Flow automation hardware self-calibration control method fusing current and vision

Abstract

The invention discloses a flow automation hardware self-calibration control method fusing current and vision, and belongs to the technical field of flow automation, the flow automation hardware self-calibration control method fusing current and vision comprises the following steps: sending an operation instruction to a target host through a USB HID simulation module; and the current sensing feedback module is connected in series on a power supply path of the USB HID simulation module. Through current and vision dual-channel physical verification, the limitation that traditional process automation depends on software for indirect inference is overcome, whether physical operation is actually executed or not can be directly verified, misjudgment caused by system blockage or signal loss is effectively avoided, verification delay is remarkably shortened through millisecond-level current feature detection, and the verification efficiency is improved. Compared with a traditional visual scheme, the order of magnitude is improved. And in combination with dual-channel time sequence alignment analysis, the misjudgment rate caused by environmental interference is greatly reduced.

Description

Technical Field

[0001] This invention belongs to the field of process automation technology, specifically relating to a hardware self-calibration control method for process automation that integrates current and vision. Background Technology

[0002] With the advancement of intelligent manufacturing and informatization, process automation is gradually expanding from software-level operations to the field of hardware and software collaborative control. Traditional automation systems mainly rely on software programs to simulate manual operations, achieving the automatic execution of regular and repetitive tasks, and are widely used in industries such as finance, manufacturing, and logistics. However, most existing automation systems rely on the execution of virtual instructions at the software level, and their judgments are mainly based on screen status or system logs, making it impossible to directly perceive the actual execution results at the physical layer.

[0003] However, existing Chinese patent CN114580360B discloses an automated correction system that uses AI algorithms to correct image recognition errors. This automated system relies on the execution of virtual instructions at the software layer, and its judgments are primarily based on screen status or system logs, making it impossible to directly perceive the actual execution results at the physical layer. In scenarios involving peripheral control (such as USB keyboards and mice, barcode scanners, and touchscreens) or industrial equipment linkage, this type of solution often suffers from high feedback latency, poor environmental adaptability, and weak hardware compatibility. Especially in secure, isolated environments or environments without a network, the lack of real-time monitoring and closed-loop verification of the physical operation process severely restricts the reliability and versatility of the automated system. Therefore, there is an urgent need for a non-invasive intelligent hardware device that can monitor physical execution signals in real time, and achieve physical-level closed-loop verification of the operation process without modifying peripherals by integrating multi-modal signal acquisition. Summary of the Invention

[0004] The main objective of this invention is to overcome the problems of lack of operational effectiveness verification, high feedback delay, weak adaptability and poor hardware compatibility in the prior art, and to provide a process automation hardware self-calibration control method and system that integrates current and vision.

[0005] The technical solution adopted to solve the above technical problems is: To achieve the above objectives, this invention provides a hardware self-calibration control method for process automation that integrates current and vision, comprising the following steps: a) Send operation commands to the target host via the USB HID emulation module; b) Capture the characteristic current waveform generated by the operation command in real time by using a current sensing feedback module connected in series in the power path of the USB HID analog module; c) Synchronously capture the screen image of the target host after executing the operation command using the VGA signal acquisition module; d) The timing relationship between the characteristic current waveform and the screen image is analyzed by the central processing unit to verify the effectiveness of the operation; e) When the operation is determined to have failed according to step d), select and execute the corresponding adaptive retry operation from the preset retry strategy library according to the type of the characteristic current waveform.

[0006] Through the above technical solution, a non-intrusive hardware-level closed-loop verification system is constructed. Utilizing the physical characteristic that USB operations inevitably cause transient current changes at the port, the current signal serves as physical evidence that the operation has been "issued," and is dually verified against the visual image representing the "result" of the operation. Specifically, when the USB HID simulation module issues a command, the current sensing feedback module immediately captures characteristic pulses at the milliampere and millisecond levels, signifying the execution of the physical layer action. Simultaneously, the VGA signal acquisition module independently acquires screen changes, and the central processing unit rigorously aligns and analyzes these two signals from different sources but with related timing within a preset time window of hundreds of milliseconds. This design fundamentally solves the misjudgment problem caused by false interface changes or system delays in traditional software solutions, achieving direct physical-level verification of the operation's validity and increasing the accuracy rate to over 99.5%. Furthermore, due to the extremely low latency of current detection, the overall verification latency is significantly reduced, providing a real-time and reliable data foundation for subsequent intelligent decision-making.

