Snapdragon Ride can provide cost-effective and energy-efficient complete system-level solutions for the complex needs of autonomous driving and ADAS (Advanced Driver Assistance Systems). Snapdragon Ride combines a unique SoC, accelerator, and autonomous driving software stack to provide automakers with a scalable solution that supports three segments of autonomous driving systems: L1 / L2 level initiative Safe ADAS-for cars with automatic emergency braking, traffic sign recognition and lane keeping assist; L2 + level "Convenience" ADAS-for autonomous driving on highways, support for self-parking, and Cars driving in urban traffic environments with frequent parking; L4 / L5 level fully autonomous driving-for autonomous driving, robotic taxis and robotic logistics in urban traffic environments.
The Snapdragon Ride platform is built on a series of different Snapdragon automotive SoCs and accelerators, and uses scalable and modular high-performance heterogeneous multi-core CPUs, energy-efficient AI and computer vision engines, and industry-leading GPUs. Based on different combinations of SoCs and accelerators, the platform can match the needs of each segment of autonomous driving and provide industry-leading cooling efficiency, from 30 TOPS-class devices for L1 / L2-level applications to L4 / L5 level driving, 130 Watt power consumption device over 700 TOPS.
Therefore, the platform can support passive or air-cooled heat dissipation design, thereby reducing costs, improving reliability, eliminating expensive liquid cooling systems, simplifying vehicle design and extending the driving range of electric vehicles. Snapdragon Ride's range of SoCs and accelerators are designed for functional safety ASIL-D (Automotive Safety Integrity Level D) systems.
Snapdragon Ride will be delivered to car manufacturers and Tier 1 suppliers for pre-development in the first half of 2020. Snapdragon Ride-equipped cars are expected to enter production in 2023.
Nakul Duggal, senior vice president of product management at Qualcomm Technologies, Inc., said, "In the past few years, we have demonstrated the company's technical strength in the automotive field through the large-scale deployment of solutions, that is, for almost all types of cars. Now we are very Happily launching the first generation of Snapdragon Ride platform as a highly scalable, open, fully customizable, and highly optimized power consumption autonomous driving solution, which can meet the requirements of New Car Evaluation Specification (NCAP) to L2 + level highway autonomous driving To a range of needs for robot taxis. "
Security SoC and accelerator software and hardware solutions
Qualcomm believes that the next innovation in the automotive sector is expected to be "convenience" ADAS at the L2 + level, and regulatory requirements are also driving the hardware solution in the Snapdragon Ride platform from a single SoC for active safety ADAS systems to a highly scalable The architecture consists of multiple SoCs and dedicated autonomous driving accelerators to support fully autonomous autonomous driving systems.
Qualcomm's ADAS SoC family and accelerator family use heterogeneous computing, which fully considers the needs of the application at the beginning of the design. Leveraging Qualcomm Technologies ’next-generation artificial intelligence engine, the aforementioned ADAS SoCs and accelerators can efficiently manage large amounts of data for in-vehicle systems, including ISPs for camera sensors, enhanced DSPs for sensor signal processing, high-performance CPUs for planning and decision-making, Cutting-edge GPUs supporting high-end visualization and immersive user experience, dedicated safety and security subsystems across SoCs and autonomous driving accelerators.
Through autonomous driving accelerators, Qualcomm Technologies brings energy-efficient computing capabilities to mainstream cars, and such equivalent functions are currently far from being supported by the automotive industry because high complexity and expensive thermal solutions cannot be achieved due to strict power consumption requirements Popularization of scale.
Launches Snapdragon Ride Autonomous Driving Software Stack
Qualcomm Technologies' new software stack dedicated to autonomous driving is a modular and scalable solution integrated into the Snapdragon Ride platform, helping automakers bring greater safety and comfort to everyday driving, such as automatic navigation Humanoid highway driving, as well as offering modular options such as perception, positioning, sensor fusion and behavior planning.
Qualcomm Technologies' integrated automotive platform enhances the company's leadership in the fields of in-vehicle information processing, information audiovisual, and in-car interconnectivity, and drives business growth. The total value of orders in these areas is currently over $ 7 billion.
