1. A customizable embedded AI model floor hardware architecture, comprising:

a customizable power supply module, a customizable detector module, a customizable AI processor module, a customizable debugging module and a customizable peripheral module,

the customizable power module is used for supplying power to the customizable detector module, the customizable AI processor module, the customizable debugging module and the customizable peripheral module;

the customizable detector module is used for acquiring data;

the customizable AI processor module is used for processing the acquired data and different embedded functions by different AI model algorithms according to the processor customized by the requirements and the scenes;

the customizable AI processor module is used for processing the customized AI signals;

the customizable peripheral module can customize different peripheral functions and external communication according to requirements and scenes.

2. The customizable embedded AI model floor hardware architecture of claim 1,

the customizable power module, the customizable detector module, the customizable AI processor module, the customizable debug module and the customizable peripheral module are all arranged on a PCB (printed Circuit Board).

3. The customizable embedded AI model floor hardware architecture of claim 1,

after the PCB locally completes a real-time embedded function and a real-time AI model algorithm in real time, the PCB is applied to different application scenes through the customizable peripheral module.

4. The customizable embedded AI model floor hardware architecture of claim 1,

the customizable AI processor module is a single processor module.

5. The customizable embedded AI model floor hardware architecture of claim 1,

the customizable AI processor module is implemented in an integrated circuit.

6. The customizable embedded AI model floor hardware architecture of claim 5,

the integrated circuit is a field programmable gate array FPGA or an application specific integrated circuit ASIC.

7. The customizable embedded AI model floor hardware architecture of claim 1,

the customizable debugging module finishes different debugging modes according to the requirements and the customized debugging interface of the scene.

8. The customizable embedded AI model floor hardware architecture of claim 1, wherein the customizable peripheral module is a peripheral interface.

Background

At present, an artificial intelligence algorithm is generally completed by complex and huge hardware equipment consisting of a GPU or an AI chip processor with strong performance, the hardware equipment has high cost and high power consumption and is generally used for accelerating a server, or the hardware equipment is used by a network to complete the artificial intelligence algorithm at a cloud end and then transmit the result back to the local. Later, it has slowly appeared that the artificial intelligence algorithm is completed by local embedded hardware equipment, so that artificial intelligence can be applied to consumption equipment to realize embedded artificial intelligence equipment, such as smart phones, vehicle-mounted cameras and the like.

The embedded artificial intelligence has high requirements on the cost, power consumption and volume of hardware, a real-time artificial intelligence algorithm needs to be realized locally on the aspect of realizing a real-time embedded function, and the embedded hardware equipment is high in flexibility and customized design and needs to meet various rich external interfaces and functions. However, the current hardware architecture based on the GPU and the general AI chip is large in size, high in power consumption and high in cost due to the fact that most of the hardware architecture aims at complex application scenes, and thus the requirements of embedded artificial intelligence are difficult to meet.

Disclosure of Invention

The invention provides a customizable floor hardware architecture of an embedded AI model, which can solve the technical problem that embedded artificial intelligence is difficult to meet.

The technical scheme for solving the technical problems is as follows:

a customizable embedded AI model floor hardware architecture, comprising:

a customizable power supply module, a customizable detector module, a customizable AI processor module, a customizable debugging module and a customizable peripheral module,

the customizable power module is used for supplying power to the customizable detector module, the customizable AI processor module, the customizable debugging module and the customizable peripheral module;

the customizable detector module is used for acquiring data;

the customizable AI processor module is used for processing the acquired data and different embedded functions by different AI model algorithms according to the processor customized by the requirements and the scenes;

the customizable AI processor module is used for processing the customized AI signals;

the customizable peripheral module can customize different peripheral functions and external communication according to requirements and scenes.

In some embodiments, in one customizable embedded AI model floor hardware architecture described above,

the customizable power module, the customizable detector module, the customizable AI processor module, the customizable debug module and the customizable peripheral module are all arranged on a PCB (printed Circuit Board).

In some embodiments, in one customizable embedded AI model floor hardware architecture described above,

after the PCB locally completes a real-time embedded function and a real-time AI model algorithm in real time, the PCB is applied to different application scenes through the customizable peripheral module.

In some embodiments, in one customizable embedded AI model floor hardware architecture described above,

the customizable AI processor module is a single processor module.

In some embodiments, in one customizable embedded AI model floor hardware architecture described above,

the customizable AI processor module is implemented in an integrated circuit.