[0007] Further, step d) specifically includes: If the characteristic current waveform is detected and the screen image changes within the preset attention area within the preset time window, the operation is considered successful. If the characteristic current waveform is detected within the time window but no change in the screen image is detected, it is determined to be a click non-response fault. If the characteristic current waveform is not detected or continuous current fluctuation is detected within the time window, it is determined to be a system delay or equipment failure.

[0008] By employing the aforementioned technical solutions, the root causes of operational failures can be meticulously analyzed. By examining the combined states of dual-channel signals, a multi-dimensional fault diagnosis model can be established. For example, "current pulses but no screen changes" indicates a system response or coordinate positioning problem, while "no current pulses or abnormal current" directly points to a USB communication or device malfunction. This fault classification method based on physical signal characteristics has the significant effect of accurately locating the root cause of the problem, avoiding the blind retries of traditional "one-size-fits-all" solutions, and providing a scientific basis for implementing targeted adaptive strategies. This greatly improves the efficiency of problem solving and the intelligence level of the automated system.

[0009] Furthermore, the adaptive retry operation in step e) includes: When a click failure is detected, perform a random coordinate offset retry; When a system delay fault is identified, dynamic delay compensation is performed for retry. When an interface occlusion fault is detected, a simulated window switching retry is performed.

[0010] Through the above technical solution, the accurate fault diagnosis results of the preceding steps are dynamically matched with the intelligent execution strategy to construct an adaptive closed loop of "diagnosis-decision-execution". The system no longer mechanically repeats the original instructions, but calls the optimal solution from the strategy library according to the identified fault mode. For example, it adjusts the delay for response problems, fine-tunes the coordinates for positioning problems, and switches the focus for interface problems. This breaks the rigid mode of fixed threshold retry, which greatly improves the retry success rate by more than 40%. It can also adaptively and compatiblely support devices with different response characteristics, demonstrating the high flexibility and robustness of the system.

[0011] Further, capturing the characteristic current waveform in step b) includes: The current of the USB port is sampled at a sampling rate of not less than 1MHz; The sampled data is low-pass filtered. The pulse waveform that meets the preset amplitude and width conditions is identified from the filtered data. The preset amplitude range is 5mA±1mA, and the preset width range is 3ms±0.5ms.

[0012] The above technical solution ensures the accuracy and reliability of physical layer signal acquisition. A sampling rate (1MHz) far exceeding the signal's characteristic frequency is used to capture transient current pulses without distortion. Then, a digital filtering algorithm eliminates noise interference. Finally, waveform matching is performed based on preset amplitude and width tolerance ranges, thereby accurately extracting effective operational features from complex electromagnetic environments. The significant effect of this high-precision signal processing technology is that it provides high-quality "evidence" data for the entire closed-loop verification system. It is the technological cornerstone for achieving high accuracy and low false positive rates, ensuring that the system can operate stably and reliably even in environments with electrical noise.

[0013] Furthermore, the method also includes a dynamic retry parameter self-learning step, the learning rule of which is: Δx_new = α·Δx_success + (1-α)·Δx_old Where Δx_new is the updated retry parameter, Δx_success is the parameter when retrying successfully, Δx_old is the current retry parameter, and α is the preset learning rate.

[0014] The above technical solution endows the system with the ability to self-optimize and evolve, enabling it to adapt to long-term factors such as equipment aging and environmental changes. Its working principle is to introduce an incremental learning algorithm with a forgetting factor. The system records the parameters used each time it successfully retries (such as coordinate offset) and dynamically updates the baseline parameters for the next retry according to a certain learning rate α. The system can continuously "learn" the best operating habits of specific equipment during use, automatically calibrate and optimize the retry strategy, making the automatic control effect better and better, and further improving the system's universality and long-term deployment stability.

[0015] This invention also provides a hardware self-calibration control system for process automation that integrates current and vision, comprising: The USB HID emulation module is used to connect to the USB port of the target host and send keyboard or mouse operation commands to it. A current sensing feedback module is connected in series in the power path of the USB HID analog module to capture the characteristic current waveform generated by the operation command in real time. The VGA signal acquisition module is used to connect to the graphics card output port of the target host in order to capture its screen image; The central processing unit is connected to the USB HID analog module, the current sensing feedback module, and the VGA signal acquisition module, respectively, and is used to execute the above method.

