre, F-18, and F-35. In addition, Geotest’s DtifEasy software has been used successfully to migrate legacy test programs from the TETS and Viper/T platforms to a PXI platform.

Geotest and Astronics DME have jointly developed a demonstration test platform that shows how PXI can be used to downsize the current VXI-based Viper/T test platform. Dewey said that using Geotest’s DtifEasy and the GX5960 PXI digital subsystem, DME and Geotest successfully demonstrated the conversion and rehosting of a legacy test program from the Viper/T to a next-generation PXI platform.8

Seamless Integration

According to John Hansen and Greg Jue, application engineers at Agilent Technologies, “Agilent supplies test and measurement equipment and systems as well as design and simulation software solutions. To enhance the usability of our instruments, we offer many different types of software tools that enable the user to quickly simulate and analyze analog, RF, and digital signals needed for MIL/Aero design and test applications.”

They explained, “Agilent provides seamless integration between hardware measurement and software solutions. For example, our vector signal analysis (VSA) software operates with many of our measurement platforms such as logic analyzers, oscilloscopes, signal analyzers, and modular test instruments as well as our design and simulation software tools. Waveforms can be created in simulation and downloaded to test equipment such as precision wideband arbitrary waveform generators to create test signals. Additionally, signals can be captured from Agilent test equipment with our VSA software and read back into simulation for further post-processing.”

Agilent offers a full line of modular instruments in PXI and AXIe form factors and software products that enable and enhance usability. For example, said Hansen and Jue, “The PXI modular platforms include soft front panels to easily verify connectivity to a module and make immediate measurements. This feature is especially useful during software development and debug and for performing benchtop measurements with one or more modular devices. Reduced development time is made possible with the supplied drivers and programming examples in Visual Studio (VB NET, C#, C/C++), VEE, LabVIEW, LabWindows/CVI, and MATLAB, enabling integration into existing environments. The PXI form-factor also conforms to government requirements for modular system design.”

Hansen and Jue said Agilent has an ongoing technology refresh program that offers services and solutions to assist customers in refreshing their existing test and measurement applications.

In addition, they said, “Many Agilent instruments support backward-compatibility of software command sets for legacy instruments to minimize the impact on automated systems. Agilent instruments can be utilized in many programming environments. “Command Expert,” they said, “is a software application for fast and easy instrument control in many PC environments.”

According to Hansen and Jue, Agilent hardware and software products are used extensively throughout the military and aerospace industries including programs with strict change-control requirements. “We work closely with these customers to ensure that our products conform to the technical and security requirements of the individual program,” they said.

Specific projects have involved single and multiple emitter simulation; nonlinear modeling and analysis of components using X-parameters; generation and analysis of custom or proprietary OFDM modulated waveforms for design and test applications; and comprehensive design, simulation, and analysis of MIL/Aero radar, EW, and communications systems. They concluded, “Combining design and simulation software with Agilent test equipment has lowered costs and improved flexibility when using COTS equipment in MIL/Aero test applications.”

Useful Models

Design, simulation, and test-program development tools certainly are a boon to MIL/Aero applications. Of course, simulations are only as good as the models you use.

Friedman at MathWorks cited a quote from George E. P. Box, professor emeritus of statistics at the University of Wisconsin, who said, “Essentially, all models are wrong, but some are useful.” Fortunately, we can employ the useful models to move as much of the development process as possible from the hardware prototype stage to the desktop. As Friedman concluded, “I don’t want to fly in an airplane for which the manufacturer never built a prototype, but I also don’t think that prototype airplanes are where you want to perform the initial debugging of the software.”

References

1. “Edwards Air Force Base Accelerates Flight Test Data Analysis Using MATLAB and Parallel Computing,” MathWorks, User Story.

2. “BAE Systems Surface Ships Develops On-Board Trainer Plant Simulation for Royal Navy,” MathWorks, User Story.

3. “NATO HAWK Management Office Accelerates Analysis of Mission-Critical Applications Using Polyspace Products for Ada,” MathWorks, User Story.

4. Seelaender, G., “Embraer Performs Full Airplane Simulation Using NI HIL Tools,” National Instruments, Case Study.

5. Dixon, R., “LM-STAR NI Software-Based Test System Saves Millions,” National Instruments, Case Study.

6. Fortenberry, M., “Building an F-35 Vehicle Systems Integration Facility (VSIF) Data Acquisition System,” National Instruments, Case Study.

7. Yazma, R. and Quan, A., ”IEEE-1445 (DTIF) Based Digital Test Solution,” Proceedings of Autotestcon, IEEE, 2009.

8. Ginn, J. and Dewey, M., Next Generation Test System Architectures for Depot and O-Level Test, Geotest-Marvin Test Systems, White Paper.

For More Information

Agilent Technologies

Astronics DME

CACI International

Geotest-Marvin Test Systems

MathWorks

National Instruments

Speedgoat

 

 

 

 

 

Click below to request this free DVD and learn more about the latest applications and technologies

  • A one-stop resource for engineers in electronic functional test, wireless and microwave
  • Access to the complete PXI product and application library including datasheets, videos & more