In the tide of continuous integration of technology and the automotive industry, the intelligence of automobiles has gradually become a new highland that major car companies strive to pursue. Gaohe Automobile, the global benchmark brand of luxury electric vehicles, frequently leads the industry trend with its unique market strategy and cutting-edge technology.
At the recently held "Gaohe Wings Day", Gaohe Automobile not only demonstrated its hard verification power in intelligent technology, R&D system engineering capabilities, etc., but also launched a self-developed intelligent cockpit platform with high computing power. Based on the FPGAs, automotive-grade MCUs and automotive-grade gateways of aviation-grade/automotive-grade dual standards, the platform is the first to use the industry-leading Qualcomm 8550 chip through chip parallel connection and automotive-level large system development, bringing users a convenient, smooth, freely expandable and sustainable intelligent cockpit experience with the advantages of high reliability and high computing power.
This initiative not only re-establishes the leading position of Gaohe Automobile in the global intelligent technology field, but also sets a new technical standard and development direction for the new energy automobile industry, which will effectively promote the entire automotive industry to a higher realm of intelligent technology.
Car intelligence is not as simple as imagined
Many consumers have such confusion, why is the experience of a luxury car worth hundreds of thousands of dollars not as good as a smartphone of thousands of yuan? Why are the convenience features of smartphones difficult to achieve in the car, or later than mobile phones?
In fact, the gap between smart phones and vehicle experience is not only a problem at the hardware level, it involves a variety of factors such as vehicle environment, driving safety, and system stability, so the development and iteration cycle is far more than that of smart phones. In order to meet the needs of users for vehicle experience more quickly, many automobile brands choose to replace or upgrade the latest automotive-grade chips, which can indeed improve the computing and response speed of the vehicle to a certain extent, but it is difficult to fundamentally solve the problems of overall system optimization, interactive experience and application ecology. If the chip upgrade path is adhered to, then the vehicle system may always be difficult to achieve an experience comparable to that of a smartphone.
This means that in order to completely solve the market pain points from the core level, we must not only stop at the technical upgrade of automotive-grade SoC chips, but need to explore to a higher system level, and even draw experience and inspiration from completely different industries. What Gaohe pursues is to provide users with a truly forward-looking, safe and reliable intelligent car experience from a deeper and broader perspective.
The practice of multiple industries has proved that distributed systems have higher reliability and stability than centralized systems. Based on this insight, Gaohe Auto has opened a new research and development path. After in-depth research and extensive testing, Gaohe has come to a very important conclusion: through the method of chip parallel connection and automotive-level large system development, the entire system can achieve automotive-level reliability. Therefore, Gaohe has developed a high-computing power intelligent cockpit platform.
Adhering to the attitude of excellence, Gaohe decomposes the development and research of key components, not only deeply optimizes the key components, but also studies and improves every interface and function of the vehicle system to ensure that every detail reaches the best working condition. For example, in the instrument display, the FPGA chip with the dual standard of vehicle specifications and aviation grade is connected in parallel with the SoC to achieve the instrument without black screen; In the gateway and vehicle control MCU interface between the SoC and the whole vehicle, mature automotive-grade products are adopted; Based on the RTOS real-time operating system, develop full-featured vehicle control software to fully avoid communication and control failures.
Knock on the blackboard here. Safety is the core of Gaohe Automobile, and the software and hardware design of its high-computing intelligent cockpit platform are strictly implemented in accordance with IEC 61508 and aviation-grade DO-254/178 standards, and functional safety has reached ASIL B level in ISO 26262 standard. At the same time, relying on nearly 100 vehicle test standards accumulated by traditional car companies for decades, Gaohe has carried out all-round development and testing of the platform to ensure that the platform fully meets the conditions of use in various extreme environments, so as to achieve high safety and reliability of the entire system. Based on this ability, Gaohe uses Qualcomm QCS8550 chips with top computing power on the vehicle machine.
Qualcomm 8550+GPT, Gaohe breaks the limit of intelligent cockpit computing power
As we all know, the Qualcomm 8295 to be used in the market is already a high-performance chip recognized by the industry, but based on chip AI computing power, application ecology and other considerations, the high-tech intelligent cockpit platform uses Qualcomm QCS8550 chips. This chip adopts TSMC's 4nm process technology, its computing power is as high as 96TOPS, and it fully supports the latest mobile Internet technology, so the performance is better. In terms of AI computing power, its 96TOPS performance is 3 times that of the 8295 chip, breaking the limit of computing power of the intelligent cockpit.
"Transformer model" may be an unfamiliar term to many, but in the field of artificial intelligence, it is a revolutionary technology that has demonstrated superior performance when performing tasks such as natural language processing and image recognition. But because of its size, few devices can run it locally, and most rely on cloud computing. Gaohe Auto has achieved a breakthrough in running this large model locally through Qualcomm QCS8550 chips. This means that compared with traditional vehicle systems, Gaohe Automobile will be faster and more accurate when processing voice commands, road condition recognition and other AI-driven tasks. All data processing can be done locally on the vehicle, and the user's privacy protection is also realized.
Another revolutionary feature brought by QCS8550 chip is its support for hardware ray tracing. Ray tracing is an advanced graphics rendering technique that simulates how light travels and reflects in a scene to create realistic effects. In the field of games and movies, this technology has been widely used. Now, Gaohe Auto has brought this technology to the vehicle-machine interface, making its display effect no different from that of top games, bringing users an immersive experience.
At the same time, based on the self-developed high-computing power intelligent cockpit platform, Gaohe and Microsoft also jointly released a GPT-based local voice model. Unlike traditional in-vehicle voice assistants, which rely on the cloud for voice recognition and processing, this model uses the latest chips to successfully deploy advanced recognition synthesis technology that was previously only available in the cloud to the vehicle. At the same time, the speech model adopts multi-language support and successfully breaks the barrier of dialect recognition with the help of massive data resources. Whether users use Mandarin, dialects, or a mix of languages, voice assistants can accurately recognize them. In the future, no matter what language users use, Gaohe can quickly and accurately serve them.
With the self-developed high-computing power intelligent cockpit platform, Gaohe can not only achieve perfect integration with the current mainstream flagship chips, but also deeply integrate with future domestic leading high-end chips such as Qilin. This means that Gaohe has not only always been at the forefront of the development of scientific and technological intelligence, but also opened up a broad path for the integration of resources between industries, and laid a solid foundation for the in-depth application of on-board artificial intelligence technology.
In June 2023, the first engineering prototype vehicle equipped with its self-developed high-computing power intelligent cockpit platform has been lit up and started middleware debugging, and it is planned to conduct small-batch internal testing on the HiPhi X at the end of this year, and achieve batch testing in the first quarter of 2024. At that time, users will experience an unprecedented intelligent driving experience.
Empowering the future travel experience with intelligent technology, Gaohe does what it says. Since its establishment, Gaohe Automobile has adhered to independent innovation, and now it has established full-stack self-research capabilities in design, vehicle engineering, powertrain and vehicle intelligence. The release of Gaohe's self-developed high-computing power intelligent cockpit platform further confirms its leading position in the field of scientific and technological intelligence. It is hoped that in the future, Gaohe will continue to uphold the original intention of "creating works", focus on technological innovation and upgrading, and promote the rapid development of the new energy industry with the attitude of a global benchmark brand of luxury pure electric vehicles.