[0016] The above technical solution provides a modular hardware entity that matches the aforementioned method. It solidifies the various functional steps of the method into specific hardware modules, and works together through explicit physical interfaces and buses (such as SPI, GPIO, and parallel buses) to form a complete and independent hardware device. This system-level integration encapsulates complex software algorithms and hardware awareness capabilities into a "plug-and-play" device. Users do not need to install any software or modify the system on the target host. They only need to connect the VGA and USB interfaces to deploy it, which greatly reduces the application threshold and ensures applicability in a secure and isolated environment.

[0017] Furthermore, the current sensing feedback module includes a high-precision current sensor and a high-speed analog-to-digital converter, wherein the sampling rate of the high-speed analog-to-digital converter is not less than 1MHz.

[0018] The above technical solution provides core hardware support for achieving high-precision current waveform capture. It utilizes a high-precision current sensor (such as INA219) to achieve milliampere-level current sensing, and then uses a high-speed analog-to-digital converter (ADC) to convert the analog current signal into a digital signal stream at a frequency of not less than 1MHz. This hardware selection and configuration ensures the resolution and speed of signal acquisition from a physical perspective, which is a prerequisite for the effective execution of the aforementioned signal processing algorithm and directly determines the sensitivity and accuracy of the entire system in identifying transient current characteristics.

[0019] Furthermore, the system also includes a device feature database connected to the central processing unit for storing preset USB device current feature templates.

[0020] The above technical solutions enhance the system's scalability and compatibility with different hardware. A storable and retrieval-enabled current waveform template library is established. When a new USB peripheral is connected to the system, its characteristic waveform can be learned through a single calibration and stored in the database. Alternatively, templates for common devices can be pre-set at the factory. When performing waveform matching, the central processing unit directly calls the corresponding template for comparison. This allows the system to be no longer limited to a single or a few fixed devices, but to flexibly adapt to the vast majority of USB HID devices on the market, greatly improving hardware versatility and reducing the cost of redevelopment due to device replacement.

[0021] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described hardware self-calibration control method for fusion current and vision process automation.

[0022] Through the above technical solution, the core algorithm is solidified and protected in software form. The data processing, logical judgment, strategy selection and other algorithmic processes in the above method are compiled into computer-executable instruction code and stored in non-volatile storage medium. When the central processing unit or other general-purpose processor loads and executes the program, all the functions of the present invention can be realized. This software provides the "soul" of the hardware system, making the updating, upgrading and maintenance of the algorithm convenient. At the same time, it also provides another important form of patent protection, covering the core logical ideas of the present invention.

[0023] The present invention also provides an application of the above method, which is applied to industrial equipment linkage control in a secure isolation environment, automated operation of financial terminals, or equipment testing scenarios in a network-free environment.

[0024] Through the above technical solutions, the optimal application scenarios of this technology are clarified to highlight its core value. Its working principle is to utilize the non-intrusive, physical-level closed-loop verification characteristics of this invention to directly deploy it in environments with extremely high security and reliability requirements. For example, in industrial control, it can ensure that operation commands are actually received by physical actuators; in financial terminals, it can prevent transaction failures caused by interface latency; in network-free testing, it can independently complete automated verification. This application provides an unprecedented high-reliability automated solution for fields such as intelligent manufacturing, financial technology, and equipment testing.

[0025] The beneficial effects of this invention are as follows: (1) This invention overcomes the limitations of traditional process automation, which relies on indirect software inference, through dual-channel physical verification using both current and vision. It can directly verify whether physical operations are actually performed, thereby effectively avoiding misjudgments caused by system blockage or signal loss. Through millisecond-level current feature detection, the verification delay is significantly shortened, achieving an order-of-magnitude improvement compared to traditional vision solutions. Combined with dual-channel timing alignment analysis, the misjudgment rate caused by environmental interference is greatly reduced. (2) This invention breaks through the rigidity of the fixed threshold retry strategy by using a fault mode driven dynamic retry strategy, significantly improving the retry success rate and adapting to different devices. The system adopts a non-intrusive design, requiring no modification to the target host or peripherals, and can be deployed through standard VGA and USB interfaces. It is particularly suitable for secure isolation environments. In addition, through temperature compensation and other algorithms, it enhances the long-term operational stability in complex environments. Attached Figure Description

[0026] Figure 1 This is a schematic diagram of the algorithm and process of the present invention. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0028] Example 1: Quantitative efficacy verification experiment based on standardized testing scenarios This embodiment aims to quantitatively verify the advantages of the present invention compared to the prior art through a strictly controlled experimental environment.

