CN113609701A - Design method for architecture of intelligent driving domain controller with scalable AI chip number - Google Patents
Design method for architecture of intelligent driving domain controller with scalable AI chip number Download PDFInfo
- Publication number
- CN113609701A CN113609701A CN202110950618.4A CN202110950618A CN113609701A CN 113609701 A CN113609701 A CN 113609701A CN 202110950618 A CN202110950618 A CN 202110950618A CN 113609701 A CN113609701 A CN 113609701A
- Authority
- CN
- China
- Prior art keywords
- intelligent driving
- chips
- domain controller
- development
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000013461 design Methods 0.000 title claims abstract description 17
- 238000000034 method Methods 0.000 title claims abstract description 14
- 230000006870 function Effects 0.000 claims abstract description 41
- 238000011161 development Methods 0.000 claims abstract description 35
- 230000007246 mechanism Effects 0.000 claims abstract description 10
- 238000012545 processing Methods 0.000 claims description 6
- 238000003860 storage Methods 0.000 claims description 4
- 238000004519 manufacturing process Methods 0.000 abstract description 4
- 238000009826 distribution Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 102100034112 Alkyldihydroxyacetonephosphate synthase, peroxisomal Human genes 0.000 description 1
- 101000799143 Homo sapiens Alkyldihydroxyacetonephosphate synthase, peroxisomal Proteins 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 238000000848 angular dependent Auger electron spectroscopy Methods 0.000 description 1
- 238000005538 encapsulation Methods 0.000 description 1
- 230000014509 gene expression Effects 0.000 description 1
- 230000017525 heat dissipation Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 229920002397 thermoplastic olefin Polymers 0.000 description 1
- 238000003466 welding Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/15—Vehicle, aircraft or watercraft design
Abstract
The invention discloses a design method of an intelligent driving domain controller framework with a tailorable AI chip quantity, which has the design concept that in the initial development stage, the AI chips in the intelligent driving domain controller are deployed according to the configuration of the highest vehicle model, then the quantity and the model of the AI chips actually participating in the development are distributed and determined according to different intelligent driving functions or calculation power required by the actual vehicle model configuration, a certain quantity and model of the AI chips are tailored from the initially deployed AI chips according to the quantity and the model, and a corresponding MCU (microprogrammed control unit) scheduling mechanism, a driving control system, an operating system and program software are designed, so that a compatible development scheme with the capability of synchronously tailoring software and hardware and aiming at different vehicle model configurations under the same hardware platform is provided for the subsequent mass production stage. The invention takes the highest configuration of the intelligent driving domain controller as a development starting point, and realizes the tailorable development according to the actually required intelligent driving function or corresponding calculation force under the same platform on the premise of keeping the rest electronic components unchanged.
Description
Technical Field
The invention relates to the technical field of intelligent driving, in particular to a design method of an intelligent driving domain controller architecture with an AI chip with a tailorable quantity.
Background
The realization of current intelligence driving function needs the hardware computing platform based on high performance, and this computing platform generally is called intelligent driving domain controller, and intelligent driving domain controller is similar to the host computer of desktop computer, and hardware such as including CPU, GPU, RAM, ROM, power module, heat dissipation module, peripheral interface module, mainboard etc. above-mentioned hardware generally passes through welding encapsulation on a mainboard, makes a whole. In the process of designing the domain controller hardware architecture, firstly, different types of chips and the number of the chips are selected based on functions to be developed, then, power supply, internal memory and the like on the periphery of the chips are developed and designed, and finally, the chips are packaged into a whole to complete the design and manufacture of the whole domain controller hardware architecture.
In the practical application process, even the same type of vehicle in the same period is often configured with different intelligent driving functions, and when the configured functions are more, the required calculation force and the types and the number of corresponding chips are increased along with the increase of the intelligent degree.
