Internal control method and device based on embedded AI edge computing system

Abstract

The invention discloses an internal control method and device based on an embedded AI edge computing system, and belongs to the technical field of networks. According to the method, efficient cooperative work of hardware modules is ensured by accurately controlling each step. Firstly, the system receives an input signal from external equipment through multiple channels, and performs denoising, format conversion and real-time analysis on the signal to ensure the accuracy and timeliness of the signal; in the system starting process, the connection state of the hardware module is automatically detected, and it is ensured that all assemblies can work normally. By monitoring the state of the system in real time, potential faults can be found and repaired in time. According to the method, through intelligent control and an optimized assembly process, the stability and reliability of the system are greatly improved, the failure rate is reduced, correct connection of each hardware component is ensured in the assembly process, and assembly errors are avoided through electrical detection and functional verification. Therefore, stable operation of the system under high load is ensured, and the method is particularly suitable for an intelligent edge computing system.

Description

Technical Field

[0001] Embodiments of the present application relate to the field of network technologies, and particularly to an internal control method and device based on an embedded AI edge computing system. Background Art

[0002] With the continuous development of artificial intelligence and edge computing technologies, more and more application scenarios require edge devices to have efficient computing capabilities. Especially in the fields of intelligent robots, autonomous driving, and the Internet of Things, embedded AI systems have become key components. These systems not only need to have powerful data processing capabilities but also require efficient and stable hardware assembly methods and internal control systems to ensure the long-term reliable operation of the system.

[0003] However, there are problems in processing signals in the prior art, especially in the collaborative control between hardware modules, system real-time detection, and fault warning. The designs of many existing system control methods in aspects such as data transmission, temperature control, and power management are not perfect, resulting in the device being prone to failure under high-load operation.

[0004] Therefore, there is an urgent need for a new method that can simultaneously meet control and assembly requirements to ensure the stable operation of the system under high-load and complex environments and improve the reliability and efficiency of the device. Summary of the Invention

[0005] Embodiments of the present application provide an internal control method and device based on an embedded AI edge computing system. The technical solutions are as follows:

[0006] On the one hand, an internal control method based on an embedded AI edge computing system is provided. The method includes:

[0007] Receiving input signals from external devices, receiving images, sensor data, or network connection signals through sensors or external interfaces;

[0008] Processing the received input signals and controlling the system to perform signal processing operations;

[0009] When the system starts, automatically detecting and verifying the working status of hardware modules and detecting all hardware modules;

[0010] Real-time monitoring of the system status.

[0011] Optionally, the external devices include cameras, lidars, USB interface devices, and Ethernet modules.

[0012] Optionally, the signal processing operations include signal denoising, calibration, format conversion, and real-time analysis.

[0013] Optionally, the processing of the received input signal and controlling the system to perform signal processing operations includes:

[0014] Signal filtering, removing the noise part in the signal through a filter;

[0015] Data format conversion, converting the received signal into a unified format;

[0016] Real-time analysis, analyzing and modeling the converted data.

[0017] Optionally, the detecting of all hardware modules includes:

[0018] Electrical connection detection, confirming that there is no open circuit or short circuit at all connection points;

[0019] Signal integrity detection, performing integrity detection on the signal transmission of each interface to determine that there is no signal loss or interference;

[0020] Power management detection, used to confirm that the power supply voltage and current of the system are within a safe range.

[0021] Optionally, the real-time monitoring of the system status includes:

[0022] Temperature monitoring, real-time monitoring of the temperature changes of each component of the system through a temperature sensor;

[0023] Voltage monitoring, detecting the voltage fluctuation of the system power supply;

[0024] Signal transmission monitoring, performing periodic checks on all transmitted signals.

