Chip AI computing power 30tops

CristianoAmon, President of Qualcomm Corp, at the press conference of SnapdragonRide, shows a combination of SoC, accelerator and automatic driving software stack, showing SnapdragonRide (Graphic source CNET/James Martin)

SnapdragonRide, which provides an extensible solution for automobile manufacturers, which can provide support for self driving vehicles in three sub fields, namely

1, L1 / L2 active safety ADAS -- for vehicles with automatic emergency braking, traffic sign recognition and lane keeping assistance

2, L2 + ADAS -- for vehicles that can drive automatically on expressways, support self-service parking, and drive in urban traffic with frequent parking

3, L4 / L5 fully automatic driving -- for automatic driving, unmanned taxi and robot logistics in urban traffic environment

snapdragonride platform is based on a series of different snapdragon SOC and accelerators. It adopts scalable and molar high-performance heterogeneous multi-core CPU, energy-efficient AI and computer vision engine, and GPU

among them, adassoc series and accelerator series adopt heterogeneous computing. At the same time, using the new generation of artificial intelligence engine of Qualcomm, ADAS and SOC can efficiently manage a large amount of data of on-board system

thanks to the combination of these different SOC and accelerators, snapdragonride platform can be equipped according to the needs of different segments of autonomous driving market, and provide good heat dissipation efficiency, from 30tops level equipment for L1 / L2 level applications to 130W power consumption equipment for L4 / L5 level driving with more than 700tops

in addition, the new snapdragonride autopilot software stack is a molar and scalable solution integrated into the snapdragonride platform

according to the introction, the software framework of snapdragonride platform can host customer specific software stack components and snapdragonride autopilot software stack components at the same time

snapdragonride platform also supports passive or air-cooled cooling design, so it can further optimize vehicle design and improve reliability while recing cost

now, arm, blackberry QNX, Infineon, Xinsi technology, elektrobit, and Anson semiconctor have joined the auto driving circle of friends of Qualcomm and become the software / hardware suppliers of snapdragonride auto driving platform

the functional security solutions of arm, the automotive designware interface IP, arc processor IP and starmemory SystemTM of Xinsi technology, the OS Security version and hypervisor Security version of BlackBerry QNX, the aurixtm microcontroller of Infineon, and the ADAS series sensors of Anson semiconctor will be integrated into the automatic driving platform of Qualcomm

elektrobit also plans to cooperate with Qualcomm to jointly develop a new generation AUTOSAR architecture that can be proced on a large scale. Ebcorbos software and snapdragonride autopilot platform will be integrated on this architecture

it is understood that snapdragonride will be delivered to automobile manufacturers and first tier suppliers for preliminary development in the first half of 2020, and according to the estimation of qcomm technologies, the vehicles equipped with snapdragonride will be put into proction in 2023

for more than a decade, Qualcomm technologies, a subsidiary of Qualcomm, has been providing advanced wireless communication solutions for GM's connected vehicle applications, including the security applications supported by the Android devices on GM

in the fields of on-board information processing, information audio-visual and in car Internet, the total value of orders of qcomm technologies has exceeded 7 billion US dollars (about 48.7 billion yuan)

according to the information released by Qualcomm at the ces2020 press conference, so far, more than one million vehicles have used the automotive solutions provided by Qualcomm

it is obvious that today, the layout of Qualcomm in the automotive field has taken another step forward

ring ces2020, in addition to snapdragonride, Qualcomm also launched a new car to cloud service, which is expected to be available in the second half of 2020

according to the introction, the car to cloud service built by Qualcomm technologies supports the soft upgrade capability of softsku chip specifications, which can not only help automobile customers meet the changing needs of consumers, but also enable chipsets to be upgraded in the field to support new functions according to new performance requirements or new features

at the same time, softsku also supports customers to develop common hardware, so as to save their special investment for different development projects. By using the cloud softsku of Qualcomm, automobile manufacturers can not only provide a variety of customized services for consumers, but also create rich and immersive in car experience through personalized features

in addition, the car to cloud service of Qualcomm also supports the global cellular connection function, which can be used to guide the initialization service and provide wireless communication connection throughout the life cycle of the car

in addition, Qualcomm technologies also announced on ces2020 that it will continue to deepen its cooperation with general motors. As a long-term partner, GM will support digital cockpit, on-board information processing and ADAS (advanced driving assistance system) through continuous cooperation with Qualcomm technologies

conclusion: Giants raging like a storm in the field of automatic driving, HUAWEI has said before it wants to build intelligent core sensors for automobile, such as lidar and millimeter wave radar. After that, Arm led the establishment of the automatic driving Computing Alliance. Now the mobile chip giant Qualcomm has released a new automatic driving platform, and has made further progress in the field of automobile and automatic driving. <