In order to solve different intelligent configurations of the same vehicle type, intelligent driving area controllers with different hardware configurations are mainly designed at present. For example, in the current vehicle models sold on the market, the vehicle model only matched with the AEB function and the vehicle with the high-version matched automatic parking and advanced ADAS functions adopt different hardware of the intelligent driving area controller. Although the intelligent driving area controller hardware can be provided according to different configured vehicle types as required, in the technical development stage, corresponding software development and a corresponding management and maintenance system need to be designed for each hardware platform independently, so that the quantity and the scale of internal development platforms are large, and the development cost and the management cost are increased.
In order to solve different intelligent configurations of the same vehicle type, another scheme is adopted in the prior art, namely the same set of intelligent driving domain controller is installed on all vehicle types, different functions are carried by adopting a payment activation mode, namely the hardware configurations of delivered new vehicles are all the same, and a user can open the corresponding intelligent driving function only by paying additionally. Although the scheme can save the development and maintenance cost of the platform, if a user does not need to activate the intelligent driving function or only needs to activate part of the intelligent driving function, unnecessary vehicle purchasing cost cannot be spent on the whole set of high-computing-power intelligent driving domain controller hardware with built-in full functions.
Disclosure of Invention
In view of the above, the present invention is directed to a design method of an intelligent driving domain controller architecture with scalable AI chips, so as to overcome the above disadvantages in the conventional development of an intelligent driving domain controller.
The technical scheme adopted by the invention is as follows:
a design method for an intelligent driving domain controller architecture with a tailorable AI chip number comprises the following steps:
carrying a plurality of AI chips for processing intelligent driving data for a main board of an intelligent driving domain controller according to the intelligent driving requirement of a top vehicle model in the initial development stage; each AI chip is preset with a corresponding intelligent driving function or configured with a set calculation value;
determining the number and the type of actually required AI chips from the initial AI chips deployed according to top vehicle models based on intelligent driving functions or computational power requirements corresponding to different vehicle model configurations;
according to the determined quantity and model of the AI chips, the hardware drive and operation system is subjected to targeted development and is preset in a storage module of the intelligent driving domain controller, and meanwhile, a scheduling mechanism of the AI chips and an intelligent driving software program are subjected to targeted development;
after the targeted development is completed, the intelligent driving area controller with the corresponding quantity and model of the AI chips is installed in the corresponding configured vehicle type.
In at least one possible implementation manner, the AI chips in the intelligent driving area controller operate independently of each other, and the MCU of the intelligent driving area controller executes the scheduling mechanism to realize cooperation between the AI chips.
In at least one possible implementation manner, the AI chip includes a CPU or a GPU.
The design concept of the invention is that in the initial stage of development, the AI chips in the intelligent driving domain controller are deployed according to the highest vehicle type configuration, then the number and the types of the AI chips actually participating in development are distributed and determined according to different intelligent driving functions or calculation forces required by the actual vehicle type configuration, a certain number and types of AI chips are cut out from the initially deployed AI chips, and a corresponding MCU scheduling mechanism, drive control, an operating system and program software are designed, so that a compatible development scheme which can be synchronously cut by software and hardware and is configured for different vehicle types on the same hardware platform is provided for the subsequent mass production stage. The invention takes the highest configuration of the intelligent driving domain controller as a development starting point, and realizes the tailorable development according to the actually required intelligent driving function or corresponding calculation force under the same platform on the premise of keeping the rest electronic components unchanged.
Drawings
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be further described with reference to the accompanying drawings, in which:
fig. 1 is a flowchart of a design method of an intelligent driving domain controller architecture with scalable AI chip number according to an embodiment of the present invention;
fig. 2 is a schematic diagram (one) of a top vehicle type AI chip deployment scheme provided in an embodiment of the present invention;
fig. 3 is a schematic diagram (ii) of a top vehicle type AI chip deployment scheme provided in the embodiment of the present invention.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are illustrative only and should not be construed as limiting the invention.
