Tracking type control system for automatically aiming and detecting surface defects of industrial equipment

Abstract

The invention discloses a control system for tracking type self-aiming detection of surface defects of industrial equipment, particularly relates to the technical field of industrial automatic detection, and comprises a control processing module, a tracking sensing module, a mobile control module, a holder attitude module and a detection marking module. According to the invention, through a tracking type self-aiming detection mechanism, all-directional non-blind area coverage detection of curved surface equipment and a special-shaped structure is realized. The system adopts active aiming and double-PID stable control, effectively overcomes external vibration interference, and ensures clear images and accurate marking. Modular cooperative control endows the system with high flexibility, and the system can quickly adapt to different product models. The intelligent decision-making system can dynamically adjust parameters and tracks, and has strong adaptive ability to deal with sudden working conditions. Closed-loop linkage of all the modules guarantees operation reliability, and the detection precision and the application range in a complex industrial environment are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of industrial automation inspection technology, and more specifically, to a control system for tracking-type self-aiming inspection of surface defects in industrial equipment. Background Technology

[0002] Currently, surface defect detection systems widely used in the industrial sector primarily employ a combination of fixed inspection stations and conveyor belts. In this existing technology, the industrial equipment to be inspected is transported to a fixed inspection station via a conveyor belt, where multi-angle industrial cameras mounted on a gantry acquire images of the equipment surface. The image data is transmitted via wired connection to an industrial control computer, where a pre-set image processing algorithm is used to identify defects. Subsequently, a robotic arm performs defect marking. The entire inspection process relies on the uniform speed of the conveyor belt and a high-precision positioning system, requiring pre-calibration of camera parameters and setting of fixed shooting intervals. The system coordinates the actions of each unit through a PLC controller and uses a timed triggering mechanism to control the timing synchronization of image acquisition and marking execution.

[0003] The aforementioned existing technologies have several inherent drawbacks: First, fixed inspection stations cannot adapt to the surface inspection needs of curved equipment or irregularly shaped structures, resulting in a large blind zone and a high rate of missed detections; second, conveyor belt vibration easily causes image blurring, severely affecting the accuracy of defect identification; third, the system lacks flexibility, requiring a complete redesign of the hardware layout when changing product models; furthermore, the response delay of the robotic arm marking mechanism often leads to mark position misalignment, and the system lacks the ability to adaptively adjust to sudden working conditions. These shortcomings severely limit the inspection accuracy and applicability in complex industrial environments. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, the present invention provides a tracking-type self-aiming detection system for detecting surface defects of industrial equipment, which solves the problems mentioned in the background art through the following scheme.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a tracking-type self-aiming detection system for detecting surface defects in industrial equipment, comprising:

[0006] The control processing module is used for system self-test and parameter configuration. During the tracking and aiming coordinated movement stage, it calculates the PWM drive command based on the trajectory deviation using the tracking PID control algorithm and determines the detection area based on encoder feedback. During the precise aiming and defect detection stage, it calculates the angle control quantity based on the current angle using the gimbal dual PID control algorithm, controls image acquisition and defect recognition, and triggers defect marking.

[0007] The tracking perception module is connected to the control processing module and is used to collect grayscale data of the surface trajectory and obtain the trajectory deviation value through the trajectory deviation calculation process.

[0008] A motion control module, connected to the control processing module, is used to drive the motor according to the PWM drive command and feed back motion data;

[0009] The gimbal attitude module is connected to the control processing module and is used to collect the current angle and drive the gimbal to rotate according to the angle control amount.

[0010] The detection marking module, connected to the control processing module, is used to acquire images of the equipment surface and laser mark the identified defects.

