At its 2024 Snapdragon Summit in Maui, HI, Qualcomm Technologies, Inc. today unveiled its most powerful automotive platforms. Powered by Oryon, the company’s fastest CPU now tailored for automotive safety and reliability, the “Elite tier” Snapdragon Digital Chassis platforms are designed to bring next-generation power and intelligence to SDVs (software-defined vehicles). The company’s CEO Cristiano Amon kicked off the Snapdragon Summit by mentioning the new Snapdragon Cockpit Elite and Ride Elite offerings, hoping to quickly add new as well as existing customers like General Motors, Rivian, BMW, Ford, Sony Honda Mobility, some of which were highlighted during the event, to the new technology’s customer list.

“Qualcomm Technologies remains at the forefront of innovation with platforms like Snapdragon Cockpit Elite and Ride Elite, as the automotive industry evolves towards centralized computing, software-defined vehicles, and AI-driven architectures,” said Nakul Duggal, Group Manager, Automotive, Industrial, and Cloud, Qualcomm Technologies, Inc. “With our strongest performing compute, graphics, and AI capabilities, coupled with industry-leading power efficiency and cutting-edge software enablement for digital cockpits and automated driving, these new elite Snapdragon automotive platforms address the industry’s needs for higher compute levels, empowering automakers to redefine automotive experiences for their customers.”

Just before the Snapdragon Summit, Qualcomm briefed a select group of media on what to expect from the latest elite version of its latest automotive tech. Mark Granger, Senior Director of Product Management for Automotive at Qualcomm Technologies, previewed the big-picture improvements of the offering.

“Qualcomm is known in the automotive space for having a very scalable platform,” said Granger, who highlighted the top capabilities of the new platform. “For instance, three times performance increase; taking the Orion CPU, customizing it for automotive versus our previous platforms; up to twelve times the AI performance. So, huge increase there as we see the advent of large language models and multimodal models that are well adept for running on the edge.”

Qualcomm aimed to build upon its reputation for industry-leading power efficiency. The platform was engineered to deliver exceptional performance while minimizing energy consumption and helping ensure that vehicles operate smarter and longer. The solution is a combination of intelligent power management hardware and software that balances core utilization and application runtime.

“The whole intention of the Snapdragon Elite products is to be pushing the boundary of what’s possible—on the AI side, bringing this huge increase in performance, but also with the DNA of Qualcomm of low power or high performance per watt,” said Granger. “We believe we differentiate there.”

 

Software-defined and virtualized with multi-OS support

Built for the industry’s shift to SDVs, the elite-tier platform is designed to take an end-to-end approach for enhanced safety, security, and upgradeability. Its unified software framework emphasizes software reuse and is designed to help automakers accelerate feature development via a cloud-based workbench, streamlining software development for continuous improvement and reducing time-to-market for new features and services.

The platform uses Qualcomm’s comprehensive software stack, virtualization, and support for multiple operating systems to drive innovation. Snapdragon Ride Elite offers an end-to-end automated driving system with advanced vision perception, sensor fusion, path planning, localization, and complete vehicle control. Snapdragon Cockpit Elite provides support for rich multimedia features, on-device AI with a fully integrated edge orchestrator, optimized gaming, and advanced 3D graphics for rich user experiences. It comes with safety, security, and long-term API compatibility support features built into the design.

“In terms of the GPU, and tying that to our display processing unit, we are delivering hundreds of millions of vehicles already to the field with virtualized environments that safely display beautiful graphics,” said Granger. “And now, you’re taking that to the next level to drive high-resolution multi-display capabilities, even utilizing, for instance, ray tracing technology to deliver a visual experience that you’ve just never heard of.”

While taking more of the best of Qualcomm’s cross-industry technology, Granger got more granular of the company’s modifications for automotive duty.