[0025] Optionally, the method further includes:

[0026] Connecting the hardware components to the main board in a predetermined order, including connecting the Jetson AGX Orin platform to other interface modules;

[0027] Connecting external devices through a standardized interface;

[0028] Performing electrical tests on all connection points;

[0029] After completion of the assembly, performing system initialization and verifying the correctness of the system through the control module, and checking whether each module during the system startup process works properly.

[0030] Optionally, the connecting of external devices through a standardized interface includes:

[0031] Connecting the power module to ensure stable power input to the system;

[0032] Connecting each interface module, including the GMSL interface, USB interface, and Ethernet module.

[0033] This method ensures the efficient collaborative work of hardware modules by precisely controlling each step. First, the system receives input signals from external devices through multiple channels, and denoises, converts the format, and performs real-time analysis on the signals to ensure the accuracy and timeliness of the signals. During the system startup process, the connection status of hardware modules is automatically detected to ensure that all components can work properly. By monitoring the system status in real time, potential faults can be detected and repaired in a timely manner. This method greatly improves the stability and reliability of the system and reduces the failure rate through intelligent control and optimized assembly processes. The assembly process not only ensures the correct connection of each hardware component, but also avoids assembly errors through electrical detection and functional verification. Thus, it ensures the stable operation of the system under high load and is particularly suitable for intelligent edge computing systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 FIG. shows a schematic flowchart of an internal control method for an embedded AI edge computing system provided by an exemplary embodiment of the present application;

[0035] Figure 2 FIG. shows a system block diagram of an internal control method for an embedded AI edge computing system provided by an exemplary embodiment of the present application;

[0036] Figure 3 FIG. shows a schematic diagram of the access methods of a GMSL camera interface and a USB camera interface provided by an exemplary embodiment of the present application;

[0037] Figure 4 FIG. shows the physical diagrams of a USB expansion card and a GM expansion card provided by another exemplary embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0038] To make the objectives, technical solutions, and advantages of the present application clearer, the following will further describe the embodiments of the present application in detail with reference to the accompanying drawings.

[0039] As used herein, "a plurality of" means two or more. "And / or" describes the association relationship of associated objects and indicates that three relationships may exist. For example, A and / or B may represent: A exists alone, A and B exist simultaneously, and B exists alone. The character " / " generally represents an "or" relationship between the associated objects before and after.

[0040] First, as Figure 1 shown, a schematic flowchart of an internal control method for an embedded AI edge computing system is illustrated.

[0041] As Figure 2As shown, a system block diagram of an internal control method based on an embedded AI edge computing system is illustrated. The architecture of the entire system is shown, including the SOC, external devices such as lidar, cameras, Ethernet interfaces, etc., which are used to describe the working state detection, external device access, and data synchronization of the hardware modules involved in the present invention.

[0042] As Figure 3 shown, the access methods of the GMSL camera interface and the USB camera interface are presented, corresponding to the part of "receiving input signals from external devices", showing how to connect the camera module to the system and obtain data. When inserting, push upward in the direction of the red frame with force. When properly inserted, the camera jack will be stuck. When pulling out, hold the plastic piece inside (marked by the red frame), and do not pull out violently.

[0043] As Figure 4 shown, the USB expansion card and the GM expansion card are presented, which are suitable for supporting the step of "connecting external devices through a standardized interface" in the present invention, specifically illustrating how to use the expansion card to increase the device access capability of the system.

[0044] Embodiment 1

[0045] As Figure 1 shown, the present application provides an internal control method based on an embedded AI edge computing system, and the method includes:

[0046] Step a, receiving input signals from external devices, receiving image, sensor data, or network connection signals through sensors or external interfaces;

[0047] Step b, processing the received input signals and controlling the system to perform corresponding operations, including signal denoising, format conversion, and real-time analysis;

[0048] Step c, automatically detecting and verifying the working state of the hardware modules at system startup, checking the electrical connections, signal integrity, and power management of all hardware modules;

[0049] Step d, monitoring the system state in real time to ensure the coordinated operation between modules, including the monitoring of temperature, voltage, and signal integrity.