As the industry's leading supplier of in-vehicle information processing and automotive Bluetooth connectivity semiconductors, Qualcomm Technologies has won information AV and digital cockpit projects from 19 of the world's leading 25 car manufacturers. At present, all the major car manufacturers in the world have adopted Qualcomm Technologies ’rich automotive information processing, information audio and video, and in-car connected automotive solutions. They are also working with Qualcomm Technologies to deliver safe, reliable and efficient vehicles, including the Snapdragon Ride platform. solution.
Chet Babla, vice president of automotive business at Arm Automotive and IoT, said: "Integrating the highest levels of functional safety solutions is critical for large-scale deployment of ADAS and autonomous driving. Future travel requires cross-industry collaboration, and we are working with Qualcomm Technologies Working together, and integrating Arm's functional safety solutions into the Snapdragon Ride platform is an example of improving autonomous driving safety. "
John Koeter, Senior Vice President, Intellectual Property Marketing and Strategy, Synopsys, said: "Synopsys' investment in intellectual property rights in the International Organization for Standardization ISO 26262 certification, American Automotive Electronics Association AEC Q100, and quality management simplifies the development of next-generation ADAS . Synopsys' automotive-grade DesignWare® interface IP, ARC® processor IP, and the combination of STAR Memory SystemTM and Snapdragon Ride platforms help Qualcomm Technologies achieve high levels of functional safety and accelerate certification of their SoCs. "
John Wall, senior vice president and co-general manager of BlackBerry QNX, said, "BlackBerry QNX is pleased to partner with Qualcomm Technologies and Snapdragon Ride platforms. QNX® OS Security and QNX® Hypervisor Security provide safe, secure and reliable autonomous driving platforms. The software foundation of Qualcomm Technologies. As Qualcomm Technologies further expands the product layout for the field of autonomous driving, we also look forward to deepening the cooperation between the two parties and ushering in new opportunities. "
Ritesh Tyagi, General Manager, Automotive Business, Infineon Silicon Valley Innovation Center, said: "Infineon has always been committed to delivering core semiconductor technology to make cars safer, smarter and more environmentally friendly. Our AURIXTM microcontrollers integrate the company's trust in Computational expertise is a fundamental module in safety-critical automotive applications. Infineon and Qualcomm Technologies collaborate to design and provide reliable, scalable, and energy-efficient computing systems that will help drive highly automated cars in a wider market Achieve deployment. "
"We are very pleased to work with Qualcomm Technologies to develop a next-generation AUTOSAR architecture that can scale to meet customer needs, integrating EB corbos software and the Snapdragon Ride platform. Our products using Qualcomm Technologies technology will enable Tier 1 suppliers and automakers can bring ADAS to production cars faster and more efficiently. "
"On Semiconductor is pleased to integrate our ADAS family of sensors into the Snapdragon Ride platform. Through close cooperation with Qualcomm Technologies, our ADAS sensors can Seamless and robust interoperability with Qualcomm Technologies processors and software, helping automakers and tier one suppliers to quickly and cost-effectively roll out next-generation solutions. In addition, digital AV cockpits and autonomous driving systems In addition, the flexible addition of in-vehicle functions such as driver monitoring and in-vehicle monitoring will also significantly accelerate the popularization of these new features of safety and convenience. "
Advantages of the Snapdragon Ride platform
• Proven and integrated secure car support package supporting secure OS and hypervisor
• Security frameworks provided by leading automotive companies, including Adaptive Automotive Open System Architecture (Adaptive
AUTOSAR)
• Optimized and complete library of basic functions for computer vision, sensor signal processing, and standard computing libraries
• AI tools to improve model efficiency and optimize the operating environment of heterogeneous computing units
• Complete autonomous driving software stack for highway driving functions, such as the perception required for highway driving
And planning functions
• Cost-effective positioning solution supporting Qualcomm® vision-enhanced high-precision positioning
• Hardware and software in a loop test environment
• Data management tools that support intelligent data collection and automatic annotation