Use self-developed algorithms and tool chains to accelerate chip mass production and application.
"Automobiles are actually the vertical industry with the fastest implementation of intelligent (concept). It reflects the essence that the public's demand for automobiles has undergone qualitative changes."
At the end of last month, at the Black Sesame Smart Shanghai Media Technology Open Day exchange event, Ding Ding, Black Sesame’s vice president of intelligent products, warmed up the event with these honest and somewhat worrying words.
As the 2021 Shanghai Auto Show shouts the slogan "Embrace Change", the automotive industry is undergoing major changes unseen in a century and is sweeping the world. Riding the tide of new energy, automobiles, a highly symbolic industrial product, are accelerating their transformation towards intelligence and connectivity.
As 2022 is coming to an end and 2023 is about to begin, providing autonomous driving functions and equipped with smart cockpits are becoming the basic configuration of various new models at this stage.
Whether it is multi-screen linkage + voice control to realize cockpit intelligence, or providing automatic/intelligent driving functions to drivers, high-performance cockpit control chips, and high-computing power automatic driving chips are all prerequisites.
There is no doubt that chips, especially automotive-grade chips, are the core keywords for China's industry in 2022. Various types of FPGAs, DSPs, MCUs, SiC power devices, etc. are used to implement various basic functions on automotive-grade chips, and CPUs and GPUs (AI) used to provide main control computing functions are becoming the core of them. Competitiveness.
As one of the recognized "unicorns" in the domestic intelligent driving industry, since its establishment in 2016, Black Sesame Intelligence has been targeting the development of large computing power intelligent driving chips and the supporting of related autonomous driving technology solutions.
The Huashan No. 2 A1000 Pro chip, which is currently independently developed and has an int 4 computing power of 196 tops, was tape-out at the end of July last year.
01. Large computing power chips are becoming a necessity
When smart driving gradually moves from cutting-edge technology to the stage of commercial application, the implementation of smart driving functions and smart cockpits means that automobile intelligence has achieved a qualitative change from scratch.
However, although going from zero to one is earth-shaking, the further development from the beginning is equally significant. What is currently certain is that this next stage of innovation is centered around chips, artificial intelligence, electronic architecture and other aspects.
Its fundamental pursuit is to achieve a leap forward in digital applications in cars and in-car digital experience needs.
The foundation of all this development is inseparable from chips. In the past five years, the computing power of smart cars has gradually increased from a few tops to thousands of tops.
Although there is still controversy in the industry about the development of intelligent driving technology around whether the implementation of high-precision digital maps is necessary, whether it is re-sensing or relocating routes, in the long run, the further improvement of computing power requirements will be A necessity.
Especially looking at the world, when the leading company NVIDIA launched Atlan, a new generation of autonomous driving chip with a computing power of 1000 tops, in the spring of 2021, it has become an industry-recognized way to promote the development of autonomous driving through the improvement of computing power.
The development of automobile intelligence requires self-driving chips with high performance and large computing power to provide a stage. Autonomous driving has entered an era of competition for computing power, and computing power has become the core driving force for the development of autonomous driving technology.
In addition to the Huashan No. 2 A1000 Pro mentioned above, the two mature products currently promoted by Black Sesame are the A1000 and A1000 L, both of which belong to the Huashan No. 2 series. These two chips are also based on the 16nm Auto process and are mainly aimed at intelligent driving domain main control chips under the current domain control architecture.
A1000 is mainly used to realize L2 intelligent driving functions. It is a standard AI chip used to realize intelligent driving functions.
As for the A1000 L, the L suffix can actually be understood as Light, which is relatively lower-level than the former. It is mainly used to realize the surround view parking and 5V5R functions, which is equivalent to integrating the active safety L2 camera function and the surround view parking low-speed function to achieve the so-called "walking and parking in one" function. It is considered an entry-level product.
Of course, it can also improve the "last 100 meters" problem for car owners while driving, and while improving the user experience, it also provides a very good sense of technology.
During the media exchange stage, Ding Ding focused on introducing the technology development plan of Black Sesame Intelligence——
The next generation of products still under development is based on higher-end "central" computing chips, while the current Huashan No. 2 A1000 series is more of an era of domain-oriented architecture. The next generation of chips will evolve from domain architecture to central architecture and make certain contributions to promoting electronic and electrical architecture.
“To put it simply, for the next generation of chips, we will consider more advanced processes and how to further concentrate the functions in different domains just mentioned into chips with greater computing power.
At the same time, its self-driving computing capabilities will be improved by about ten times from the previous generation of chips to the next generation of large-scale computing chips. At the same time, we will greatly enhance functional safety and essential attributes of cars. "
02. Our positioning is to be a good Tier 2
Regarding the industry positioning issue of Black Sesame Intelligence that has been speculated by the outside world before, Black Sesame Intelligence made a clear statement in its official capacity through the platform of this Open Day: the positioning of our company is to play the role of Tier 2 in the standard industry. Through cooperation with Tier1, we will eventually provide our products to OEMs.
In short, we must not follow Mobileye's original approach and provide enterprises with a "black box" that has completed all development and integration processes, and only requires the OEM to complete the vehicle adaptation work.
