At each turn negotiated by the autonomous car, the stakes are considerable. Indeed, on this test track, all the variables of each driving scenario are taken into account and measured repeatedly and reliably. Although you are not driving the car, you are in control of the main mission of this test drive: training your car’s self-driving algorithm to make the optimal decision every time, without serious errors.
This may seem impractical in the real world, where virtually every variable surrounding a self-driving car is unpredictable, and crashes during testing are inevitable. However, thanks to new fleets of digital twins, these types of idealistic scenarios are becoming a reality in today’s digital test environments. Although the principle of repeatedly measuring each variable of an autonomous car is still in the development phase, this model is becoming more precise with each new advance in the field of car emulation. Automakers are increasingly using digital twins to accurately test and measure specific automotive components and features before launching them in the real world, while bringing innovations to market faster, safer and more cost-effectively. .
Digital twins are digital representations of real-time data that help build predictive models to determine the probability of success of physical prototypes. According to a recent report by research firm Future Market Insights, the global market for digital twin technologies, which today is worth $9.5 billion, is expected to increase at an annual growth rate of 22.6% to reach 77, $65 billion by 2032. Various sectors, from healthcare to telecommunications, real estate and retail, are using digital twins, but the largest portion of this market, 15%, is held by the transport and communication industry. car. In the automotive industry, the increasing use of simulation, design, maintenance, repair and overhaul (MRO), manufacturing and the consequences of car accidents justifies the use of the technology of digital twins. According to the report, this technology can test complex safety scenarios for autonomous vehicles and measure the importance of maintaining components of a race car engine that are at risk of damage or burnout.
Thanks to digital twins, automakers are now leveraging vast amounts of data gathered in the lab from testing self-driving cars to perform complex simulations. These simulations allow product developers to dig deeper into how the artificial intelligence (AI) of an autonomous vehicle will react to unpredictable situations, such as weather conditions (hail or snow) or a traffic jam. It also allows product developers to program a much larger set of functional tests in much less time. For example, researchers can perform crash test simulations faster and safer using digital twins. At their core, self-driving cars are essentially robots operating in the real world. Digital twins are the digital environments that humans can build, manipulate and control to teach these autonomous vehicles to operate safely in the virtual world before integrating their intelligence into real-world scenarios where they will interact with spontaneous events and human behaviors.
Today, automakers use digital twins for every automotive function they test, from radar and C-V2X to in-vehicle networking and cybersecurity. By integrating multiple digital twins, they can build a comprehensive test platform, where they can train the car’s self-driving algorithm to accurately see and react to its complex and dynamic environment. For their part, automakers can reproduce real-life driving scenes in the lab by varying traffic density, speed, distance and total number of targets, which can speed up test times for the worst cases. more common while minimizing the risks.
Designers use digital twins to model sensors, test them in the lab against real-world scenarios, and explore new sensor designs and combinations. Additionally, digital twins are used to model a car’s in-vehicle network to test network bandwidth and data speed to improve response times. EWhen it comes to cybersecurity testing, automakers are using digital twins to test all endpoints in a given design environment to get a clearer picture of a vehicle’s security around the world reality, while eliminating security risks.
It is certain that the use of these digital twins in the development of automotive products presents certain challenges. Emulation, in general, can never completely represent the real world, but by adding, for example, noise or stress to a test environment, the digital twin can come close to reproducing real-world scenarios. On the other hand, since variables are not completely random, product developers can control and manipulate them, increasing the chances of getting the result right faster and minimizing costly rework and schedule delays. .
What does this mean for the future of automotive innovation? As we said before, we are about to live a unique experience. By bringing the road into the lab, the automotive industry is shifting its innovation paradigm, and digital twins are the quiet but effective actors that unlock these new possibilities behind the scenes.
Digital twins: an asset for the autonomous vehicles of tomorrow