Arm Community
Arm Community
  • Site
  • User
  • Site
  • Search
  • User
Arm Community blogs
Arm Community blogs
Embedded and Microcontrollers blog A Starter's Guide to Arm Processing Power in Automotive
  • Blogs
  • Mentions
  • Sub-Groups
  • Tags
  • Jump...
  • Cancel
More blogs in Arm Community blogs
  • AI blog

  • Announcements

  • Architectures and Processors blog

  • Automotive blog

  • Embedded and Microcontrollers blog

  • Internet of Things (IoT) blog

  • Laptops and Desktops blog

  • Mobile, Graphics, and Gaming blog

  • Operating Systems blog

  • Servers and Cloud Computing blog

  • SoC Design and Simulation blog

  • Tools, Software and IDEs blog

Tags
  • Powertrain
  • automotive
  • In Vehicle Infotainment (IVI)
  • Security
  • Advanced Driver Assistance Systems (ADAS)
  • Electrification
  • functional safety
  • Electronic Control Unit (ECU)
  • Autonomous Car
Actions
  • RSS
  • More
  • Cancel
Related blog posts
Related forum threads

A Starter's Guide to Arm Processing Power in Automotive

James Scobie
James Scobie
July 25, 2018

Co-authored by James Scobie and Govind Wathan

The automotive industry is rapidly transforming, with technology driving innovation, automotive standards shaping requirements and consumer preferences changing demands.

As a key player in the automotive space, Arm is enabling these radical transitions from Advanced Driver-Assistance Systems (ADAS) to Autonomous Driving, In-Vehicle Infotainment (IVI) to Digital Cockpit, traditional telematics to fully connected cars and from the Internal Combustion Engine (ICE) to vehicle electrification.

Arm offers different classes of processor, each with a broad range of capabilities, specifically designed to address the needs for each of these automotive applications. To help you take advantage of this processing power, this guide will help show how Arm’s different classes of processors can be applied to various automotive applications.

What are the Different Classes of Arm Processors for Automotive?

Processor Class Designed For Automotive Applications
Arm Cortex-A
  • High performance 
  • Optimized for rich Operating Systems and hypervisors 
  • Autonomous 
  • ADAS vision and LiDAR
  • IVI 
  • Digital cockpit 
  • Connectivity  
  • Gateway
Arm Cortex-R
  • Optimized for high performance 
  • Hard real-time deterministic applications and RTOS 
  • Autonomous 
  • ADAS radar 
  • Connectivity  
  • Powertrain 
  • Chassis  
  • Central body electronics
Arm Cortex-M
  • Smallest area and lowest power 
  • Optimized for discrete processing and microcontrollers 
  • Body electronics 
  • Gateway 
  • SoC management 
  • Sensors  
  • End point actuators

List of Arm processors in a car

Four Automotive Applications Designed with Arm Processing Power

ADAS to Autonomous Driving

ADAS in car cockpit

In some form or another, assisted driving features can be found in most of today’s vehicles. They serve to improve both our driving experience and road safety. However, these features are increasingly transitioning from playing a supporting role in our decision-making process to taking decisions on their own. Taking as an example the parking use case; parking assistance systems have gone from parking sensors that deliver proximity information to the driver (who is still in control), to fully automated parking at the touch of a button, or even voice command!

The processing power and vehicle architecture looks significantly different for these two systems. In its most basic form, the application of processing for assisted driving features fall in the following four stages:

Sense > Perceive > Decide > Actuate.

Graph of Arm technology in autonomous vehicle systems

To help explain the different processing requirements, the five levels of automated driving (Levels 1 to 5) as defined by the Society of Automotive Engineers (SAE) serve as a useful reference.

Assisted driving features found in Level 1 are basic and support the driver for a specific process. Typically, features at this level do not influence other processes in the vehicle and nearly all of the sense, perceive and decide processing happen at the edge of the vehicle (e.g. a camera system for lane departure warning). Cost (i.e. silicon area) and thermal efficiency are important considerations in these systems. Cortex-A CPUs, such as the Cortex-A55 and Cortex-A65 are suitable for these systems due to their small size and high-efficiency, as well as diagnostic and systematic capabilities. 

In Level 2 systems, some aspects of the vehicle are directly managed by the assisted driving feature, such as is the case with adaptive cruise control. In these systems, the processing power and functional safety requirements (typically up to ASIL D) are higher, but most of the processing still happens at the edge and therefore power and thermal efficiency is still critical. Here, the Cortex-A65AE is a suitable choice due to its unique combination of high power efficiency and high area efficiency. It also scales in performance with up to eight cores in a single cluster and can operate in either split-mode or, for systems that require ASIL D systematic and diagnostic capabilities, in lock-mode.