“Our packaging of the Snapdragon Cockpit and Ride Elite products are such that you have the device on a substrate and we’re having memory there in a specific module a thermal enclosure. This is for things like the DRAM,” he said. “Then we tie that with our PMIC solution to get the absolute best power efficiency as well as fine grain control across the different blocks. For UFS and telematics, that will be potentially in the same ECU or separately. Just to recap, the Snapdragon Cockpit and Ride Elite involve SOC, DRAM, and PMIx, all in a co-optimized packaging solution.”

 

Flexible, centralized processing to support consolidation

Automakers can choose Snapdragon Cockpit Elite to power advanced digital experiences and Ride Elite to power automated driving capabilities separately or take advantage of the new, more flexible architecture to combine both digital cockpit and automated driving functionalities on the same SoC.

“This is truly an example of consolidation,” said Granger. “It’s a performant device, the Snapdragon Ride Elite, bringing ADAS and cockpit onto a single device for a centralized controller. We also bring in this functional safety with a so-called Safety Island that allows you to minimize additional MCUs in the vehicle. So, it’s perfectly dovetailed to the centralization. Something that Qualcomm has done with each generation already, bringing more capability, more ability to integrate.”

The platform is said to run multiple applications without performance loss, offering concurrency and multitasking for numerous cameras, sensors, rich user experiences, and advanced AI-enabled audio with virtualization. Automakers can create configurable SDVs for all model or brands in their portfolios, providing flexibility and scalability while simplifying vehicle architecture. The architecture results in accelerated deployment schedules, ensuring customers can enjoy the latest innovations and features more quickly than before.

 

Accelerated AI performance for intuitive and safe experiences

Regarding performance specs, the company like many in the industry is downplaying some typical measures like TOPS (tera operations per second) and instead focusing on functionality, “having the AI performance for ADAS and cockpit, being able to bring that all together,” said Granger.

The accelerated AI performance is enabled by a dedicated neural processing unit (NPU) designed for multimodal AI. It’s a primary ingredient in the secret sauce that offers a twelve times performance boost over previous company cockpit platforms, enabling real-time external environment and cabin data processing.

This advancement facilitates live decision-making, adaptive responses, and proactive assistance for more personalized in-cabin experiences. Equipped with transformer accelerators and vector engines, along with mixed precision support, the NPU in Snapdragon Ride Elite is designed to deliver low-latency, highly accurate, and efficient end-to-end transformers, maintaining optimal power and performance.

The platform is designed to enable hands-free, unsupervised automated driving that anticipates needs, along with real-time driver monitoring and enhanced object detection for a safer ride. It meets the highest level of automotive safety standards, ASIL-D, as a result of the dedicated safety island controller and robust hardware architecture for isolation and interference-free operation, helping to ensure reliable quality-of-service for specific ADAS functions.

Qualcomm’s elite-tier platforms feature a powerful but efficient camera system with an advanced image signal processor (ISP) for clear, responsive visuals in extreme driving conditions. They support over 40 multimodal sensors, including up to 20 high-resolution cameras for 360-degree coverage and in-cabin monitoring.

 

Cockpit ‘sound bubbles’ and NPUs

The accelerated AI performance benefits intuitive experiences in the cockpit. The platform’s improved Adreno GPU is targeting a three-times performance boost with advanced rendering capabilities, meeting greater demands for gaming, multimedia, and dynamic driver information.

“From the cockpit side, we see a huge need for edge AI to handle multimodal, but also multiple large parameter models running on the edge so that you can have wonderful virtual assistants,” said Granger. “Thinking about multimodal, using all that camera feed that you have coming into the vehicle, using all the context of the geolocation, orchestrating that together, and being able to deliver compelling use cases that start to tie in the visual, audio, as well as all that contextual data available in the car to bring accurate natural assistance.”

With the dedicated AI audio engine of the elite family, Qualcomm is “massively increasing” its capabilities over its current-generation products, said Granger.