[0029] 1. Experimental environment setup: Hardware: The hardware self-calibration control system of this invention (including modules such as VGA acquisition, USB HID simulation, current sensing, and central processing unit). The comparison device is a high-performance PC running automated software based on pure visual recognition (OCR / pixel matching).

[0030] Target host: A standard PC with Windows 10 operating system.

[0031] Controlled peripherals: Logitech M100 optical mouse and Lenovo M110 optical mouse, alternately connected to the target host.

[0032] Environment: A laboratory environment with a constant temperature of 25℃ and a stable illuminance of 500lx (normal office illuminance).

[0033] 2. Experimental methods and procedures: Task: Run a custom test program on the target host that displays a button at a random location on the screen. The automated system's task is to simulate clicking this button.

[0034] Test process: a. The system of the present invention and the comparison software each performed 1000 click operations.

[0035] b. The workflow of the system of the present invention is as follows: send a click command -> capture a characteristic current pulse of 5-10mA / 3ms -> capture a VGA screen within a 100ms window period -> perform frame difference analysis (ROI area pixel difference >5% is considered successful).

[0036] c. For the comparison software, its workflow is as follows: send click command -> wait 100ms -> capture screen -> perform image matching.

[0037] d. Record the result (success / failure) and total time taken for each operation.

[0038] 3. Experimental Data and Results Analysis: After 1000 click tests, the system of this invention demonstrated significant advantages in all key indicators. In terms of the accuracy rate of operation failure judgment, the system of this invention achieved 99.6%, far exceeding the 88% of the pure visual comparison scheme. Regarding response speed, the average verification latency of the system of this invention was only 22ms, while the comparison scheme required 115ms, reducing the verification time to approximately one-fifth of the original scheme. Furthermore, in 1000 tests, the system of this invention only had 4 false positives, compared to 120 false positives for the pure visual scheme, resulting in a 96.7% reduction in the false positive rate. These data fully demonstrate the superior performance of this invention in improving the reliability and efficiency of automated operations.

[0039] Results analysis: Direct verification of physical operation results: The accuracy of the system of this invention reaches 99.6%. The four failures were mainly due to current noise interference under extreme conditions. Compared with the 88% accuracy of the other solution, this verifies that the present invention, through dual-channel physical feedback at the hardware layer, effectively overcomes the limitations of indirect inference in software and avoids misjudgments caused by system blockage or signal loss.

[0040] Reduced feedback latency and false positive rate: The average verification latency of this invention is 22ms, far lower than the 115ms of the comparative scheme. This is due to the millisecond-level response with a current signal processing latency of ≤0.1ms. The extremely low number of false positives (4 times) proves the effectiveness of timing alignment analysis, eliminating the impact of screen refresh latency and dynamic interference.

[0041] 4. Environmental interference test: Based on the above experiments, an environmental interference test was conducted: the light intensity fluctuated between 500 lx and 800 lx (±300 lx), while scrolling text was played on the screen (the animation area accounted for about 15%).

[0042] The false positive rate of the system of the present invention increased to 0.8%, which is still at an extremely low level; the false positive rate of the pure visual contrast scheme soared to 18%. This result strongly proves that the present invention has extremely strong robustness to changes in lighting and dynamic elements on the screen, and verifies its objective effectiveness in controlling the false positive rate of environmental interference to an extremely low level (<1%).

[0043] Example 2: Specific Application of Dynamic Adaptive Retry Mechanism This embodiment details how the present invention executes a dynamic adaptive retry strategy based on different fault modes, and verifies its effectiveness.

[0044] 1. Scene Setup: The target host is running an old version of industrial software that is slow to respond. After clicking a button on its interface, the system response time is unstable, and occasionally the click will fail due to other pop-up windows obstructing the view.

[0045] 2. Operational procedures and strategy execution: First click: The system sends a click command, and the current sensor module captures a normal pulse (7.5mA / 3.2ms). However, after 100ms, the VGA screen shows no change in the ROI area.

[0046] Fault diagnosis: The decision engine determined the fault to be "click not responding".

[0047] Retry Strategy 1 (Random Coordinate Offset): The system performs a random coordinate offset retry, with an offset of ±2 pixels. Assume the new coordinates are (802, 599). This retry is successful, and the pixel difference in the ROI region is 12%.