Arriving at level 3, the vehicle is now able to drive autonomously in some conditions, however the driver is still expected to control the vehicle in all other conditions and to intervene in the event the vehicle has a failure. In these systems, multiple assisted driving features work together in order to control the vehicle. Some of the processing still happens at the edge, but this is increasingly moving towards a centralized system for sensor fusion, perception and decision making. In these systems, the performance and safety requirements are even higher. Cortex-A CPUs, such as the Cortex-A65AE and Cortex-A76AE are suitable for high performance and up to ASIL D in lock-mode.

As we get to Levels 4 and 5, the assisted driving features in the vehicle achieve full autonomous driving. The vehicle architecture of these systems is still being defined, but already we can expect it to need higher processing power. Examples of early prototypes use off-the-shelf “server-in-the-trunk” systems that are not widely deployable. Arm understands the challenges faced by automotive OEMs and designs deployable compute for Autonomous systems. The Cortex-A76AE is the first such example. It has been designed to fit into multi-cluster systems that deliver more than 250K DMIPS within an industry-leading power budget of under 15W. The Cortex-A76AE also supports lock-mode for designs that require ASIL D systematic and diagnostic capability.

In all of the systems above there is a need to assure the functional safe operation of the system. The creation of a safety island where this can be performed is frequently used to provide an independent area where the operation of the system and scheduling of diagnostic checks can be performed. This safety island itself must have the highest level of safety and so having features such as Dual Core Lock-Step (DCLS) are important. It also greatly benefits from being capable to react in a real-time and deterministic manner allowing the rapid control and management of the system in the presence of a fault. The Cortex-R52 processor delivers the mix of both the highest level of functional safety combined with efficient real-time performance making it an ideal choice for the safety island.

In-Vehicle Infotainment (IVI) / Digital Cockpit

In-Vehicle Infotainment (IVI) Digital Cockpit in car

In recent years, and largely influenced by growing demands for premium user experiences (UX), hardware consolidation and the increase in volumes of Electrical Vehicles (EVs), the expectations placed on premium cockpits have been on the rise. Drivers want rich and relevant information such as speed, navigation and warning signs displayed in such a way as to avoid taking their eyes off the road. Heads-Up-Displays (HUDs) which have featured in cockpit systems from as early as 2012, are now incorporating Augmented Reality (AR) to further enhance the driver’s UX and improve road safety.

Advances in AI and voice recognition technology will improve the ease-of-use of in-vehicle functions, through advanced Human-Machine Interfaces (HMI) such as gesture and voice control, as well as increased security and safety through voice recognition.

As well as control for your seats, ambience lighting and air conditioning, the total number of touchscreen displays in a vehicle has also gone up, extending the IVI user experience to passengers and all-in-all, increasing the demands for performance.

The next generation of IVI systems will see a high degree of consolidation from other ECUs within the vehicle. These IVI systems are expected to merge with cluster systems, which display vehicle and driver information that pertain to safety. This combined system is referred to as the digital cockpit. The inclusion of the cluster which displays instrumentation and control for the vehicle's operation, mean that digital cockpit systems have mixed-criticality from a functional safety perspective.

Moreover, these systems also demand increasing levels of performance. Cortex-A CPUs are a good fit for this, such as the Cortex-A76, which delivers a 20% uplift in performance over its predecessor and is designed on a rigorous design flow to avoid systematic faults.

This merging of what have been traditionally separate systems brings interesting changes and challenges for both 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 display these different applications safely on their respective displays. The latest Cortex-A processors provide the performance required to run a Rich OS next to a RTOS or Autosar stack, and the Cortex-A offers hardware support for the hypervisor platform that enables it.

IVI digital cockpit in car displays

Powertrain

Powertrain in car

Electrification of the vehicles powertrain system is currently one of the largest growth areas in the automobile. Demand for cleaner and more efficient vehicles is not only influenced by rising consumer demands but also increasingly stringent legislation. Conventional Internal Combustion Engines (ICE) will continue in production for some years to come, but increasingly in hybrid configurations. Widespread adoption of full Electric Vehicles (EV) will remain dependent on energy storage technology and the available infrastructure for charging and energy distribution. Over time, we’ll see the balance move to electric drives with higher power battery management. 

Today, there is high volume market demand for both ICE and increasingly electrified systems, so solutions to meet these needs are essential. Arm can address the needs for both ICE and electric drives.

