ARM's latest chip is its first one built for self-driving cars
World class processor with integrated safety
Lakshmi Mandyam – VP of Automotive at ARM’s technology stated in her blog, “If consumers don’t trust the autonomous systems in their cars and fail to accept that these vehicles are safe, and then mass market acceptance of this technology will be slow to happen.”
There is a truth in that, irrespective of how advanced these technologies gets, if we don’t put our trust in them, we will never use it, and us the creators of these technologies become the hindering factors. This is why tech firms such as ARM, walk the path of creating breakthroughs that we can trust.
There is the only number of tech firms currently in the industry that can match ARM’s track record of delivering products which embeds the latest functional safety standards across the entire vehicle infrastructure. In fact, looking at the published statistics, the presence of Arm intellectual property (IP) is truly phenomenal with 65 percent of today's automated driver assistance (ADAS) applications running on Arm IP as well as 85 percent of in-vehicle information systems.
Announcing the Cortex-A76AE
Cortex-A76AE and the future automotive enhanced (AE) processors are a direct result of close working relationships with car manufacturers aimed to provide higher performance and efficiency, without sacrificing functional safety standards and certifications.
Arm’s AE processors will be the most safety-critical products currently available which has undergone through a rigorous functional safety process and include safety innovations like dual-core lock-step to ensure redundancy in CPU cores, and as Arm develops more purpose-built automotive compute IP such as ML processors and GPUs, you can be certain the same safety-first principle will guide those designs.
A faster processor coupled with higher-bandwidth and lower-latency 5G network built for billions of connected devices will keep vehicles connected, enabling faster communication between the vehicles on the road and real-time cloud update, which quickly alert other vehicles to potential hazards another vehicle's data may provide.
Main benefits of Cortex-A76AE
The Cortex-A76AE is purpose-built for functional safety applications such as ADAS and autonomous vehicles. Let’s have a look at the main three benefits of the Cortex-A76AE.
1. Safety for autonomous systems
Where the Cortex-A76AE really stands out, is in its ability to deliver the aforementioned performance, at high safety integrity that includes the Dual Core Lock-Step (DCLS) and Split-Lock features.
The flexibility offered through Split-Lock confers additional safety and provide the ability to continue to operate in a degraded mode rather than completely shutting the system down. If one core starts to exhibit a failure condition, the system could be quiesced and the faulty core be taken off-line (split) allowing continuation in a degraded mode of operation. This ‘split available" capability is critical for any autonomous system.
2. Performance for ADAS and Autonomous Driving
Cortex-A76AE has been designed to act as the decision engine in next generation ADAS and Autonomous Vehicle systems. It delivers 30% uplift in performance over its predecessor, the Cortex-A75, and 60% increase in performance over Cortex-A72. This massive boost in performance meets the emerging CPU requirements for autonomous driving.
3. Flexibility in mixed criticality systems
As mentioned the next generation ADAS and autonomous driving systems will consist of multiple applications running at different levels of safety criticality. This presents a challenge to processor providers when scoping the design.
Cortex-A76AE solves the challenge of mixed criticality applications through its ability to operate in two modes, performance mode and safety mode. In performance mode, all cores within a cluster operate as Symmetrical Multiprocessors (SMP). In safety mode, pairs of Cortex-A76AE cores in a cluster are configured to run in Lock-Step as discussed previously.
The mode of operation can be changed at any time through reset. This means that a Tier1 or car manufacturer is able to tune the platform to fit any mix of safety critical applications, post production.
Efforts to develop fully autonomous self-driving cars have continued to close the gap over the past few years, and now ARM has unveiled its first chip designed specifically for the task. The company's technology already powers many mobile devices, IoT and is in cars today, but this Cortex-A76AE (Automotive Enhanced) chip combines the high power and efficiency we're used to in smartphones with "Split-Lock" tech focus on safety.
If one of these chips is processing the code that drives your car, a crash would be disastrous, which is why the DynamIQ multi-core processing we've seen used to boost AI recently is being paired with Split-Lock. Now you have a fast 64-bit processor that can switch between modes, with the "Lock" ability running identical code on two cores for safety and checking, or "Split" where different cores can deal with different tasks. There's even support for an additional CPU core that can act as a "Safety Island" checking the other outputs for redundancy, with the ability to disable some and run in a lower-speed mode if any problems are detected.
With all the information flowing through the processor of an autonomous car, getting both performance and reliability -- while maintaining efficiency, which ARM claims can measure at 250K DMIPS at less than 15 Watts -- out of the 7nm hardware will be key. As far as competition, look no further than Intel & Mobileye. ARM's roadmap for these AE chips extends to 2020 and beyond, and it's likely that a self-driving car you see on the road on the future will have something relying on the design inside of it.