The automotive industry is not known for its speed… in adopting new technologies and incorporating them into new vehicles, that is. It is well known that it took decades for power windows to become standard on most cars, and even now, the migrations to electric vehicles, software-defined vehicles and autonomous driving are occurring more slowly than originally anticipated.
Arm is betting that the sector is going to be much more aggressive about adopting AI, and it is ready to help automotive companies reduce the amount of time it takes to engineer Ai capabilities into new vehicles. The company this week launched its Arm Zena Compute Subsystems (Zena CSS), described as a “pre-integrated and pre-validated platform for AI-defined vehicles” that consists of 16 Armv9-based Cortex-A720AE cores, which the company said have been “performance-optimized for ADAS and IVI applications.”
As automotive companies look to energize traditionally “static” technology development processes, Zena CSS will allow them “to launch new vehicle models at least one year faster,” enable software development to start “before silicon even exists,” and ultimately allow then to use up to 20% fewer engineers per project, said Dipti Vachani, senior vice president and general manager, Automotive Line of Business at Arm.
The 16-core CSS platform launch comes about 15 months after Vachani said as part of an earlier automotive announcement that Arm would debut a new CSS for automotive in 2025.
Zena CSS also includes the following:
- Arm’s Safety Island system for real-time ASIL D processing capabilities such as fault management, safety monitoring, system control and SoC boot.
- Runtime Security Engine with safety-capable Hardware Root of Trust (RoT), leveraging Arm TrustZone technology to manage SoC-level security.
- CPU coherency and chip-to-chip connectivity provided by the Arm NeoVerse CMN S3AE.
- Optional Image Signal Processing powered by Arm’s Mali-C720AE and Mali GPU for ADAS use-cases including surround view and driver-monitoring.
- Support for easy integration of accelerators and partner-specific logic to meet evolving workload demands for advanced, AI-capable SoC design.
The Zena CSS launch was supported by statements from many of Arm’s automotive partners. “The transition to software-defined vehicles demands a shift in how we approach compute architectures,” said Magnus Östberg, Chief Software Officer, Mercedes-Benz AG. “Standardized, pre-verified compute subsystems like Arm Zena CSS can significantly accelerate development timelines and reduce complexity across the industry. By integrating such solutions early in the development chain, the automotive industry can streamline complexity and establish a stronger foundation for delivering the safe, intelligent, and connected experiences that define the next generation of vehicles.”
Shipments of Arm-based chips to the automotive industry have tripled over the last five years, Vachani said, as its devices have been engineered into vehicles from the likes of Tesla, Rivian, NIO, Mercedes-Benz, Honda, and Geely.
Arm said its automotive ecosystem partners can begin software development on Zena CSS immediately using cloud-based virtual platform support from AWS, Cadence, Siemens and Synopsys.