P>

game player is concive to the faster and better landing of autopilot cars, but on the other hand, as more hardcore players expand their business boundaries, competition in the market will become more intense. p>

this article comes from the author of car home, which does not represent the standpoint of car home

The types of chips that provide computing power for AI include GPU, FPGA and ASIC

GPU is a kind of microprocessor specialized in image operation on personal computers, workstations, game machines and some mobile devices (such as tablet computers, smart phones, etc.). It is similar to Cu, except that GPU is designed to perform complex mathematical and geometric calculations, which are necessary for graphics rendering

FPGA can complete any digital device function chip, even high-performance CPU can be implemented with FPGA. In 2015, Intel acquired the FPGA long alter head with us $16.1 billion. One of its purposes is to focus on the development of FPGA's special computing power in the field of artificial intelligence in the future

ASIC refers to the integrated circuits designed and manufactured according to the requirements of specific users or the needs of specific electronic systems. Strictly speaking, ASIC is a special chip, which is different from the traditional general chip. It's a chip specially designed for a specific need. The TPU that Google recently exposed for AI deep learning computing is also an ASIC

extended data:

chips are also called integrated circuits. According to different functions, they can be divided into many kinds, including those responsible for power supply voltage output control, audio and video processing, and complex operation processing. The algorithm can only run with the help of chips, and because each chip has different computing power in different scenarios, the processing speed and energy consumption of the algorithm are also different. Today, with the rapid development of the artificial intelligence market, people are looking for chips that can make the deep learning algorithm perform faster and with lower energy consumption

About

} are black sesame CEO and coo Liu Weihong

black sesame CEO. Previously, he was a member of the founding team of omnivision, a global vision chip leader. He has worked in Silicon Valley chip instry for more than 20 years and has rich experience and technology accumulation in image processing chip and software algorithm

CTO Qi Zheng is the main design member of Intel Pentium II chip, CSO Zeng Daibing is the chief engineer of ZTE microelectronics, and coo Liu Weihong was the president of chassis and control system division of Bosch China ADAS

because of its strong R & D team, heima, a start-up company, can make the ADAS chip huashan-1 A500 within three years and launch it in mass proction. This year, it will launch huashan-2 a1000 chip and launch the fad automatic driving platform

since the beginning of this year, if new cars are not equipped with L1 / L2 automatic driving, they are "embarrassed to sell", the popularity of automatic driving is rapidly increasing, and the higher level of L3 or even L4 automatic driving is on the eve of mass proction, and the demand for automatic driving chips and computing platform solutions in the instry is explosive growth. The market scale of autopilot chips alone is expected to reach trillion US dollars, becoming the largest single market in the semiconctor instry

in August this year, FAW Intelligent Network Development Institute reached a technical cooperation agreement with heima. FAW Intelligent Network Development Institute will start the development of intelligent driving platform based on huashan-2 a1000 to meet the demand of subsequent mass proction models. The two sides will jointly promote the application of artificial intelligence technology in the automotive instry and accelerate the instrialization of domestic intelligent driving chips

in addition, heima has signed a number of fad fixed-point models, and it is expected that there will be models equipped with fad automatic driving platform on the market next year. In addition, many enterprises at home and abroad have begun to test fad automatic driving platform, and the test vehicles have been on the road

the great success of black sesame in automatic driving chip and domain controller makes instry research institutions begin to pay attention to this start-up company which has just been established for more than four years. In April this year, cbinsights, one of the strongest think tanks in Silicon Valley, released the list of Chinese chip design enterprises. Black sesame was listed in the field of vehicle chip, becoming one of the top 65 Chinese chip design enterprises

in July this year, black sesame Huashan No.2 a1000 chip also appeared in the world artificial intelligence conference, together with pingtouge, Yitu, Cambrian and other high-end artificial intelligence chips

it can be said that after more than four years of development, black sesame has become the world's leading autopilot chip design company, and even has the ability to compete with the eldest brother in the chip instry. At the same time, the rapid progress of black sesame also promotes the domestic automatic driving chip design to a new level

in the conversation with the two founders, they also revealed a colorful egg. Next year, hei mahjong will release a chip with stronger performance. At that time, the highest computing power of fad automatic driving platform equipped with this chip is expected to exceed 1000tops, and its computing power can be fully automatic driving

this article comes from the author of car home, which does not represent the standpoint of car home

in the proct library of pailitao, 1 billion new proct pictures are added every day. In order to accurately identify these huge amounts of images and give users correct feedback, Alibaba needs GPU computing power to identify them for one hour. Cost aside, this speed seems to be a bit slow for Ali to pursue efficiency