The invention provides an embodiment of an architecture design method of an intelligent driving domain controller with a tailorable AI chip number, and specifically, as shown in FIG. 1, the method may include:
s1, carrying a plurality of AI chips for intelligent driving data processing for the main board of the intelligent driving domain controller according to the intelligent driving requirement of the top vehicle model in the initial development stage; each AI chip is preset with a corresponding intelligent driving function or configured with a set calculation value;
step S2, determining the number and model of AI chips needed actually from the AI chips initially deployed according to top-matched vehicle types based on intelligent driving functions or computing power requirements corresponding to different vehicle type configurations;
step S3, according to the determined quantity and model of AI chips, the hardware drive and operation system is pertinently developed and preset in the storage module of the intelligent driving domain controller, and meanwhile, the dispatching mechanism of the AI chips and the intelligent driving software program are pertinently and synchronously developed;
and step S4, after the targeted development is completed, installing the intelligent driving area controller with the corresponding AI chip quantity and model in the corresponding configured vehicle type.
Furthermore, the AI chips in the intelligent driving area controller run independently, and the MCU of the intelligent driving area controller executes the scheduling mechanism to realize the cooperation among the AI chips.
Further, the AI chip includes a CPU or a GPU.
To facilitate an understanding of the above embodiments and their preferred versions, the following schematic illustrations are provided:
in the initial stage of development, according to the intelligent driving requirement of a top-loading vehicle model, a plurality of AI chips (initial configuration can be realized by adopting an entity or virtual mode) are firstly carried on a main board of an intelligent driving domain controller, wherein the AI chips can be a CPU (central processing unit), a GPU (graphics processing unit) or other types of chips and are mainly used for data processing and calculation of an intelligent driving system. The AI chips operate independently, and each AI chip is provided with a plurality of different intelligent driving functions or configured with a set calculation value.
And determining the number and the type of actually required AI chips based on the intelligent driving function schemes configured by different vehicle types. Specifically, as shown in fig. 2, three AI chips are initially loaded on the domain controller motherboard and are respectively responsible for different intelligent driving functions. When the top vehicle allocation type scheme needs the first intelligent driving function to the sixth intelligent driving function, determining that all three AI chips are needed to participate in subsequent development; for the medium-sized vehicle model, the vehicle model can comprise a first intelligent driving function to a fourth intelligent driving function, and the first two AI chips are determined to be needed to participate in subsequent development; and for the low-mix vehicle type, the first intelligent driving function to the second intelligent driving function can be included, and then the first AI chip is determined to be needed to participate in the subsequent development. Alternatively, as shown in fig. 3, three AI chips are initially mounted on the domain controller main board, and each AI chip has a computational power of 30 TPOs. When the intelligent driving function corresponding to the design scheme of the top-loading vehicle type needs 90TPOs calculation force, all three AI chips are determined to be equipped; for the intelligent driving function of the medium-vehicle-distribution type, when 60TPOs calculation force is needed, two AI chips are determined to be equipped; the intelligent driving function of the low-vehicle-distribution type is simple, for example, only 30TPOs calculation force is needed, and an AI chip can be determined to be needed.
According to the determined number and specific model of the required AI chips, starting to purposefully develop a hardware drive and operating system and deploy the hardware drive and operating system in a storage module of the intelligent driving area controller; meanwhile, an AI chip scheduling mechanism of the MCU and an intelligent driving software program need to be developed in a targeted manner.
After the targeted development is completed, the intelligent driving area controllers with the corresponding AI chip quantity and model are installed in the corresponding configured vehicle types. For example, a highly-distributed vehicle is provided with an intelligent driving area controller with three AI chips, which has 6 functions or 90TPOs calculation power; the middle-distribution vehicle model is provided with an intelligent driving area controller with two AI chips, which has 4 functions or has 60TPOs calculation power; the low-vehicle-distribution type can be provided with an intelligent driving area controller with an AI chip, and has 2 functions or 30TPOs calculation power. Therefore, the purpose of collocating different AI chips based on the required functions under the same hardware platform is realized, and the number, the built-in functions and the calculation power value of the AI chips are all referred to, which is not limited by the invention.