[0050] In a possible implementation manner, the system first receives external data through external device interfaces (such as USB, GMSL, Ethernet, etc.). The received signals are preprocessed, including noise removal, format standardization, and real-time data analysis. At the system startup stage, the electrical connection status, signal transmission condition, and power management function of the hardware modules are automatically detected. The system will monitor the hardware working state in real time, such as temperature, voltage, etc., to ensure normal operation.

[0051] Among them, signal preprocessing is completed by an embedded digital signal processing (DSP) unit to ensure the accuracy and real-time performance of data. The system conducts automated hardware detection to verify whether all modules are working properly, reducing the risk of system failures. The monitoring function integrates sensors such as temperature, voltage, and signals to provide real-time feedback on the system status and ensure the coordinated operation of the hardware.

[0052] Thus, the automated module detection and status monitoring ensure the stable operation of the hardware modules, reducing system failures. Through real-time monitoring, the system can respond promptly when abnormalities occur, preventing the spread of failures. Moreover, signal processing and conversion ensure the accuracy of input signals and optimize the system's response speed.

[0053] Embodiment 2

[0054] Optionally, the external devices in step a include cameras, lidars, USB interface devices, and Ethernet modules.

[0055] In a possible implementation, the system supports the access of multiple external devices, including cameras, lidars, USB interfaces, and Ethernet modules, and receives various formats of images, sensor data, and network signals. The signals of each device are uniformly processed and input into the system to ensure the unity of data formats.

[0056] Among them, each device accesses the system through a standardized interface, avoiding compatibility issues and ensuring the stable transmission of data. The system coordinates the inputs of different external devices through an internal bus, improving the efficiency of multi-device collaborative work.

[0057] Thus, it is realized that the system supports the access of multiple external devices, can be flexibly configured according to requirements, and the input of multi-device data enhances the system's ability to handle multiple tasks and improves the computing efficiency.

[0058] Embodiment 3

[0059] Optionally, step b further includes signal denoising, calibration, and format conversion to ensure signal accuracy and the system's response speed.

[0060] Optionally, step b includes the following sub-steps:

[0061] b1. Signal filtering, removing the noise part of the signal through a filter;

[0062] b2. Data format conversion, converting the received signal into a unified format for easy system processing;

[0063] b3. Real-time analysis, analyzing and modeling the converted data for decision-making.

[0064] In a possible implementation, the input signal passes through a digital signal processing module to perform denoising, calibration, and format conversion. After signal preprocessing, the system transmits the signals in a unified format to the main processor for further processing.

[0065] This includes the processes of denoising, calibration, and format conversion. External interference is removed through a filtering algorithm to ensure signal clarity; deviation correction is performed on the received data according to system requirements to ensure signal accuracy; data from different sources and formats is unified into a standard format that the system can process.

[0066] Thus, denoising and calibration ensure that the data received by the system is more accurate, and format conversion makes signal processing more efficient, improving the overall performance of the system.

[0067] Example 4

[0068] Optionally, the hardware module detection in step c includes:

[0069] c1. Electrical connection detection to confirm that all connection points are not open or short-circuited;

[0070] c2. Signal integrity detection to perform integrity detection on the signal transmission of each interface to ensure no signal loss or interference;

[0071] c3. Power management detection to ensure that the power supply voltage and current of the system are within a safe range to avoid hardware damage caused by overvoltage or overcurrent.

[0072] In a possible implementation, the signal processing module uses a hardware-accelerated filtering algorithm for denoising to ensure signal quality. After all received signals are unified in format, data modeling and decision support are performed through a real-time analysis module.

[0073] Among them, a high-performance filter is used to remove noise, improve signal quality, ensure that signals in multiple formats can be recognized and efficiently processed by the system, and real-time modeling is performed on the processed signals through an algorithm to provide a basis for decision-making.