"Of course we have also developed a lot of software and a lot of supporting tool chains, but our starting point is more that when customers get such complex chips, they can use our chips faster and better, so we don't It’s not that we invest a lot of manpower in making software and tools and set relatively high threshold fees.”
Ding Ding explained the software and tool chain issues for the supporting development of Black Sesame Intelligence: "We hope to rely on the capabilities of the chip itself to allow customers to quickly apply it. When the scale is established, we will return to the business logic of semiconductors."
Regarding the A1000 and A1000 L, which are now very mature and reliable, Ding Ding further added that there are already many Tier1 suppliers and partners in China, including car manufacturers, who are also adopting them.
In particular, the advantage of the A1000 L is that it can provide higher computing power while the cost of the current car factory's parking system remains basically unchanged. Just a small change to the original system can bring about a leap in parking experience.
So in the subsequent media exchange, someone asked about Black Sesame Intelligence’s current partners. In this regard, although it is a matter of trade secrets, Yang Yuxin, chief marketing officer of Black Sesame Intelligence, has revealed some information to everyone——
Black Sesame Intelligence has currently appointed more than 10 car companies. 80% of them are passenger cars and 20% are commercial vehicles.
"The samples currently on display in our company are all products that customers are developing and will actually be mass-produced. In terms of progress, we think they are in line with our expectations. Our resources cannot be spread too thin, so we We will focus on current car companies and implement our mass production cooperation first.”
According to the public news, at the end of May this year, Black Sesame Intelligence and JAC Automobile reached a cooperation. Black Sesame's Huashan No. 2 A1000 series autonomous driving chips will be installed on JAC's Sihao brand models, providing Provide L2 level intelligent driving system and provide core computing power support.
From a technical perspective, a very important driving force for the popularization of autonomous driving is the evolution of automotive electronic and electrical architecture, and there are two main technical points to promote the evolution: data and computing power.
Data is the "blood" of the next generation of smart cars. The evolution direction of electronic and electrical architecture is to ensure the high-speed flow of very large amounts of data to further support the functions deployed on it. This involves data processing, which requires stronger computing power. Chips to support the evolution of electronic and electrical architecture.
"For the next generation of chips, Black Sesame Intelligence will consider more advanced processes and how to further concentrate functions in different domains into chips with greater computing power. At the same time, in terms of autonomous driving computing capabilities, the next generation of chips will It will be more than ten times more powerful in terms of computing power. At the same time, the functional safety and essential attributes of the car will be greatly enhanced," Ding Ding finally added.
Previous article:Storage drops, but car memory doesn't
Next article:For blind-filling lidar, is the smaller the angular resolution really, the better?
- Popular Resources
- Popular amplifiers
- A new chapter in Great Wall Motors R&D: solid-state battery technology leads the future
- Naxin Micro provides full-scenario GaN driver IC solutions
- Interpreting Huawei’s new solid-state battery patent, will it challenge CATL in 2030?
- Are pure electric/plug-in hybrid vehicles going crazy? A Chinese company has launched the world's first -40℃ dischargeable hybrid battery that is not afraid of cold
- How much do you know about intelligent driving domain control: low-end and mid-end models are accelerating their introduction, with integrated driving and parking solutions accounting for the majority
- Foresight Launches Six Advanced Stereo Sensor Suite to Revolutionize Industrial and Automotive 3D Perception
- OPTIMA launches new ORANGETOP QH6 lithium battery to adapt to extreme temperature conditions
- Allegro MicroSystems Introduces Advanced Magnetic and Inductive Position Sensing Solutions
- TDK launches second generation 6-axis IMU for automotive safety applications
- LED chemical incompatibility test to see which chemicals LEDs can be used with
- Application of ARM9 hardware coprocessor on WinCE embedded motherboard
- What are the key points for selecting rotor flowmeter?
- LM317 high power charger circuit
- A brief analysis of Embest's application and development of embedded medical devices
- Single-phase RC protection circuit
- stm32 PVD programmable voltage monitor
- Introduction and measurement of edge trigger and level trigger of 51 single chip microcomputer
- Improved design of Linux system software shell protection technology
- What to do if the ABB robot protection device stops
- Huawei's Strategic Department Director Gai Gang: The cumulative installed base of open source Euler operating system exceeds 10 million sets
- Download from the Internet--ARM Getting Started Notes
- Learn ARM development(22)
- Learn ARM development(21)
- Learn ARM development(20)
- Learn ARM development(19)
- Learn ARM development(14)
- Learn ARM development(15)
- Analysis of the application of several common contact parts in high-voltage connectors of new energy vehicles
- Wiring harness durability test and contact voltage drop test method
- Ltspice .save
- SinlinxA33 development board uses MIPI DSI screen
- Install Runhe Neptune Hongmeng system on w801kit
- DSPC6678 on-chip storage space allocation mechanism
- Ask a silly question about f_mount()!
- [Evaluation of SGP40] + Cloud platform configuration for IoT remote monitoring application development
- [RVB2601 creative application development] + AHT10 temperature and humidity sensor
- A nice frame diagram
- Problem with the start address of SDRAM W9825G6KH
- How to analyze this circuit