In summary, the design concept of the present invention is that, in an initial development stage, AI chips in an intelligent driving domain controller are deployed according to the highest vehicle model configuration, and then the number and the model of the AI chips actually participating in the development are allocated and determined according to different intelligent driving functions or computing power required by the actual vehicle model configuration, and accordingly, a certain number and model of AI chips are cut out from the initially deployed AI chips, and a corresponding MCU scheduling mechanism, driving control, operating system and program software are designed, thereby providing a compatible development scheme in which software and hardware can be synchronously cut out and configured for different vehicle models on the same hardware platform in a subsequent mass production stage. The invention takes the highest configuration of the intelligent driving domain controller as a development starting point, and realizes the tailorable development according to the actually required intelligent driving function or corresponding calculation force under the same platform on the premise of keeping the rest electronic components unchanged.
In the embodiments of the present invention, "at least one" means one or more, "a plurality" means two or more. "and/or" describes the association relationship of the associated objects, and means that there may be three relationships, for example, a and/or B, and may mean that a exists alone, a and B exist simultaneously, and B exists alone. Wherein A and B can be singular or plural. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. "at least one of the following" and similar expressions refer to any combination of these items, including any combination of singular or plural items. For example, at least one of a, b, and c may represent: a, b, c, a and b, a and c, b and c or a and b and c, wherein a, b and c can be single or multiple.
The structure, features and effects of the present invention have been described in detail with reference to the embodiments shown in the drawings, but the above embodiments are merely preferred embodiments of the present invention, and it should be understood that technical features related to the above embodiments and preferred modes thereof can be reasonably combined and configured into various equivalent schemes by those skilled in the art without departing from and changing the design idea and technical effects of the present invention; therefore, the invention is not limited to the embodiments shown in the drawings, and all the modifications and equivalent embodiments that can be made according to the idea of the invention are within the scope of the invention as long as they are not beyond the spirit of the description and the drawings.
Claims (3)
1. A design method of an intelligent driving domain controller architecture with the number of AI chips capable of being tailored is characterized by comprising the following steps:
carrying a plurality of AI chips for processing intelligent driving data for a main board of an intelligent driving domain controller according to the intelligent driving requirement of a top vehicle model in the initial development stage; each AI chip is preset with a corresponding intelligent driving function or configured with a set calculation value;
determining the number and the type of actually required AI chips from the initial AI chips deployed according to top vehicle models based on intelligent driving functions or computational power requirements corresponding to different vehicle model configurations;
according to the determined quantity and model of the AI chips, the hardware drive and operation system is subjected to targeted development and is preset in a storage module of the intelligent driving domain controller, and meanwhile, a scheduling mechanism of the AI chips and an intelligent driving software program are subjected to targeted development;
after the targeted development is completed, the intelligent driving area controller with the corresponding quantity and model of the AI chips is installed in the corresponding configured vehicle type.
2. The architecture design method for an intelligent driving domain controller with tailorable AI chips according to claim 1, wherein the AI chips in the intelligent driving domain controller operate independently of each other, and the scheduling mechanism is executed by the MCU of the intelligent driving domain controller to realize the cooperation between the AI chips.