[0074] It can be seen that signal filtering and format conversion optimize the processing speed of the data stream. Through real-time data analysis, the system can make fast and accurate decision responses.

[0075] Example 5

[0076] Optionally, the real-time monitoring in step d includes:

[0077] d1. Temperature monitoring to monitor the temperature changes of each component of the system in real time through a temperature sensor to prevent overheating;

[0078] d2. Voltage monitoring, detecting voltage fluctuations in the system power supply to avoid system instability caused by abnormal voltage;

[0079] d3. Signal transmission monitoring, periodically checking all transmitted signals to ensure no packet loss or delay.

[0080] In a possible implementation, the electrical connection and signal integrity detection are completed through an embedded test tool, and the power management detection is monitored using a dedicated power module.

[0081] An electrical test tool is used to check the stability of the connection to ensure the stable operation of the system. A high-precision signal detection module is used to monitor the data integrity during transmission. The voltage and current are monitored through a power management module to prevent equipment damage caused by overcurrent or overvoltage.

[0082] Thus, the electrical connection and signal integrity detection ensure the long-term stability of the hardware module, and the power management detection ensures the stable power supply of the system and avoids electrical faults.

[0083] Example 6

[0084] Optionally, the method further includes:

[0085] a. Connecting the hardware components to the motherboard in a predetermined order, including connecting the Jetson AGX Orin platform to other interface modules;

[0086] b. Connecting external devices through standardized interfaces, such as USB interfaces, GMSL camera interfaces, Ethernet interfaces, etc.;

[0087] c. Conducting electrical tests on all connection points, using tools such as multimeters to check the electrical stability of the hardware component connections;

[0088] d. After completion of the assembly, initializing the system and verifying the system correctness through the control module, checking whether each module during the system startup process is working properly.

[0089] In a possible implementation, the system is equipped with temperature, voltage and signal monitoring sensors to adjust the device state in real time through the control system. Among them, the temperature sensor monitors the working temperature of the device to prevent hardware damage caused by overheating; the voltage monitoring ensures stable power output and avoids the impact of voltage fluctuations on the device; the signal transmission quality is detected in real time to ensure smooth communication between the system and external devices.

[0090] Thus, the temperature and voltage monitoring helps to extend the service life of the hardware device and prevent failures caused by overheating or abnormal voltage; the signal monitoring ensures the smooth flow of data and improves the overall efficiency of the system.

[0091] Further, the step a includes:

[0092] a1. Connect the power supply module to ensure stable power input to the system;

[0093] a2. Connect each interface module, including the GMSL interface, USB interface, and Ethernet module, to ensure stable data transmission.

[0094] In a possible implementation, first connect the power supply module to ensure stable voltage input. Then connect each interface module to ensure unobstructed data paths.

[0095] Among them, a stable power supply is provided through the power supply module to avoid voltage fluctuations. The connection of the interface module ensures smooth data flow between devices and meets the system's computing requirements.

[0096] Thus, ensuring power supply achieves stable system power and avoids the impact of voltage fluctuations on device operation. The correct connection of the interface module guarantees stable data transmission and avoids performance degradation caused by connection problems.

[0097] The following is an example in combination with the application scenarios applicable to this application.

[0098] Application Scenario 1, an internal control method in an autonomous driving system.

[0099] In an autonomous driving system, external devices such as lidar, cameras, GPS modules, and IMUs (Inertial Measurement Units) provide real-time images, sensor data, and position and speed information. The system imports this data into the main control system through interfaces such as USB and Ethernet.

[0100] The autonomous driving system needs to process the received data and perform analysis. For example, the image signal undergoes denoising processing, and real-time analysis of obstacles in front of the vehicle is carried out. The accuracy of lidar data is improved through calibration. Format conversion converts data from different sources into a unified format for easy processing and fusion.

[0101] When the system starts, it automatically detects the connection status of modules such as cameras, radars, computing platforms (such as Jetson AGX Orin), and power supplies. Ensure normal data transmission for each sensor, good status of the computing platform, and sufficient power supply.

