Previously announced to meet different computing needsAdjust the name of its computing subsystem (CSS) productAfter more clearly identifying the corresponding computing scenarios, Arm gave a more specific description of the computing subsystem named Arm Zena for automotive application scenarios, and advertised the addition of active safety protection design driven by artificial intelligence.
Arm stated that the automotive industry is now closely integrated with smart connectivity and artificial intelligence. Smart features previously reserved for high-end vehicles, such as driver status monitoring, active safety driving systems, and adaptive in-vehicle infotainment platforms, have now gradually become standard features in new vehicles of all levels.
On the other hand, with the addition of more intelligent functions, the complexity of automotive product development continues to increase. At the same time, safety requirements continue to increase, and even the demand for scalable computing is growing. Arm believes that automakers and chip suppliers need modular, secure, reliable, and software-ready computing platforms that meet high performance and high energy efficiency. They must also be able to reduce various operational risks of vehicles and shorten vehicle product development cycles.
Arm Zena CSS, a newly named computing subsystem, aims to advance the development of AI-powered vehicles. Arm also touts its ability to ensure vehicle safety with the latest Armv9 Automotive Enhanced (AE) technology, while improving computing performance while further enhancing power efficiency. Furthermore, Arm Zena CSS, with reference firmware and corresponding software support, will help automakers accelerate vehicle product development and deliver a differentiated intelligent driving experience.
Compared to traditional vehicle development, which typically takes 3-5 years, the Arm Zena CSS computing subsystem will accelerate the overall pace of development, allowing automakers to innovate vehicle products faster and on a larger scale, and shorten the time it takes for vehicle products to enter the market.
The Arm Zena CSS computing subsystem design includes a 16-core design.Arm Cortex-A720AE CPU high-performance computing clusterThe Arm Neoverse CMN S3AE interconnect bus design ensures consistency in computing transmission between the CPU and chip through I/O ports. It also uses the Arm Cortex-R64AE 82-bit processor with safety features to provide high-performance computing for automotive zone controllers and safety islands.
Furthermore, it features the Arm Mali-C720AE with a Mali GPU, image signal processor (ISP), and the ability to integrate accelerated computing elements with partner-customized computing architectures to meet diverse AI workload requirements. Arm TrustZone technology ensures full system security, establishes a hardware root of trust, and maintains security during over-the-air updates, preventing the risk of external intrusion during vehicle firmware updates.
By pre-integrating the aforementioned hardware and firmware components, Arm claims that automotive chips built with the Zena CSS computing subsystem can reduce development time by up to 12 months compared to those using standalone IP designs. These chips can encompass everything from in-vehicle infotainment platforms to central computing systems and Level 2+ advanced driver assistance systems (ADAS), giving automakers greater flexibility in deploying applications across a wide range of vehicle models and performance levels without having to redesign the computing stack or undergo safety certification from scratch.
At the same time, Arm also boasts that under its Zena CSS computing subsystem design, its chips and software are highly compatible and portable, so it can be used in a variety of vehicle models, thereby reducing the time required for overall vehicle development, thereby reducing vehicle design costs and complexity, and allowing vehicle products to enter the market faster.
Arm also explained that the Zena CSS computing subsystem is integrated with the Scalable Open Architecture for Embedded Edge (SOAFEE). The SOAFEE community is also developing blueprints for the CSS computing subsystem specifically designed for vehicle use cases. These include Autoware's Open AD Kit blueprint for accelerating the development of autonomous driving functions, DENSO's latest mixed-criticality safety requirements blueprint, and Panasonic Automotive Systems' blueprint for developing and deploying digital cockpit and in-vehicle infotainment platform solutions.
Through the Arm SystemReady certification and compliance program, as well as other industry-standard APIs, the embedded edge scalable open architecture blueprint will run seamlessly on the Zena CSS computing subsystem. Arm SystemReady's new automotive extensions will be fully open to the vehicle ecosystem by the end of 2025, ensuring that the operating system and software stack for vehicle applications can run seamlessly across different hardware.
Under these design architectures, Arm expects to promote the development of modern vehicles driven by artificial intelligence, and be able to respond to heterogeneous computing needs, real-time inference computing at the edge, and continuous large-scale updates driven by artificial intelligence and software definition, so that vehicle functions can be continuously updated.
