Access Official Technical PDF

Share Summary:LinkedInEmail

Virtual vehicle development is becoming an important way to manage the growing complexity of electrified systems. Projects supported by the European Green Vehicles Initiative (EGVI) explored how digital models, simulation and connected test environments can help engineers develop and validate vehicle components and systems. For industrial automation professionals, the approaches are relevant beyond automotive: they highlight practical ways to coordinate multidisciplinary engineering, connect virtual and physical testing, and assess system interactions earlier in development.

The EGVI portfolio covered 73 projects and involved 1,317 participants, including 208 small and medium-sized enterprises. Across the portfolio, work ranged from batteries and electric powertrains to virtual product development and testing. Several projects in the 2018–2019 calls focused specifically on the digital engineering methods needed to integrate complex vehicle subsystems.

Why virtual vehicle development matters

Electrified vehicles bring together mechanical, electrical, electronic and software-based systems. Developing components independently and integrating them only near the end can create additional work: teams must resolve interface issues, exchange models and data, and calibrate the overall system. Simulation-based development offers another path. Engineers can combine component and subsystem models into virtual prototypes, examine system-level behavior and explore design choices before relying on physical test vehicles or rigs.

This does not eliminate physical testing. Rather, it can shift some development and verification activity earlier in the process and provide a structured connection between models and experiments. That principle will be familiar to automation teams working with machine models, control software, drives, sensors and plant-level systems.

Digital models across the engineering lifecycle

The PANDA project addressed how digital models of electrified vehicles and components could be organized for use in both virtual and real testing. Its approach used a “W-model” to represent development activities that include virtual verification alongside more conventional product development and testing. A dedicated model library and cloud-based computing were intended to support the reuse and integration of models.

PANDA also used Energetic Macroscopic Representation (EMR), a graphical formalism for organizing models of systems and components. The project reported that a simulation using its EMR-based library required 15% of the computation time of a simulation using classical structural libraries in the comparison described. This result illustrates why model structure and interoperability matter: a model is more useful when teams can connect it consistently to other models and use it across development tasks.

VISION-xEV took a complementary approach, developing methods for component and system integration modelling. The project brought together models across domains, including electrical and thermal battery behavior, inverters, electric machines and powertrain thermal management. Its aim was to support virtual prototyping and optimization, reducing the need to defer discovery of system-level issues until laboratory or road testing.

Connecting simulation with physical tests

Simulation is most valuable when engineers can relate its outputs to test evidence. Hardware-in-the-loop (HIL) testing provides one such link: real hardware can be evaluated within a simulated system or operating environment. PANDA planned HIL testing for an electrical subsystem of a hybrid vehicle, alongside virtual testing of battery-electric, fuel-cell and hybrid vehicle references.

XILforEV extended the idea from an individual test setup to connected experimental environments. The project aimed to link test platforms in different physical domains and locations, allowing coordinated real-time test scenarios. Its planned use cases included brake blending, ride blending, integrated chassis control and fail-safe studies. The general engineering lesson is that test assets can reveal interactions that are difficult to investigate when each subsystem is assessed in isolation.

For system integrators, this raises familiar implementation questions:

  • How will models and test equipment exchange signals and data?
  • Which interfaces and operating conditions must be represented in the simulation?
  • How will teams verify that virtual results correspond to physical behavior?
  • Can test platforms be reused across projects or accessed by geographically distributed teams?

These questions apply wherever controls, hardware and software must work together, including industrial machinery and automated production systems.

Using connectivity and data to manage system behavior

Digital development also depends on the information available to a system. CEVOLVER explored connected energy and thermal management for electric vehicles. Its proposed functions used connectivity and cloud computing to support route selection, driving behavior and charging decisions. The project’s work included defining communication architectures and identifying required signals for the demonstrator vehicles.

This is a useful reminder for automation projects: advanced control functions depend on clearly defined data flows, communication interfaces and system responsibilities. Cloud resources may support calculations that are not practical on a vehicle control unit, but the system still needs a sound architecture connecting onboard controls, external services and the relevant data.

What automation teams can take from these projects

The EGVI work was directed at vehicles, but its engineering methods have broader relevance. When a project spans disciplines and suppliers, teams can benefit from:

  • Shared model conventions: A common structure makes it easier to integrate subsystem models and exchange engineering information.
  • Early system-level simulation: Virtual prototypes help teams examine interactions before all physical components are available.
  • Virtual and physical test planning: Linking simulation with HIL or other test setups helps connect predicted behavior with measured behavior.
  • Explicit interface definition: Signals, data ownership and communication requirements should be established as part of system design.
  • Reusable test environments: Connected test assets can support investigation of cross-domain behavior and facilitate collaboration.

These practices do not remove the need for engineering judgment or physical validation. They provide a framework for organizing complex development work and making better use of models, test equipment and specialist expertise.

Conclusion

Virtual vehicle development demonstrates how digital models, simulation and connected testing can support the integration of increasingly complex systems. The EGVI projects show several complementary approaches: organizing reusable models, building virtual prototypes, applying HIL testing and connecting experimental setups. For automation engineers and technical decision-makers, the practical takeaway is to treat model integration, interfaces and test strategy as core parts of system architecture—not as activities left until commissioning or final validation.

Access Official Technical PDF

About The Author