Connected and autonomous vehicles (CAVs) have the potential to transform the way people travel, live and work.
However, developing a safe CAV involves designing vehicles that include a complete array of sensors. All of these sensors must be rigorously tested – both on their own and in concert with other systems – to ensure vehicles are safe and reliable.
But when costs, logistics and legislation stand in the way of field testing, how can you truly test your CAV prototypes in the full range of conditions your vehicles may encounter on the road?
Increasingly, simulation holds the answer.
Breaking free of real-world testing
By combining hardware and software simulation tools, CAV designers can create digital car models and model the manoeuvres they take, the environments they drive in, and the obstacles they meet. By combining software, hardware and model-in-the-loop testing in this way, designers can assess how prototypes perform in a diverse range of situations.
Spirent recently worked with automobile design and development company Italdesign to integrate accurate, realistic GNSS signal simulation into its CAV test platform.
From test track to lab
Italdesign began testing CAV prototypes on its own test track, but the speeds it could achieve and the conditions it could create there were too limited for rigorous testing. And because Italian legislation doesn’t allow CAVs to be tested on public roads, simulation was the only way to put the vehicle through its paces in a wide range of conditions.
The solution was to build a complete, lab-based co-simulation platform. Using a combination of dSPACE SCALEXIO hardware and IPG CarMaker software, Italdesign can now simulate its test track and the vehicle’s dynamics (motion and trajectory) in real time.
Integrating true GNSS simulation
A crucial part of Italdesign’s research is to understand how the vehicle’s GNSS receiver responds to the constantly changing satellite signal environment as it proceeds on its journey. To test the vehicle’s GNSS receiver, Italdesign wanted a highly controllable GNSS signal simulator that was capable of generating realistic signals from multiple satellite constellations (GPS, Glonass, BeiDou etc.) on multiple frequencies (L1, L2, L5 etc.) Without accurate GNSS signal simulation, tests run the risk of failing to identify performance issues in GNSS receivers that could result in a CAV that isn’t for purpose. Because of this, Italdesign chose Spirent’s GSS7000 multi-constellation, multi-frequency GNSS signal simulator for its testing lab. With the Spirent GSS7000, Italdesign was able to get a highly accurate, yet flexible approach to GNSS testing.
A crucial challenge was ensuring the GSS7000 would integrate with into Italdesign’s core testing platform. To do that, we developed a module that used the GSS7000’s PosApp software API to send, receive and process data relating to any car model in real time from both the dSPACE and IPG systems. With the module in place, Italdesign’s test systems can automatically feed vehicle position, velocity, acceleration, altitude, angular velocity and angular acceleration data to the Spirent simulator.
Flexibility, accuracy and integration
Since integrating the Spirent GSS7000 into its simulation platform, Italdesign has been able to accurately evaluate how its prototypes receive and process GNSS signals in a full range of signal conditions.
Thanks to the capabilities of the Spirent GSS7000, Italdesign can also be confident that its prototypes could operate safely using GNSS signals from any constellation and frequency. And because of the ability to test multi-GNSS, it can also be sure its prototypes have built-in redundancy to minimise the risk of navigation errors.
Between the accuracy of Spirent’s simulation and the close integration with dSPACE SCALEXIO and IPG CarMaker, Italdesign now has a powerful, flexible simulation platform that overcomes the costs and complexity associated with testing autonomous vehicles in the field.
For more information, read the full Italdesign case study.