HIL testing simulates real-world conditions in a virtual environment to test complex components before physical products are made.
As manufactured products become more complex, testing of these products becomes more challenging. Full-scale testing of assembled products can be costly, as manufacturers build finished prototypes and attempt to test every conceivable scenario before deploying the product. The testing of finished products also has limitations, potentially delaying the identification of flaws that could have been identified if testing had been conducted earlier.
The increased role of computer software in manufactured products has added to the complexity of product testing. For example, automobiles and aerospace vehicles contain numerous electronic control units (ECUs) that handle various functions and input/output (I/O), making comprehensive testing difficult.
Software can also become part of the solution, as computer-based simulation offers ways to perform virtual testing. Instead of physically testing finished prototypes, manufacturing and engineering teams can simulate actual conditions and perform testing on digital models of products, saving time and expenses. While this helps accelerate testing, software simulations sometimes fail to identify issues encountered in physical testing.
To handle testing complexities, manufacturers often turn to hardware-in-the-loop (HIL) testing, which simulates real-world conditions for assembled products or components. By connecting a controller to a system simulating the operation of the product in real-world conditions, product teams can test products early and often in the design cycle while keeping testing as realistic as possible. HIL testing essentially replaces a physical model, such as an automobile transmission system, with a virtual representation of that system that simulates the physical model.
HIL process overview
HIL testing typically connects real controller hardware with a simulated system via a combination of analog and digital I/O connections. The simulated system is typically a mathematical model of the actual system. The model is executed in real time to simulate the behavior of the actual system based on inputs from the controller hardware.
Specific HIL testing processes vary by industry and application, but in general, the process starts with creating a digital model that simulates the physical system. The model may be created with commercial or custom-built software capable of modeling electrical, mechanical and hydraulic components as well as physical behavior related to fluid mechanics, thermodynamics and other physics-based specialties.
Before connecting the system to the controller, the model is typically run on a test system to simulate responses from controller hardware inputs and verify the system reacts in a realistic manner. Once verified, the controller hardware is connected to the simulation hardware to interact with the simulated system. The I/O connections may use communication protocols such as Ethernet, CAN and ARINC to send signals that emulate physical behaviors.
HIL testing with the real controller hardware generates a variety of data acquired from the controller and simulated system. The data is used to provide feedback to the controller, enabling the controller to adjust its behavior based on the simulated system’s responses. Measurement and verification include a wide variety of tests, ranging from normal operating conditions to fault conditions. Numerous iterations and refinements can be run to verify model accuracy and system performance.
Benefits of HIL testing
HIL testing offers numerous benefits to product designers and manufacturers. Because testing can be conducted before final assembly, it can identify potential flaws when they can be fixed more affordably and efficiently. Tests can be rerun to determine the efficacy of major fixes and minor adjustments. Such testing can also consider numerous scenarios without the time and expense required for physical tests. Automation can be used to conduct multiple scenarios, sometimes simultaneously, to accelerate testing and development.
By connecting real controller hardware to a simulated system, HIL testing offers more realistic testing than software-only simulation. It essentially combines aspects of both physical testing and software-based modeling. This combination also enables more collaboration amongst professionals with various backgrounds. Using primarily digital methods, test results can be shared readily with other stakeholders. Pure physical and software testing are often conducted by experts with specific expertise in those areas who report results back to product managers and others less interactively. HIL testing often involves collaboration up front and more ongoing access to hardware and system data throughout the process.
HIL testing also offers safety benefits, enabling teams to simulate conditions without exposing humans to dangerous situations. For example, an automobile braking system or other safety-critical systems can be tested using HIL techniques before testing in actual conditions.
Challenges of HIL testing
While typically not as expensive as physical testing of finished prototypes, HIL testing can be time-consuming and costly. Significant planning and development are required to build a simulation model, prepare the hardware controller for connection to the HIL system and monitor the testing.
The accuracy of HIL simulation can also be a challenge. Even with sophisticated hardware and software, it may not perfectly emulate actual systems, requiring ongoing testing and adjustment.
Collaboration among multidisciplinary teams can also prove challenging. While a wide range of perspectives can be beneficial, it can also require additional coordination and “cat herding” to achieve consensus on the approach and interpretation of results.
Present and future applications
A wealth of opportunities are available for teams able to properly address the various challenges. For example, the automotive industry has been particularly active in employing HIL testing. In addition to the previously mentioned transmission and braking examples, it can be used to test vehicle dynamics, steering systems, cruise control systems, advanced driver assistance systems (ADASs) and other systems employing ECUs.
In the aerospace industry, HIL testing can be used for flight control systems, avionics, navigation modules and a host of other areas. With the ability to perform real-time simulation, HIL has proven helpful for certifying aerospace systems and components.
It has also been used in the energy industry to simulate power plant behavior and grid reliability, in the electronics industry to test components and systems, and in industrial automation systems to evaluate system effectiveness before deployment. Infrastructure systems, such as water and wastewater treatment facilities, can use HIL to simulate scenarios such as peak demand and emergency scenarios.
Looking ahead, HIL testing will likely employ artificial intelligence and machine learning to further automate and refine testing procedures. A variation called virtual HIL (vHIL) testing is being used to create and execute tests before the actual ECU hardware is available. With the vHIL approach, testing can begin earlier and be automated to guide subsequent testing. As manufactured products become more complex, HIL and other methods will also become more advanced to meet the needs of product manufacturers.
