ARM adds functional safety IP to CPU and GPU architectures
ARM has introduced a series of processors designed to support decision making in autonomously operating systems such as cars and robots. And because autonomy is not possible without compliance with appropriate standards to ensure functional safety, the new IP include additional circuitry to ensure reliability and failsafe performance.
The new suite of IP includes the Cortex-A78AE general-purpose CPU, the Mali-G78AE Graphics processor, and the Mali-C71AE image signal processor (ISP). In combination with supporting software, tools and system IP, these products enable silicon providers and OEMs to design computing platforms for autonomous workloads. They will be deployed in a range of applications, from enabling more intelligence and configurability in smart manufacturing to enhancing ADAS and digital cockpit applications in automotive.
The Cortex-A78AE CPU is Arm’s latest, highest performance safety capable CPU, offering the ability to run different, complex workloads for autonomous applications such as mobile robotics and driverless transportation. According to the IP vendor, the CPU delivers a 30% performance uplift compared to its predecessor. Plus, it supports features to achieve the relevant automotive and industrial functional safety standards, ISO 26262 and IEC 61508 for applications up to (Automotive) Safety Levels ASIL D and SIL 3. To ensure compliance with these standards, the processor has an enhanced Split Lock functionality which enables system designers to choose between maximum safety or maximum performance; ARM calls this feature “Hybrid Mode”.
It is designed to specifically enable applications that target lower levels of ASIL requirements without compromising performance and allow the deployment of the same SoC compute architecture into different domain controllers.
The Mali-G78AE brings functional safety to embedded GPUs. Its outstanding feature is flexible partitioning: Up to four independent partitions offer the possibility to separate workloads for different safety use cases – in the car, for example, such a feature enables designers to use the device for safety-critical tasks such as the instrument panel and less safety-critical applications such as the infotainment system.
The Mali-C71AE ISP is addressing a broad range of vision applications across automotive and industrial. It offers the flexibility needed to support both human and machine vision applications such as production line monitoring and ADAS camera systems and is equipped with enhanced safety features to achieve ASIL B / SIL2 safety capability. Offering a high throughput of 1.2 giga pixel per second, it targets use cases such as real-time cameras or up to 16 buffered cameras.
These IP products will also be incorporated into existing platforms by Arms’ licensing partners, a company spokesperson said. A prominent example is Nvidia’s announced Orin processor, which in turn will be used to power – among others – carmaker Daimler’s Autonomous Driving Platform for future Mercedes vehicles.
More information: https://devsummit.arm.com/
Related articles:
Daimler, Nvidia co-develop software-defined vehicle architecture
Virtualization on Microcontrollers
ARM sale to Nvidia agreed at $40 billion
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