In some embodiments, in one customizable embedded AI model floor hardware architecture described above,

the integrated circuit is a field programmable gate array FPGA or an application specific integrated circuit ASIC.

In some embodiments, in one customizable embedded AI model floor hardware architecture described above,

the customizable debugging module finishes different debugging modes according to the requirements and the customized debugging interface of the scene.

In some embodiments, in one customizable embedded AI model floor hardware architecture described above,

the customizable peripheral module is a peripheral interface.

The invention has the beneficial effects that: the embodiment of the invention provides a customizable embedded type AI model floor hardware architecture, which comprises: the system comprises a customizable power supply module, a customizable detector module, a customizable AI processor module, a customizable debugging module and a customizable peripheral module, wherein the customizable power supply module is used for supplying power to other modules; the customizable detector module is used for acquiring data; the customizable AI processor module is used for processing the acquired data by different AI model algorithms according to the processor customized by the requirements and the scenes; the customizable AI processor module is used for processing the customized AI signals; the customizable peripheral module can customize different peripheral functions and external communication according to requirements and scenes. According to the method and the system, various modules can meet the requirements of different AI models for landing on the ground and realizing locally and different embedded application scenes through flexible customization, realize complex AI model algorithms and embedded functions, and meet the AI processing and embedded processing capabilities of local real-time performance.

Drawings

Fig. 1 is a first schematic diagram of a customizable embedded AI model floor hardware architecture according to an embodiment of the present invention.

Detailed Description

The principles and features of this invention are described below in conjunction with the following drawings, which are set forth by way of illustration only and are not intended to limit the scope of the invention.

In order that the above objects, features and advantages of the present application can be more clearly understood, the present disclosure will be further described in detail with reference to the accompanying drawings and examples. It is to be understood that the embodiments described are only a few embodiments of the present disclosure, and not all embodiments. The specific embodiments described herein are merely illustrative of the disclosure and are not limiting of the application. All other embodiments that can be derived by one of ordinary skill in the art from the description of the embodiments are intended to be within the scope of the present disclosure.

It is noted that, in this document, relational terms such as "first" and "second," and the like, may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions.

Those skilled in the art will appreciate that although some embodiments described herein include some features included in other embodiments instead of others, combinations of features of different embodiments are meant to be within the scope of the application and form different embodiments.

Fig. 1 is a first schematic diagram of a customizable embedded AI model floor hardware architecture according to an embodiment of the present invention.

The embodiment of the invention provides a customizable embedded type AI model floor hardware architecture, which is combined with a figure 1 and comprises the following steps:

a customizable power supply module 101, a customizable detector module 102, a customizable AI processor module 103, a customizable debug module 104, a customizable peripheral module 105,

the customizable power module 101 is configured to supply power to the customizable detector module 102, the customizable AI processor module 103, the customizable debug module 104, and the customizable peripheral module 105;

the customizable detector module 102 is used to collect data;

the customizable detector module is a detector module, collects and acquires data to be processed, and can be customized according to scheme requirements.

The customizable AI processor module 103 is used for processing the acquired data and different embedded functions by different AI model algorithms according to the requirements and scenes.

The customizable debug module 104 is configured to provide online debugging for the customizable AI processor module;

the customizable peripheral module 105 can customize different peripheral functions to communicate with the outside according to the needs and scenarios.

Specifically, the customizable power module in the embodiment of the application provides stable power supply for the whole architecture, and can be customized according to a scheme scene; the customizable detector module is a detector module, acquires data to be processed, and can be customized according to the requirements of a scheme; the customizable AI processor module is a core module of the whole framework, and can complete the core algorithm and the real-time embedded function of the AI for the acquired data in real time, and different processors can be selected and customized according to the requirements of the scheme; the customizable debugging module provides an interface for debugging Debug communication for the system, and different debugging interfaces can be customized according to the requirement of a scheme; the peripheral function modules which can be customized are mainly some peripheral function modules with embedded function requirements and can also be customized according to scheme requirements. Each module can meet the requirements of different AI models for landing on the ground and realizing locally and different embedded application scenes through flexible customization, realize complex AI model algorithms, realize complex embedded functions and meet the AI processing and embedded processing capabilities of local real-time.

