Arm technology is at the brim of computing and data revolution that is transforming the way people live and businesses operate.
Recently Arm announced the latest computing technologies to improve automated decision-making and safety capabilities through automotive and industrial applications. The new IP suite includes the Arm ® Cortex ® -A78AE CPU, Arm Mali™-G78AE GPU, and Arm Mali-C71AE ISP, designed to work together in conjunction with support applications, tools, and IP framework to help silicone providers and OEMs to develop autonomous workloads. These technologies can be used in a variety of applications, from allowing more intelligence and automation in smart manufacturing to improving ADAS and automated cockpit applications in the automotive industry.
Cortex-A78AE: strong performance in critical safety applications
The new Arm Cortex-A78AE CPU is Arm’s newest, high-performance safety-enabled CPU, providing the ability to run separate, dynamic workloads for stand-alone applications such as mobile robotics and driverless transport. It’s delivering:
A 30 per cent improvement in output relative to its predecessor.
Supports software to meet applicable automobile and industrial practical protection requirements, ISO 26262 and IEC 61508 for implementations up to ASIL D / SIL 3.
New improved Split Lock feature (Hybrid Mode) for optimum flexibility. Hybrid Mode is designed to explicitly allow applications that reach lower levels of ASIL specifications without losing performance and allow the same SoC computing architecture to be implemented in different domain controllers.
Mali-G78AE: protection redefining for embedded GPUs, with modular partitioning
Mali is the world’s leading shipping GPU, and the new Mali-G78AE is Arm’s first GPU designed for safety, bringing rich user interface and heterogeneous computing to safety-critical autonomous applications. The latest Mali-G78AE allows:
A modern solution to autonomous GPU workloads with Modular Partitioning, with up to four entirely functional workload isolation partitions for safety use cases. GPU resources can now be used for security-enabled human-computer interfaces or heterogeneous computing required in autonomous systems.
Mali-C71AE: an evolution in the defence of ISPs
Autonomous workloads need to be conscious of their environments, often by cameras that may run under a wide variety of lighting conditions. The functionality needed to support both human and machine vision applications such as production line surveillance and ADAS camera systems. Improved safety enhancements support ASIL B / SIL2 safety capabilities.
As autonomous systems migrate towards more user-defined capabilities, Arm is working to promote the development of applications that can completely understand the advantages of these emerging innovations through projects such as Project Cassini, aimed at laying the groundwork for the implementation of cloud-based native software paradigms across the entire edge of computing. Arm is also collaborating with several open-source organizations and specialized software suppliers to make the autonomous computing platform more available, to implement technologies from existing cloud-native ecosystems, and to collaboratively push new technology to enable the functionality needed for autonomous workloads.
