Electric Vehicle Development From System To Software

Finding the proper tradeoffs is a major challenge for EV manufacturers, and the difficulty of building bench rigs makes virtual prototypes even more useful.


There is perhaps no higher profile electronic product category today than electric vehicles (EVs). Aggressive startups pioneered the market, but many major automobile manufacturers are now participating as well. Rising fuel costs, improved battery technology and increased environmental sensitivity have all helped to drive public enthusiasm for EVs. In response, developers are driving continuous innovation. EV performance now meets or exceeds that of traditional combustion-driven cars. Engineers are addressing “range anxiety” with improved efficiency, higher battery capacity and decreased charge time. Today’s EVs can drive faster and farther between charges, making them more practical for a wider range of drivers.

This innovation does not come for free and, for most EV buyers, cost is a significant consideration. Making the proper tradeoffs across performance, efficiency, driving range and cost is a major challenge for manufacturers. These tradeoffs must be evaluated and iterated many times during the development process as the design evolves, the market changes and competitors emerge. A successful EV product requires near-flawless execution of this process. In addition, the development team must deal with the general challenges of vehicular design, including close interaction among electrical and mechanical components and an operating environment with variable temperature and humidity plus noise and vibration.

Traditional automotive design has relied heavily on bench testing. There are three major limitations to this approach. Building bench rigs is an expensive undertaking, limiting the number of instances and the number of engineers who can access them. This is a particular concern for EVs, since they have significant software content and programmers have needed hardware to test their code. Second, testing for real-world failures by injecting faults damages or destroys the hardware, and some types of faults cannot be injected. Above all, finding design problems in bench testing is much too late in the EV development process. Bugs detected after silicon fabrication or software rollout result in project delays and cost over-runs.

Fortunately, there is now an alternative solution: virtual prototypes for EVs that can verify complete electro-mechanical-hydraulic-thermal systems, including their embedded software, purely in simulation. This permits early architectural exploration of subsystem topologies and design tradeoffs without having to reconfigure actual hardware. Implementation problems can be detected and investigated early, allowing improvements before any hardware prototypes are built. Physical testing time can be reduced since fault injection and analysis can be performed with the simulation model. Virtual prototypes can be replicated at will, making them available to all programmers for early software development. The result is a faster, more efficient and less expensive process for EV development.

Creating a unified virtual prototype for EV design and verification requires a mix of technologies to cover the full range of system to software, including electro-mechanical components and complex interaction between embedded software and the hardware components.

The Synopsys unified virtual prototyping solution for EV development supports hardware, software, and the complete system. The process starts with product requirements and the use of SaberRD to perform high-level, abstract studies. The system model is refined, with Silver running SIL simulations to develop ECU software and test the interaction among EV components. As the software evolves to create the production code that will run in the vehicle, Virtualizer runs the actual binaries for highly accurate simulations of system behavior. The use of Virtualizer continues as the entire system is integrated together. Silver comes back into play to run system-level tests for a high level of coverage, using TestWeaver to create the tests, manage the process and maximize coverage. Finally, SaberRD provides high fidelity modeling capabilities for power electronics and integration with software control.

Development of electro-mechanical systems for vehicular applications is always challenging, but EVs raise the bar significantly. The cost of physical prototypes is too high to make them available to all engineers who can benefit, plus it’s vital to “shift left” the design and verification process to much earlier in the project timeline. An EV development flow based on industry-leading products, like Synopsys, provides the solution. This flow enables exploration of design options, evaluation of tradeoffs and development of embedded software before any hardware is built. The “shift left” effect is realized by a significant decrease in time consumed in development iterations. This happens because the solution is all virtual and small changes can be implemented and tested almost instantaneously from anywhere. The virtual prototype can be replicated easily for much more efficient and robust EV development. The flow enables development of safer vehicles by testing faults and corner cases that would be dangerous or impossible to reproduce in hardware. These benefits greatly increase the chance of launching successful products in the competitive EV automotive sector. To learn more, please read a recent white paper.

Leave a Reply

(Note: This name will be displayed publicly)