3. The AI chip quantity-tailorable intelligent driving domain controller architecture design method of claim 1 or 2, wherein the AI chip comprises a CPU or GPU.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110950618.4A CN113609701A (en) | 2021-08-18 | 2021-08-18 | Design method for architecture of intelligent driving domain controller with scalable AI chip number |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110950618.4A CN113609701A (en) | 2021-08-18 | 2021-08-18 | Design method for architecture of intelligent driving domain controller with scalable AI chip number |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113609701A true CN113609701A (en) | 2021-11-05 |
Family
ID=78341103
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110950618.4A Pending CN113609701A (en) | 2021-08-18 | 2021-08-18 | Design method for architecture of intelligent driving domain controller with scalable AI chip number |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113609701A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006143016A (en) * | 2004-11-19 | 2006-06-08 | Denso Corp | Construction method of control system for vehicle and control system for vehicle |
| CN108388532A (en) * | 2018-03-13 | 2018-08-10 | 算丰科技（北京）有限公司 | The AI operations that configurable hardware calculates power accelerate board and its processing method, server |
| CN108556767A (en) * | 2018-03-01 | 2018-09-21 | 李洪运 | A kind of expansible intelligent driving auxiliary system |
| CN109855646A (en) * | 2019-04-30 | 2019-06-07 | 奥特酷智能科技(南京)有限公司 | It is distributed centralized automated driving system and method |
-
2021
- 2021-08-18 CN CN202110950618.4A patent/CN113609701A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006143016A (en) * | 2004-11-19 | 2006-06-08 | Denso Corp | Construction method of control system for vehicle and control system for vehicle |
| CN108556767A (en) * | 2018-03-01 | 2018-09-21 | 李洪运 | A kind of expansible intelligent driving auxiliary system |
| CN108388532A (en) * | 2018-03-13 | 2018-08-10 | 算丰科技（北京）有限公司 | The AI operations that configurable hardware calculates power accelerate board and its processing method, server |
| CN109855646A (en) * | 2019-04-30 | 2019-06-07 | 奥特酷智能科技(南京)有限公司 | It is distributed centralized automated driving system and method |
| WO2020207504A1 (en) * | 2019-04-30 | 2020-10-15 | 奥特酷智能科技(南京)有限公司 | Distributed centralized automatic driving system |
Non-Patent Citations (3)
| Title |
|---|
| AI汽车网: "全球自动驾驶AI芯片最新、最全盘点", Retrieved from the Internet <URL:https://auto.vogel.com.cn/c/2020-07-12/1047972.shtml> * |
| 望天涯: "高通自动驾驶平台 - Snapdragon Ride", pages 1 - 2, Retrieved from the Internet <URL:https://zhuanlan.zhihu.com/p/103310894> * |
| 杨东: "智能车辆自动驾驶域控制器设计与实现", 工程科技Ⅱ辑, no. 2, pages 035 - 490 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7971203B2 (en) | Method, apparatus and system for dynamically reassigning a physical device from one virtual machine to another | |
| US7483974B2 (en) | Virtual management controller to coordinate processing blade management in a blade server environment | |
| US5796941A (en) | Method for supervising software execution in a license restricted environment | |
| CN101344899B (en) | Simulation test method and system of on-chip system | |
| CN102279766B (en) | Method and system for concurrently simulating processors and scheduler | |
| CN104778083A (en) | Method and device for task scheduling on heterogeneous multi-core reconfigurable computing platform | |
| US11003429B1 (en) | Compile-time scheduling | |
| Xie et al. | Minimizing development cost with reliability goal for automotive functional safety during design phase | |
| US20090049214A1 (en) | Graphics card test method | |
| CN111176739A (en) | System starting method, device, equipment and storage medium | |
| CN104715503A (en) | Primitive Processing in Graphics Processing System | |
| CN110750312A (en) | Hardware resource configuration method and device, cloud side equipment and storage medium | |
| US20210158131A1 (en) | Hierarchical partitioning of operators | |
| CN107992411A (en) | A kind of method for building task management system software virtual testing environment | |
| CN110750359B (en) | Hardware resource configuration method and device, cloud side equipment and storage medium | |
| CN101339516A (en) | Efficient patching using a data structure | |
| EP2325747A2 (en) | Virtual platform for prototyping system-on-chip designs | |
| KR20110100070A (en) | Method and apparatus for simulation of multi core system using reconfigurable processor core | |
| CN114637598A (en) | Vehicle controller and scheduling method of operating system thereof | |
| CN102331961B (en) | Method, system and dispatcher for simulating multiple processors in parallel | |
| CN111353263A (en) | Software and hardware design and verification platform system | |
| CN113609701A (en) | Design method for architecture of intelligent driving domain controller with scalable AI chip number | |
| CN113253978A (en) | Embedded software development method, model, electronic device, and medium | |
| CN112667373A (en) | Task scheduling method, device and system based on heterogeneous CPU architecture and storage medium | |
| CN111694787A (en) | Chip starting method, network equipment and machine readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |