1. Background
The automotive industry is undergoing rapid transformation, driven by technology innovation, automotive standards shaping demand, and changing consumer preferences.
As a major player in the automotive field, Arm is realizing the transition from Advanced Driver Assistance Systems (ADAS) to autonomous driving, from In-Vehicle Infotainment (IVI) to digital cockpits, from traditional telematics to fully connected vehicles, and from Internal Combustion Engine (ICE) to vehicle electrification.
Arm provides different categories of processors, each with a wide range of functionalities, specifically designed to meet the needs of these automotive applications. To help you leverage this processing power, below showcases how Arm’s different categories of processors are applied in various automotive applications.
2. What are the different categories of Arm processors for automotive?
3. Applications of Arm processors in automobiles
3.1. From ADAS to Autonomous Driving
Most vehicles today feature some form of assisted driving functionality. These systems help improve our driving experience and road safety. However, these functionalities are increasingly transitioning from playing a supportive role in our decision-making process to making decisions autonomously. Take the parking use case as an example; parking assistance systems have evolved from providing proximity information to drivers (who are still in control) to fully automated parking achievable with just a button press or even voice commands!
The processing capabilities and vehicle architectures of these two systems look quite different. In their most basic forms, the applications processed by assisted driving functions are divided into the following four stages:
Sensing > Perception > Decision > Execution.
To help explain the different processing requirements, the Society of Automotive Engineers (SAE) defines five levels of automation from L1 to L5.
The L1 assisted driving functionalities are basic features that support the driver in completing specific processes. Typically, the functionalities at this level do not impact other processes in the vehicle, and nearly all sensing, perception, and decision processing occurs at the vehicle’s edge (e.g., camera systems for lane departure warnings). Cost (SOC area) and thermal efficiency are significant considerations in these systems. Cortex-A CPUs, such as Cortex-A55 and Cortex-A65, are suitable for these systems due to their small size, efficiency, and diagnostic and system capabilities.
In L2 systems, certain aspects of the vehicle are directly managed by the assisted driving functionalities, such as adaptive cruise control. In these systems, the processing power and functional safety requirements (often up to ASIL D) are higher, but most processing still occurs at the edge, so power and thermal efficiency remain critical. Here, the Cortex-A65AE is an appropriate choice as it uniquely combines high power efficiency with high area efficiency. It can also scale performance up to eight cores in a single cluster and can operate in split mode or locked mode for systems requiring ASIL D and diagnostic functionalities.
Once we reach L3, the vehicle is now capable of autonomous driving under certain conditions but still expects the driver to control the vehicle in all other cases and intervene in the event of a vehicle failure. In these systems, multiple assisted driving functionalities work together to control the vehicle. Some processing still occurs at the edge, but this is increasingly shifting towards centralized systems for sensor fusion, perception, and decision-making. In these systems, performance and safety requirements are higher. Cortex-A CPUs, such as Cortex-A65AE and Cortex-A76AE, are suitable for high performance and can reach up to ASIL D in locked mode.
When we reach L4 and L5, the assisted driving functionalities in the vehicle will achieve full autonomous driving. The vehicle architectures for these systems are still being defined, but we can anticipate that they will require higher processing power. Early prototypes use off-the-shelf “backbone server” systems, which are not widely deployable. Arm understands the challenges faced by automotive OEMs and designs deployable computing for autonomous systems. The Cortex-A76AE is the first such example. It is designed to fit multi-cluster systems, providing over 250K DMIPS within an industry-leading power budget of below 15W. The Cortex-A76AE also supports locked mode for designs requiring ASIL D systems and diagnostic functionalities.
In all of the above systems, ensuring the functional safety of the system is crucial. Creating a Safety Island that can execute system operations and diagnostic checks is often used to provide an isolated area for this purpose. This Safety Island itself must have the highest level of safety, so features like Dual-Core Lockstep (DCLS) are crucial. It also greatly benefits from the ability to react in a real-time and deterministic manner, allowing for quick control and management of the system in case of failures. The Cortex-R52 processor combines the highest level of functional safety with efficient real-time performance, making it an ideal choice for a Safety Island.
3.2. In-Vehicle Infotainment (IVI) / Digital Cockpit
In recent years, largely driven by the increasing demand for premium user experience (UX), hardware integration, and the growing number of electric vehicles (EVs), expectations for high-quality cockpits have been on the rise. Drivers want to display rich and relevant information such as speed, navigation, and warning signs without taking their eyes off the road. Head-Up Displays (HUD) appeared in cockpit systems as early as 2012 and are now being integrated with augmented reality (AR) to further enhance the driver’s user experience and improve road safety.
Advancements in artificial intelligence and voice recognition technologies will enhance the usability of in-vehicle functionalities through advanced human-machine interfaces (HMI) such as gesture and voice control, improving safety and security through voice recognition.
In addition to controlling seats, ambient lighting, and air conditioning, the number of touchscreen displays in vehicles is also increasing, extending the IVI user experience to passengers and all aspects, raising performance requirements.