[0048] Parameter self-learning: The system records the success offset Δx_success = (+2, -1). According to the learning rule Δx_new = α·Δx_success + (1-α)·Δx_old (α=0.2), the baseline offset is updated to (+0.4, -0.2).

[0049] The second click (simulating a new fault): The system sends a command, and the current sensing module captures a normal pulse. However, after 100ms, the entire screen freezes without any change.

[0050] Fault diagnosis: The decision engine determined the fault to be "system delay".

[0051] Retry Strategy Two (Dynamic Delay Compensation): The system performs a delay retransmission, initially adding a 50ms delay before retransmitting the command. If it still fails, the delay is increased in increments of 50ms (Δt=50ms) until success or the 500ms upper limit is reached.

[0052] The third click (simulating interface occlusion): The system sends a command, and the current sensing module captures a normal pulse. After 100ms, the screen changes as a whole, but the target ROI area (button) does not update and is covered by a pop-up window.

[0053] Fault diagnosis: The decision engine determined the fault to be "interface obstruction".

[0054] Retry Strategy 3 (Simulated Window Switching): The system sends the Alt+Tab key combination via the USB HID simulation module to switch windows. After waiting 200ms (T=200ms), the original click command is resent.

[0055] 3. In simulating 100 click tasks containing the above-mentioned types of faults: The system of this invention successfully executed 92 times through dynamic adaptive retries, with a retry success rate of 92%; the fixed threshold retry scheme (fixed delay of 300ms, fixed offset of ±5 pixels) only succeeded 65 times, with a retry success rate of 65%. This result strongly proves that the adaptive retry mechanism of this invention can improve the retry success rate by more than 40% and can intelligently adjust according to device characteristics (such as touch screens requiring a ±10 pixel tolerance).

[0056] Example 3: Specific Cases of Systems, Media, and Applications This embodiment describes a specific hardware system example that integrates all the functional modules required to implement the method of the present invention.

[0057] Central Processing Unit: A Xilinx Spartan-6 series FPGA chip (model XC6SLX16) is selected. The internal logic resources of this chip are sufficient to implement current signal processing, VGA frame difference analysis, and decision engine in parallel.

[0058] Current sensing feedback module: The current sensor is Texas Instruments' INA219, and the ADC is Analog Devices' ADS8320. Both are connected to the FPGA via the SPI bus, and the sampling rate is set to 1MHz.

[0059] VGA signal acquisition module: The video decoding chip uses Analog Devices' ADV7181C, and its output is connected to the FPGA's general-purpose I / O port via a 16-bit parallel bus.

[0060] USB HID emulation module: It adopts Microchip's USB protocol control chip PIC18F4550. This chip has a built-in USB engine and can be controlled by FPGA through GPIO ports to emulate a standard mouse or keyboard.

[0061] Device characteristic database: Winbond's W25Q128 serial Flash chip with a capacity of 128Mbit is selected. It is connected to the FPGA via the QSPI interface and is used to store the current characteristic templates and system firmware of various USB devices.

[0062] The specific implementation of the medium describes the specific form of the computer-readable storage medium.

[0063] Media type: The W25Q128 serial Flash chip in the above system example.

[0064] Storage content: The chip contains a FPGA configuration file written in Verilog HDL hardware description language. Once loaded onto the FPGA, this file enables all steps of the method of this invention to be implemented, including: The algorithm logic for controlling ADC sampling and performing low-pass filtering and peak detection.

[0065] The logic for receiving VGA data streams and performing frame buffering and ROI region pixel difference calculation.

[0066] Implement a decision engine, a state machine that aligns timings, determines fault modes, and selects retry strategies.

[0067] An arithmetic logic unit that implements a dynamic retry parameter self-learning algorithm.

[0068] Controls the USB HID emulation module and the interface logic for communication with the Flash memory.

[0069] When the system is powered on, the FPGA automatically reads the configuration file from the Flash chip, thus enabling it to perform all the functions of this invention.

[0070] Specific examples of applications: Application 1: Interlocking control of industrial equipment in a secure, isolated environment In the central control room of a chemical production workshop, operators need to periodically click the "Data Acquisition" button on an industrial control computer interface. This industrial control computer is located in a control network physically isolated from the external network. Using the system of this invention, no software needs to be installed on the industrial control computer. Only a VGA and USB cable connection is required to achieve 24 / 7 unattended automated data acquisition. Hardware-level verification ensures that every click is valid and avoids the loss of production data due to operation failure.