[0102] Monitor the temperature, power, voltage, etc. of each hardware module of the vehicle in real time. Through continuous monitoring by system sensors, if the sensor temperature is too high or the voltage is unstable, the system will trigger an alarm and adjust the status or reduce the load to avoid hardware damage or system crashes.

[0103] In the context of autonomous driving, signal denoising and format conversion ensure that the system can extract useful information from sensor data from different sources, guaranteeing the accuracy of real-time computing decisions. The real-time monitoring function ensures the stability of the system during long-term operation, preventing overheating or voltage problems from affecting the system. Automatically detecting the status of hardware modules can effectively prevent sensor failures or power supply issues from affecting driving safety.

[0104] Application scenario 2, intelligent robot control system.

[0105] Various sensors equipped on intelligent robots (such as RGB cameras, depth cameras, temperature sensors, etc.) will input images and sensor data into the main control system through interfaces such as USB, GMSL, and Ethernet.

[0106] In the robot control system, image signals and sensor data need to undergo denoising processing to reduce errors generated by sensors. At the same time, through format conversion, various signals are unified into a format suitable for calculation.

[0107] Real-time analysis will make decisions based on image recognition and sensor data fusion and command the robot to execute actions.

[0108] When the system starts up, the robot control platform will detect each hardware module, such as the power module, sensor module, and computing module. The system will ensure that all modules are correctly connected and in working condition.

[0109] The system will monitor the temperature, power, and signal integrity of each hardware module in the robot control system in real time. If there is a situation of insufficient voltage or too high temperature, the system will make adjustments or issue a fault report.

[0110] By automatically detecting hardware modules, it is ensured that the hardware devices of intelligent robots are always in a normal state during operation, reducing fault shutdowns. Real-time monitoring guarantees that the system can operate stably for a long time under complex operations or high loads, preventing the robot from getting out of control due to hardware problems. Data processing and signal format conversion improve the robot's perception ability of the environment, enabling it to efficiently complete tasks.

[0111] Application scenario 3: Intelligent manufacturing and Industrial Internet of Things (IIoT) system

[0112] In the intelligent manufacturing scenario, sensors, PLCs (programmable logic controllers), and other devices (such as temperature, humidity, and pressure sensors) are connected to the AI edge computing system through standardized interfaces (such as Ethernet, Modbus, etc.) to collect real-time data.

[0113] In this system, data from industrial equipment needs to be denoised and format-converted before real-time analysis. For example, temperature and pressure data are denoised to remove interference generated by sensors, and then the operating parameters of the production line are adjusted according to the converted data.

[0114] The system will automatically detect all industrial equipment connections at startup, including sensors, electrical equipment, and control units. It detects the stability of the electrical connections to ensure smooth data flow.

[0115] For an intelligent manufacturing system, the system continuously monitors the status of hardware modules, including key parameters such as voltage, current, and temperature. Once an abnormal situation (such as overheating or unstable voltage) is detected, the system will trigger an alarm and automatically adjust the operation of the production line.

[0116] Through denoising and format conversion, the system can ensure that the industrial data collected from sensors is accurate and reliable, facilitating further analysis and control. It avoids production line shutdowns caused by hardware failures or connection problems, ensuring the efficient operation of the intelligent manufacturing system. Through continuous monitoring, the system can respond quickly when potential problems are discovered, ensuring that the production line is not affected by equipment failures.

[0117] Application Scenario 4: Smart City Monitoring System

[0118] In a smart city, external devices such as cameras, environmental monitoring sensors, and traffic monitoring systems are connected to the edge computing system through various interfaces (such as USB, Ethernet, etc.) to provide real-time urban operation data.

[0119] The system denoises the video streams and sensor data received from different monitoring devices to ensure clear images and accurate sensor data. The system also unifies the data formats of different devices for subsequent processing and analysis.