  • ICE: High performance, determinism and real-time requirements are required to control the combustion efficiency and after treatment for both Spark Ignition (SI) and Compression Ignition (CI) engines. Cortex-R processors provide the high performance multi-core configurations and functionality demanded by these challenging applications and include support for floating point calculations frequently used in these systems.
  • Electrification: Efficient field-oriented motor control to optimize range together with energy storage and charge management for large high voltage battery packs are critical for successful EV or hybrid systems. The high efficiency of Cortex-M processors and high performance Cortex-R processors both offer a range of solutions to meet the needs of electrified systems.
  • Functional safety: The drivetrain system is responsible for control of the vehicle's enormous kinetic energy. In electrified systems this control extends to the stored energy within the battery. The system must be capable of energy management without causing hazards and so the highest level of functional safety is required. This can be addressed through the use of a processor with systematic capability for ASIL D and built in safety mechanisms, such as the Cortex-R5.

Drivetrain systems are increasingly looking to manage large parts of the electrified and ICE system within a single high performance controller. Encapsulating this within a single domain provides an efficiently interconnected solution. This enables the ability to manage improved control of the complete system from charging, energy storage, drive source selection and balancing to energy recovery and demand prediction. Processors such as the Cortex-R52 enable the delivery of real-time high performance multi-core products. Offering the highest level of functional safety and the ability to efficiently manage multiple separated workloads though real-time virtualization, the Cortex-R52 can deliver on these tough demands.

Body

Body electronics in car

Applications of body electronics are ubiquitous, covering the entire vehicle. Body applications range from simple, small sense and actuation nodes, like door locks, temperature and position sensors in Heating Ventilation Air Conditioning (HVAC) control, to more complex Central Body Control Unit applications, sometimes incorporating vehicle network control. With a range of different Tier 1 suppliers adopted in many different vehicles, the ability to build multiple applications onto a single platform based on a common architecture gives the flexibility, scalability and improved reuse for both tools and software. Arm’s Cortex-M class of processors delivers a simple programming model and well supported ecosystem of tools and software, together with highly configurable options to tailor the hardware. This reduces development cost and time to market whilst meeting the demanding needs of these systems.

Many of these applications are key-off, meaning they need to remain operational when the car is parked, draining battery power for days or even weeks. Low energy consumption is important both at run-time and during these "key off" periods to maintain battery life. Cortex-M processors enable body applications to wake up, complete an action and go back to sleep rapidly, with the ultra-low energy sipping power normally associated with wearable applications. Cortex-M processors easily meet these needs, and also benefit from a broad, supportive ecosystem, along with tools and software that allow simple, consistent and rapid implementation which improves efficiency and reduces development cost. 

Security is increasingly important across many automotive applications and this includes those in body control. The introduction of Arm TrustZone within the Cortex-M family offers an opportunity for added security support in this profile of processors for these applications. 

Cortex-R processors also offer a flexible solution for high performance centralized body control systems and network controllers. The desire to keep a check on the increasing number of ECUs in the vehicle makes Cortex-R processors an ideal fit where consolidation of functions into fewer ECUs is needed by simplifying the migration of multiple applications into a single processor and maintaining their isolation.

Development Solutions

Alongside its processor technology, Arm also creates professional software development tools, simulation models and critical software components to accelerate innovation in the automotive segment. With 2020’s cars likely to embed over 10x as many lines of code as those produced in the previous decade, it has become success-defining for the entire supply chain to be able to go from design through to safety certification in less time. Arm’s development solutions are highly optimized for the entire range of Arm processors and have been externally assessed by TÜV SÜD for use in applications up to ASIL D where applicable. Together with Arm’s vast ecosystem of third-party software, tools and service providers, these solutions enable shorter product cycles on Arm-based systems.

More information 

For more information on Arm's Automotive solutions, please visit our web page below. 

Arm Automotive Solutions

Anonymous
Embedded and Microcontrollers blog
  • Adapting Kubernetes for high-performance IoT Edge deployments

    Alexandre Peixoto Ferreira
    Alexandre Peixoto Ferreira
    In this blog post, we address heterogeneity in IoT edge deployments using Kubernetes.
    • August 21, 2024
  • Evolving Edge Computing and Harnessing Heterogeneity

    Alexandre Peixoto Ferreira
    Alexandre Peixoto Ferreira
    This blog post identifies heterogeneity as an opportunity to create better edge computing systems.
    • August 21, 2024
  • Demonstrating a Hybrid Runtime for Containerized Applications in High-Performance IoT Edge

    Chris Adeniyi-Jones
    Chris Adeniyi-Jones
    In this blog post, we show how a hybrid runtime and k3s can be used to deploy an application onto an edge platform that includes an embedded processor.
    • August 21, 2024