“We’ll be leading there to bring, for instance, personalized sound bubbles in the car so that users can have their own space, two kids in the backseat playing two different video games, your partner in the passenger seat watching a video,” he said. “And you driving, with your eyes on the road, but at the same time being able to do productivity and also interacting in an audio bubble, so that you have privacy as well as unique experiences within the vehicle, all enabled by the processing.”

Voice recognition is getting a boost, especially for Chinese and Japanese languages that have been very tricky for traditional natural language understanding tech.

“Utilizing even a small LLM (large language model), you can get great voice recognition and speech-to-text performance that has never been seen,” said Granger. “We have this dedicated audio processing block that has embedded NPUs to enable that type of a use case.”

The NPUs can also benefit electric-vehicle battery monitoring and extending range.

“Today most EVs have a battery management ECU that’s “been designed with frankly quite little performance,” said Granger. “We’re working with our third-party ecosystem who are bringing algorithms to run on our device to utilize the NPU for battery life, charging, the health of the cells. We bring in low-power capabilities, the DNA of Qualcomm, which is ultimately extending the battery life of the vehicle.”

 

Roll out for accelerating auto development

While all performance targets compared to the previous generation are based on preliminary internal testing and subject to change upon final validation, final numbers for Snapdragon Cockpit Elite and Ride Elite will be available as sampling begins in 2025. That validation will be followed more quickly by start-of-production, to fit with the acceleration in product development cycles brought by tech companies, startups, and new OEMs entering the space, especially in Asia.

“The average cycle time has certainly come down over the past many years and continues to accelerate, especially you see that here in China is very rapid right now,” said Granger, who was answering the media questions virtually from there. “We’re sampling these devices next year to our customers, and for one of our lead SOPs, the launch of the vehicle is expected to be in 2026, which is a very fast cycle time for automotive as compared to 10 years ago, where three years was considered a quite fast cycle time.”

As the auto portion of the Snapdragon Summit began, Qualcomm announced that technology collaborations are underway with leading car manufacturers Li Auto and Mercedes-Benz AG, who will feature Snapdragon Elite platforms in their future vehicles.

“We are incredibly excited about the new Snapdragon Cockpit Elite and Snapdragon Ride Elite solutions and the transformative potential they hold for next-generation vehicles,” said Donghui Ma, President of Li Auto. “The automotive industry is on the cusp of a revolution, leveraging the latest advancements in computing, artificial intelligence, and software to deliver unparalleled experiences for drivers and passengers alike.”

In 2015, Mercedes-Benz and Qualcomm first announced a strategic collaboration on connected car technologies.

“We are pleased to continue our trusted collaboration with them by integrating Snapdragon Cockpit Elite into our future vehicles,” said Magnus Östberg, Chief Software Officer, Mercedes-Benz AG. “This powerful and efficient central compute technology will enable us to provide our customers with an unmatched in-vehicle experience—one that is uniquely tailored, seamlessly connected, and remarkably intuitive.”

Qualcomm also announced a multi-year technology collaboration with Google leveraging technologies from the Snapdragon Digital Chassis, Android Automotive OS, and Google Cloud to produce a new standardized reference platform for the development of cockpit solutions using generative artificial intelligence (Gen AI). Underpinning the delivery of this framework will be Google AI, to create Gen AI-enabled in-car experiences like intuitive voice assistants, immersive map experiences, and real-time updates to anticipate driver needs, powered by Snapdragon edge AI SoCs and Qualcomm AI Hub, which simplifies deploying AI models for vision, audio and speech applications to the cockpit.

“Our technology collaboration with Google to drive innovation and develop cutting-edge, Gen AI-enabled solutions is a significant milestone in unlocking new possibilities for automakers and Tier One suppliers while empowering the industry to create safe and digitally advanced experiences,” said Duggal.

“The automotive industry is on the cusp of major transformation enabled by advancements in generative AI, automated driving solutions, and the ability to update software features throughout the platform’s lifecycle,” added Patrick Brady, VP of Engineering at Google.

So, the market will see Snapdragon Elites on the road sooner than expected.