[0071] Application 2: Automated Operation of Financial Terminals At bank counters, tellers frequently need to perform a series of standardized keyboard inputs and mouse clicks on financial terminals. This invention's system can simulate these operations and ensure accurate execution of each step through physical feedback, preventing transaction failures or errors caused by interface lag or slow response, thus improving the accuracy and efficiency of business processing.

[0072] Application 3: Device testing in a network-free environment In a product quality inspection laboratory, a batch of newly manufactured wireless routers needs to undergo functional testing. The test PC is configured to operate without a network connection, and testers need to automate a series of clicks and inputs to check the router's web management interface. The system of this invention works perfectly in this environment; its non-intrusive nature does not rely on any network connection, and physical feedback ensures the reliability and repeatability of the testing process.

[0073] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention.

Claims

1. A hardware self-calibration control method for process automation that integrates current and vision, characterized in that, Includes the following steps: a) Send operation commands to the target host via the USB HID emulation module; b) Capture the characteristic current waveform generated by the operation command in real time by using a current sensing feedback module connected in series in the power path of the USB HID analog module; c) Synchronously capture the screen image of the target host after executing the operation command using the VGA signal acquisition module; d) The timing relationship between the characteristic current waveform and the screen image is analyzed by the central processing unit to verify the effectiveness of the operation; e) When the operation is determined to have failed according to step d), select and execute the corresponding adaptive retry operation from the preset retry strategy library according to the type of the characteristic current waveform.

2. The hardware self-calibration control method for process automation that integrates current and vision as described in claim 1, characterized in that, Step d) specifically includes: If the characteristic current waveform is detected and the screen image changes within the preset attention area within the preset time window, the operation is considered successful. If the characteristic current waveform is detected within the time window but no change in the screen image is detected, it is determined to be a click non-response fault. If the characteristic current waveform is not detected or continuous current fluctuation is detected within the time window, it is determined to be a system delay or equipment failure.

3. The method according to claim 1 or 2, characterized in that, The adaptive retry operation in step e) includes: When a click failure is detected, perform a random coordinate offset retry; When a system delay fault is identified, dynamic delay compensation is performed for retry. When an interface occlusion fault is detected, a simulated window switching retry is performed.

4. The hardware self-calibration control method for process automation that integrates current and vision as described in claim 1, characterized in that, The capture of the characteristic current waveform in step b) includes: The current of the USB port is sampled at a sampling rate of not less than 1MHz; The sampled data is low-pass filtered. The pulse waveform that meets the preset amplitude and width conditions is identified from the filtered data. The preset amplitude range is 5mA±1mA, and the preset width range is 3ms±0.5ms.

5. The hardware self-calibration control method for process automation that integrates current and vision as described in claim 1, characterized in that, It also includes a dynamic retry parameter self-learning step, the learning rule of which is: Δx_new = α·Δx_success + (1-α)·Δx_old Where Δx_new is the updated retry parameter, Δx_success is the parameter when retrying successfully, Δx_old is the current retry parameter, and α is the preset learning rate.

6. A hardware self-calibration control system for process automation that integrates current and vision, characterized in that, include: The USB HID emulation module is used to connect to the USB port of the target host and send keyboard or mouse operation commands to it. A current sensing feedback module is connected in series in the power path of the USB HID analog module to capture the characteristic current waveform generated by the operation command in real time. The VGA signal acquisition module is used to connect to the graphics card output port of the target host in order to capture its screen image; The central processing unit is connected to the USB HID analog module, the current sensing feedback module, and the VGA signal acquisition module, respectively.

7. The hardware self-calibration control system for process automation integrating current and vision as described in claim 6, characterized in that, The current sensing feedback module includes a high-precision current sensor and a high-speed analog-to-digital converter, wherein the sampling rate of the high-speed analog-to-digital converter is not less than 1MHz.

8. The hardware self-calibration control system for process automation that integrates current and vision as described in claim 6, characterized in that, It also includes a device feature database, connected to the central processing unit, for storing preset USB device current feature templates.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the hardware self-calibration control method for process automation that integrates current and vision as described in any one of claims 1 to 5.

10. An application of a hardware self-calibration control method for process automation that integrates current and vision, characterized in that, The method of any one of claims 1 to 5 is applied to industrial equipment linkage control, financial terminal automated operation, or equipment testing scenarios in a secure isolation environment or a network-free environment.

Images