Are digital twins changing the automotive industry? 3 things you need to know

Somewhere in Los Angeles, the Department of Transportation is collaborating on a solution to create a digital twin of the city’s transportation infrastructure. Elsewhere in the world, automotive use cases for digital twins are expanding. Given their far-reaching capabilities, the automotive industry is indeed one of the few industries integrating the use of digital twins.

With new trends such as electrification and autonomous driving developing, today's vehicles can be described as an ever-evolving system of subsystems and often feature different electronic modules, sometimes as many as 100 or more. The complexity of the subsystems requires complex real-time virtual representations. This is exactly what a digital twin is: a virtual copy of a physical object or system that can be used to simulate real-world conditions and test different scenarios in a virtual environment. It monitors and reflects physical objects by using sensors, machine learning, and artificial intelligence. Each type of digital twin has its own unique capabilities and benefits and can be used together to create a comprehensive vehicle digital model.

Hierarchy-based classification of digital twins

Unit-level digital twins: At lower hierarchical levels, this can reflect components of equipment, materials, or environmental factors.

System-level digital twins: This includes multiple unit-level digital twins within a production system or complex physical product.

System-level (SoS) digital twin: Multiple system-level digital twins form an SoS-level digital twin, thereby enhancing collaboration in supply chain, design, service, and maintenance.

Digital twins contain large amounts of data

Digital twins not only generate large amounts of data about physical objects, but they also require data to be captured from the physical objects. In cars, sensors are affixed to multiple components of the vehicle to generate and forward large amounts of data about the vehicle's functions and status. Data streaming is an important component of modern digital twins and can provide multiple benefits throughout the vehicle’s lifecycle.

Figure 1. Composition and characteristics of the digital twin ecosystem

Monitoring of vehicle components is a special advantage brought by data availability. One example is tracking the battery's state of health (SoH), which is critical for correcting anomalies. Similar to SoH, digital twins can also help track the remaining useful life (RUL) of vehicle components. SoH can help automakers manage things like battery charging and thermal control of components, while RUL can help insurance companies assess the value of damaged vehicles or remind owners to order new components. For example, if a battery system's RUL expires, the batteries can be replaced and reused in other applications.

All of these possibilities require large amounts of data. While digital twins’ reliance on data isn’t unique to cars, modern cars have become more complex, giving the technology room to shine. Not only do these complex vehicles generate more and more data, they also need to forward and process the data as quickly as possible for optimal performance.

Connected vehicles are changing the driver experience. Unlock the value of connected vehicles with NXP vehicle networking solutions.

Some changes to vehicle architecture

Vehicle architecture must become more flexible to accommodate the bidirectional transmission of data. Specifically, high-performance in-vehicle networks (IVNs) are required to meet the processing requirements of increasing data usage. A growing trend in this area is the shift to Zonal E/E architecture.

Regional architecture delivers vehicle functionality over the network, allowing problems to be localized to specific parts of the vehicle architecture. This enables faster response and error acknowledgment. Digital twins can reside anywhere, including close to vehicle end nodes (near actuators and sensors), more centrally in high-performance computing clusters, or in a connected cloud. Latency and decision response time determine where the digital twin should reside. Tools and processing capabilities allow digital twins to be implemented during the migration of signals to the cloud, allowing system developers to easily migrate functionality.

Figure 2: Architecture evolution trend: from domain to region

Additionally, by increasing the processing power of central vehicle computers, regional E&E architecture enables vehicle innovation driven by software and over-the-air updates. Partitioning the architecture simplifies the integration of data from multiple sources, which can facilitate the use of digital twins.

Vehicle prototyping, performance optimization and other benefits

Although complex, virtual prototypes can clearly illustrate how vehicle components work together in different scenarios. For OEMs, digital twins not only facilitate virtual prototyping but also offer greater possibilities to upgrade vehicle designs, deploy new features and thereby reduce costs. By collecting and analyzing operational data, they simulate vehicle performance and share insights about potential improvements. This can provide multiple benefits for optimizing the performance of vehicle components.

One of the benefits is improved battery management to extend battery life. Using AI-driven solutions can adjust to any changes in battery health, allowing for continuous improvement, especially in control decisions. This support can help overcome major issues with EV batteries and BMS, including improving EV range, ensuring battery safety, and extending battery life.

In addition to BMS improvements, predictive maintenance has the added advantage of enabling real-time monitoring of vehicle components and systems. This data can be used to predict when maintenance is needed, reducing downtime and maintenance costs. Other use cases for the technology in the automotive industry include simulating the behavior of self-driving cars in different driving scenarios. This possibility helps refine and test the software and hardware components of autonomous vehicles before they are deployed on the road.

Overall, the use of digital twins in cars will become more widespread, especially as vehicle digitization continues to advance. Its use in the industry has the potential to improve product design, manufacturing processes and vehicle maintenance, leading to better products and a more efficient and reliable automotive industry. As this development continues, a key matter will be the need to prioritize functional safety and cybersecurity requirements for various automotive processes.