Arm Enables Autonomous System Of All Kinds Posted on : Nov 07 - 2020

Arm DevSummit was one of several virtual tech conferences and launch events taking place over the past month week and it included everything from a fireside chat with the CEOs of Arm and Nvidia about the proposed acquisition to new technology announcements and the deep dive training tracks one would expect. One of the most interesting announcements was around a new platform of cores for SoCs specifically targeting autonomous machines. While there is a growing focus on autonomous vehicles, few companies look at the broader opportunity for autonomous control across a wide variety of platforms that include autonomous systems for materials handling and production for everything from consumer products to agriculture. Arm announced a complete platform that can span a wide range of applications that meet the safety critical requirements of autonomous systems.

As Arm expanded its technology offerings in mobile, it began releasing a complete platform of compute cores and related technologies together as a single platform. Now, Arm is doing the same for autonomous systems. Arm introduced the Cortex-A78AE Central Processing Unit (CPU) core, the Mali-G78AE Graphics Processing Unit (GPU) core, and the Mali-C71AE Image Signal Processor (ISP) all based on the standard Arm products but designed for the unique requirements of autonomous systems. For example, the Cortex-A78AE not only provides 30% more performance than the previous generation Cortex-A76AE, it also offers a new hybrid operating mode.

To understand the value of these products and the new operating mode of the Cortex-A76AE, it is important to understand that various applications within an autonomous platform will have different levels of safety requirements depending on the applications ability to impact the proper operation of the platform and the potential to cause damage or harm. In vehicles, for example, applications like the instrument cluster, external lights, and sensor systems may require a relatively low level of safety certification referred to as ASIL-B, while the airbag, engine, braking, and steering systems will require the highest level of safety certification ASIL-D. Arm has developed CPU cores to operate independently, referred to as “split mode” for those lower safety levels and in “locked mode” where certain cores and resources are completely synchronized to check for any operating errors. The locked mode has a performance impact because certain cores and resources will not be available when operating at the highest safety levels. But, even in split mode, there is a performance hit because cores or shared logic are unavailable during periods of error checking. The new hybrid mode allows the shared logic to be locked to prevent having to take the cores or logic offline for even brief periods. This allows for even higher performance and efficiency when running lower safety level applications.

The Mali-G78AE is Arm’s first GPU designed specifically for the safety critical requirements of autonomous systems. The Mail-G78AE supports up to four independent partitions allowing for multiple simultaneous workloads. Additionally, it supports virtualization so that there can be multiple workloads within each partition. It also supports the ISO 26262 and IEC 61508 safety standards, which is suitable for certifications like ASIL-B in automotive. View More