The next generation of IVI systems will see a high level of integration with other ECUs in the vehicle. These IVI systems are expected to merge with cluster systems that display safety-related vehicle and driver information. This combined system is referred to as the digital cockpit. The dashboard, which includes displays for vehicle operation instruments and controls, means that from a functional safety perspective, the digital cockpit system has mixed criticality.
Moreover, these systems will require higher performance levels. Cortex-A CPUs are well-suited for this, such as Cortex-A76, which boasts a 20% performance improvement over its predecessor and employs a stringent design process to avoid system failures.
This merging of traditionally independent systems presents interesting changes and challenges for hardware and software developers. The software stack now needs to be able to run safety-critical and commercial applications on the same system-on-chip (SoC) and securely display these different applications on their respective displays. The latest Cortex-A processors provide the necessary capabilities to run both Rich OS alongside RTOS or Autosar, and Cortex-A offers hardware support for the hypervisor platform that supports it.
3.3. Powertrain
The electrification of vehicle powertrain systems is one of the largest growth areas in today’s automotive sector. The demand for cleaner, more efficient vehicles is influenced not only by the growing consumer demand but also by increasingly stringent legislation. Traditional Internal Combustion Engines (ICE) will continue to be produced in the coming years, but hybrid configurations will become more prevalent. The widespread adoption of all-electric vehicles (EVs) will still depend on energy storage technologies and the available infrastructure for charging and energy distribution. Over time, we will see a shift towards higher power battery management.
Today, there is substantial demand in the market for both ICE and increasingly electrified systems, making it essential to provide solutions that meet these demands. Arm can cater to the needs of both ICE and electric-driven systems.
-
ICE: Requires high performance, determinism, and real-time requirements to control the combustion efficiency and after-treatment of spark-ignition (SI) and compression-ignition (CI) engines. Cortex-R processors provide the high-performance multi-core configurations and functionalities needed for these challenging applications, including support for floating-point calculations commonly used in these systems.
-
Electrification: Efficient field-oriented motor control to optimize range and energy management for large high-voltage battery packs is crucial for successful EV or hybrid systems. Both Cortex-M processors for high efficiency and high-performance Cortex-R processors provide a range of solutions to meet the demands of electrified systems.
-
Functional Safety: The powertrain is responsible for controlling the vehicle’s immense kinetic energy. In electrified systems, this control extends to the energy stored within the battery. This system must be able to manage energy without causing hazards, thus requiring the highest level of functional safety. This can be addressed by using processors with ASIL D system capabilities and built-in safety mechanisms, such as Cortex-R5.
Powertrains are increasingly expected to manage most electrified and ICE systems within a single high-performance controller. Encapsulating them within a single domain can provide an efficient interconnected solution. This enables improved control of the entire system from charging, energy storage, drive source selection and balancing, to energy recovery and demand forecasting. Processors like Cortex-R52 can deliver real-time high-performance multi-core products. The Cortex-R52 provides the highest level of functional safety and the ability to effectively manage multiple isolated workloads through real-time virtualization, meeting these demanding requirements.
3.4. Body
The application of body electronics is ubiquitous, covering the entire vehicle. Body applications range from simple, small sensing and actuation nodes, such as door locks and temperature and position sensors in HVAC control, to more complex central body control unit applications that sometimes include vehicle network control. As many different vehicles adopt a range of different tier-one suppliers, the ability to build multiple applications on a single platform based on a common architecture provides flexibility, scalability, and improved reusability for tools and software. Arm’s Cortex-M class processors offer a simple programming model and a well-supported tool and software ecosystem, along with highly configurable options for custom hardware.
Many of these applications are in a shutdown state, meaning they need to remain operational while the vehicle is parked, draining battery power for days or even weeks. Low power consumption during operation and during these “shutdown” periods is crucial for maintaining battery life. Cortex-M processors enable body applications to quickly wake up, perform an action, and quickly return to sleep mode, with ultra-low power consumption typically associated with wearable applications. Cortex-M processors easily meet these demands and benefit from a broad supportive ecosystem, along with tools and software that allow for simple, consistent, and rapid implementation, thereby enhancing efficiency and reducing development costs.
Security is becoming increasingly important in many automotive applications, including those in body control. Arm TrustZone in the Cortex-M series provides an opportunity for increased security support for these application processor profiles.
Cortex-R processors also provide flexible solutions for high-performance centralized body control systems and network controllers. The desire to inspect the increasing number of ECUs in vehicles makes Cortex-R processors an ideal choice for those needing to consolidate functionalities into fewer ECUs by simplifying the migration of multiple applications to a single processor while maintaining their isolation.
4. Development Solutions
In addition to processor technology, Arm has also created specialized software development tools, simulation models, and critical software components to accelerate innovation in the automotive sector. The lines of code embedded in vehicles in 2020 could be more than ten times that produced in the past decade, which has become a successful definition for the entire supply chain to complete from design to safety certification in a shorter time frame. Arm’s development solutions are highly optimized for the entire Arm processor family and have undergone external evaluation by TÜV SÜD for applications up to the highest ASIL D level (where applicable). Together with Arm’s vast ecosystem of third-party software, tools, and service providers, these solutions can shorten the product cycle of Arm-based systems.