[0120] After startup, the system checks all hardware modules, such as sensors, cameras, control servers, etc., to ensure that they can be normally connected and start working.

[0121] The smart city monitoring system needs to continuously monitor the status of key locations in the city to ensure the normal operation of system hardware (such as cameras, servers, etc.). Once an abnormality (such as equipment downtime, power problems, etc.) is discovered, the system can adjust or report the problem in a timely manner.

[0122] Through denoising and format conversion, the system can extract useful information from massive data, providing reliable support for decision-making. It ensures that the city monitoring system can still work stably under high-load environments, reducing the impact of failures. Data format conversion and processing ensure that the system can process large-scale monitoring data in real time, providing accurate data support for smart city management.

[0123] An embodiment of the present application further provides a computer-readable storage medium, in which at least one instruction is stored, and the at least one instruction is loaded and executed by a processor to implement the internal control method based on the embedded AI edge computing system provided in the above various embodiments.

[0124] Optionally, the computer-readable storage medium may include: read-only memory (ROM, Read Only Memory), random access memory (RAM, Random Access Memory), solid state drive (SSD, Solid State Drives), or optical disc, etc. Among them, the random access memory may include resistive random access memory (ReRAM, Resistance RandomAccess Memory) and dynamic random access memory (DRAM, Dynamic Random Access Memory).

[0125] The serial numbers of the embodiments of the present application above are only for description and do not represent the advantages and disadvantages of the embodiments.

[0126] Those of ordinary skill in the art can understand that all or part of the steps to implement the above embodiments can be completed by hardware, or can be completed by instructing relevant hardware through a program. The program can be stored in a computer-readable storage medium. The storage medium mentioned above can be read-only memory, magnetic disk, or optical disc, etc.

[0127] The above are only optional embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. An internal control method based on an embedded AI edge computing system, characterized in that: The method comprises: Receive input signals from external devices, such as images, sensor data, or network connection signals through sensors or external interfaces; Processing received input signals and controlling the system to perform signal processing operations; When the system starts, it automatically detects and verifies the working status of the hardware modules and tests all hardware modules; Monitor system status in real time.

2. The method according to claim 1, characterized in that The external devices include cameras, laser radars, USB interface devices and Ethernet modules.

3. The method according to claim 1, characterized in that The signal processing operations include signal denoising, calibration, format conversion and real-time analysis.

4. The method according to claim 1, characterized in that: The processing of the received input signal and controlling the system to perform the signal processing operation comprises: Signal filtering, removing the noise part of the signal through the filter; Data format conversion, converting the received signal into a unified format; Real-time analytics to analyze and model transformed data.

5. The method according to claim 1, characterized in that The detection of all hardware modules includes: Electrical connection test to confirm that there is no open circuit or short circuit at all connection points; Signal integrity detection: perform integrity detection on the signal transmission of each interface to determine if there is no signal loss or interference; Power management detection is used to confirm that the system's power supply voltage and current are within a safe range.

6. The method according to claim 1, characterized in that The real-time monitoring system status includes: Temperature monitoring: real-time monitoring of temperature changes of each component of the system through temperature sensors; Voltage monitoring, detecting voltage fluctuations in the system power supply; Signal transmission monitoring, periodic inspection of all transmission signals.

7. The method according to any one of claims 1 to 6, characterized in that: The method further comprises: Connect the hardware components to the mainboard in a predetermined order, including connecting the Jetson AGX Orin platform with other interface modules; Connect external devices via standardized interfaces; Conduct electrical tests on all connection points; After the assembly is completed, the system is initialized and the system correctness is verified through the control module, and each module is checked to see if it is working properly during the system startup process.

8. The method according to claim 7, characterized in that The connecting of external devices via a standardized interface comprises: Connect the power module to ensure the system's power input is stable; Connect various interface modules, including GMSL interface, USB interface and Ethernet module.

Images