Recently, Great Wall Motors held a coffee smart 2.0 upgrade conference, announcing the launch of its third-generation autonomous driving computing platform ICU3.0 based on the Qualcomm Snapdragon Ride platform. level autonomous driving capability. This release shows the latest cooperation progress between the two parties in the field of mass production and commercial use of high-level autonomous driving technology.

As the world’s highest computing power for mass-produced autonomous driving computing platform, Great Wall Motors autonomous driving computing platform ICU 3.0 is equipped with Qualcomm Snapdragon Ride solution. While bringing high computing power, the Snapdragon Ride platform also provides ICU3.0 with an extremely high energy efficiency ratio of up to 5.5TOPS/W and excellent heat dissipation performance.

In January 2020, the Qualcomm Snapdragon Ride platform was officially released, providing the automotive industry with economical benefits for the automotive industry with its high-performance and energy-efficient hardware, industry-leading AI technology, flexible and scalable open architecture, and a groundbreaking autonomous driving software stack. Efficient complete system level solution. The Snapdragon Ride platform can provide different levels of computing power, ranging from 10 TOPS for automotive windshield ADAS cameras to more than 700 TOPS for fully autonomous driving scenarios with a power consumption of less than 5 watts. The cooperation between Qualcomm and Great Wall Motor is greatly improved by the combination of Snapdragon Ride ADAS SoC and AI accelerator, thus bringing L4-level autonomous driving capability in limited scenarios to the scalable computing power center of Great Wall Motor.

Highly complex autonomous driving places extremely stringent requirements on computing platforms. For example, sensor fusion and image processing have brought great performance challenges to CPUs, GPUs, and dedicated AI accelerators. The computing platform must also meet vehicle-level requirements such as safety, robustness, high performance, and efficient heat dissipation. Integrating solutions with the highest level of functional safety is crucial for commercial implementation and large-scale deployment of ADAS and autonomous driving. Snapdragon Ride’s ADAS SoC series and accelerator series use heterogeneous computing, fully considering the needs of the application at the beginning of the design. Leveraging Qualcomm’s next-generation artificial intelligence engine, its ADAS SoCs and accelerators can efficiently manage large amounts of data from in-vehicle systems, including ISPs for camera sensors, enhanced DSPs for sensor signal processing, high-performance CPUs for planning and decision-making, support Cutting-edge GPUs for high-end visualization and immersive user experience, dedicated safety and security subsystems across SoCs and autonomous driving accelerators. Through the self-driving accelerator, Snapdragon Ride can bring energy-efficient computing power and powerful cooling performance to mainstream models (supporting passive or air-cooled cooling design), and the same function is still far from being supported by the automotive industry due to high complexity. and expensive cooling solutions cannot achieve mass adoption due to stringent power consumption requirements.

In December 2020, Great Wall Motor and Qualcomm announced their cooperation in the field of autonomous driving for the first time, and will use the Qualcomm Snapdragon Ride platform in the high-end models mass-produced in 2022 to build the Great Wall Motors coffee intelligent driving system. Only half a year later, the cooperation between the two parties has progressed rapidly. Relying on Snapdragon Ride, Great Wall Motors released the third-generation autonomous driving computing platform ICU3.0 and specified the mass production time as the second quarter of 2022.

In addition to announcing the mass production time, Great Wall Motor also announced that its ICU 3.0 platform will become the autonomous driving computing platform with the largest computing power in China in 2022 and the first platform in China to support L4 autonomous driving capabilities in composite scenarios. The period from 2020 to 2030 will be the “golden decade” for the development of autonomous driving. In the Chinese market, Great Wall Motors and Qualcomm fully integrated the advantages of both parties and took the lead in accelerating the implementation of autonomous driving.

Source: Gasgoo