In some embodiments, in one customizable embedded AI model floor hardware architecture described above,

the customizable power module 101, the customizable detector module 102, the customizable AI processor module 103, the customizable debug module 104, and the customizable peripheral module 105 are all disposed on a PCB board.

Specifically, in the embodiment of the present application, 5 modules, namely, a customizable power module 101, a customizable detector module 102, a customizable AI processor module 103, a customizable debug module 104, and a customizable peripheral module 105, are integrated on a single hardware PCB, so as to implement small volume, low power consumption, and integrated embedded processing and local AI processing capabilities.

In some embodiments, in one customizable embedded AI model floor hardware architecture described above,

after the PCB locally completes the real-time embedded function and the real-time AI model algorithm in real time, the PCB is applied to different application scenarios through the customizable peripheral module 105.

In some embodiments, in one customizable embedded AI model floor hardware architecture described above,

the customizable AI processor module 103 is a single processor module.

Specifically, in the embodiment of the application, a single processor module not only realizes local real-time embedded processing capability, but also realizes local real-time artificial intelligence processing capability.

In some embodiments, in one customizable embedded AI model floor hardware architecture described above,

the customizable AI processor module 103 is implemented as an integrated circuit.

Specifically, the embodiment of the application completes the artificial intelligence algorithm locally in real time through the integrated circuit, and uses the artificial intelligence algorithm result in local embedded control processing and the like.

In some embodiments, in the customizable embedded AI model floor hardware architecture, the integrated circuit is a field programmable gate array FPGA or an application specific integrated circuit ASIC.

In some embodiments, in one customizable embedded AI model floor hardware architecture described above,

the customizable debug module 104 completes different debug modes according to the requirements and the customized scene debug interface.

In some embodiments, in one customizable embedded AI model floor hardware architecture described above,

the customizable peripheral module 105 is a peripheral interface.

The invention is described below with reference to two specific embodiments.

The first embodiment: the customized power supply module supplies power for a customized AI processor module, a customized detector module, customized debugging module debugging and customized peripheral equipment, and specifically can provide a power supply input source by a portable power supply, such as a lithium battery, a mobile power supply and the like; the customized detector module is an MIPI camera module and is connected with a camera to acquire image data; the customized AI processor module realizes the functions of image classification AI algorithm and image transmission by a programmable FPGA device ZYNQ, and simultaneously ZYNQ completes the control function after classification results, such as sorting of abnormal samples for industrial detection; the customized debugging module is a JTAG module and provides an online debugging function for the FPGA; the customized peripheral interfaces are HDMI, UART, GPIO and the like, and transmit images and results to the display.

The customizable embedded AI model floor hardware architecture provided in the first embodiment realizes the function of image classification on the PCB board and the function of real-time video transmission by customizing each module, has a small volume, a powerful function and a low cost, and can be applied to various embedded application fields.

Second embodiment: the customized power supply module supplies power for a customized AI processor module, a customized detector module, customized debugging module debugging and customized peripheral equipment, and specifically can provide a power supply input source by a portable power supply, such as a lithium battery, a mobile power supply and the like; the customized detector module is an MIPI camera module and is connected with a camera to acquire image data; the customized AI processor module realizes the functions of target detection AI algorithm and image transmission by a programmable FPGA device ZYNQ, and simultaneously transplants a Linux system on the ZYNQ to realize the abnormal monitoring and alarming of interaction and security protection of a complex GUI interface; the customized debugging module is a network interface module, provides an online debugging function for the FPGA, and can update configuration data and the like through a network; the customized peripheral interface is a network interface and transmits the image and the result to the display through a network; the debugging function and the peripheral interface can be simultaneously completed through a network interface.

The customizable embedded AI model floor hardware architecture provided in the second embodiment implements a target detection function on a PCB board and a real-time video transmission function by customizing each module, has a small volume, a powerful function, and a low cost, and can be applied to various embedded application fields.

Those skilled in the art will appreciate that the description of each embodiment has a respective emphasis, and reference may be made to the related description of other embodiments for those parts of an embodiment that are not described in detail.

Although the embodiments of the present application have been described in conjunction with the accompanying drawings, those skilled in the art will be able to make various modifications and variations without departing from the spirit and scope of the application, and such modifications and variations are included in the specific embodiments of the present invention as defined in the appended claims, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of various equivalent modifications and substitutions within the technical scope of the present disclosure, and these modifications and substitutions are intended to be included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

While the invention has been described with reference to specific embodiments, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.