[0011] Preferably, the control processing module first performs a system self-test, confirming the initialization status by checking the communication connectivity of the tracking perception module, motion control module, and gimbal attitude module; if communication is interrupted, the alarm module is triggered; then, it receives preset parameters sent by an external terminal via the wireless communication module, including the detection trajectory type, detection area interval, target angle, laser marking duration, and PID control parameters; during the tracking, aiming, and coordinated movement phase, the control processing module calculates PWM drive commands based on the trajectory deviation data fed back by the tracking perception module using the tracking PID control algorithm and transmits them to the motion control module; simultaneously, it determines whether the preset detection area has been reached based on the vehicle's movement distance fed back by the encoder; after reaching the detection area... The control processing module sends a stop command to the motion control module and initiates the precision aiming and defect detection phase. During this phase, the control processing module receives the current angle data from the gimbal attitude module, calculates the yaw and pitch control values ​​using the gimbal's dual PID control algorithm, and drives the gimbal attitude module to adjust its angle. After aiming stabilizes, the control processing module sends a shooting command to the detection marking module and performs image preprocessing on the acquired image. After identifying defects, it controls the laser marker to mark the defects and records their location information. After detection is completed, the control processing module determines whether all areas have been detected. If not, it returns to the tracking movement phase; if so, it stores the detected data in the data storage module and uploads the data summary via the wireless communication module.

[0012] Preferably, after the control processing module is powered on, the system self-test sends handshake signals to the tracking perception module, the motion control module, and the gimbal attitude module in sequence. If no response is received within a set time, it is determined that the communication is interrupted, and the alarm module startup process is executed.

[0013] Preferably, the parameter configuration receives preset parameters in JSON format sent by an external terminal via a wireless communication module, parses them, and stores them in an internal register.

[0014] Preferably, the tracking PID control algorithm receives the trajectory deviation value e(t) calculated by the tracking sensing module, and calculates the PWM drive command using the PID formula: Where K_p, K_i, and K_d are preset parameters, t is a time variable, and U(t) is the PWM duty cycle output to the motion control module.

[0015] Preferably, the movement distance calculation is based on the number of pulses N fed back by the encoder in the movement control module, and the movement distance S of the trolley is calculated by a formula, where D is the wheel diameter and P is the number of pulses per encoder revolution.

[0016] Preferably, the gimbal dual PID control algorithm includes:

[0017] Calculation of yaw angle control parameters:

[0018]

[0019] Pitch angle control value calculation:

[0020]

[0021] Where θ_target and φ_target are preset target angles, θ_act and φ_act are the current angles fed back by the MPU6050, and K_p1, K_i1, K_d1 and K_p2, K_i2, K_d2 are PID parameters.

[0022] Preferably, the image preprocessing includes:

[0023] Gaussian filtering noise reduction: Convolution operations are performed on images captured by an industrial camera, using a Gaussian kernel matrix G:

[0024] Where σ is the filter intensity parameter, and x and y are pixel coordinates, histogram equalization enhancement: maps the image grayscale distribution to a uniform distribution, and transforms it through the cumulative distribution function CDF. Where r_k is the original gray level, s_k is the converted gray level, n_j is the number of pixels at gray level j, and N is the total number of pixels.

[0025] Preferably, the data storage and uploading process packages the defect image, location, detection time, and device status data into a binary file, writes it to the SD card of the data storage module, and simultaneously uploads the data summary to an external terminal via the wireless communication module.

[0026] Preferably, the tracking sensing module: activates the grayscale sensor according to the initialization command of the control processing module, continuously collects grayscale data of the preset trajectory on the surface of the industrial equipment; obtains the deviation value between the actual position of the trajectory and the target position through the trajectory deviation calculation process, and feeds it back to the control processing module in real time to provide input data for the tracking PID control algorithm. This module works in conjunction with the motion control module. When the control processing module outputs the PWM drive command, the tracking sensing module continuously monitors the trajectory changes and dynamically updates the deviation value to support the differential correction of the motion control module.

[0027] Preferably, the grayscale data acquisition is performed by scanning the trajectory area at a sampling frequency of 100Hz using a grayscale sensor, outputting a 0-5V analog voltage signal, which is then converted into an 8-bit grayscale value G by an ADC.

[0028] Preferably, the trajectory deviation calculation is performed by setting a grayscale value G_target corresponding to the center of the target trajectory, and the difference e between the actual collected grayscale value G_actual and G_target is the deviation value. G_target is set by a preset parameter.

[0029] Preferably, the motion control module receives PWM drive commands output by the control processing module and controls the DC motor speed through the motor drive process; the encoder collects the motor speed in real time and converts it into the trolley's moving distance and speed data, which is then fed back to the control processing module. During the tracking and aiming coordinated movement stage, the motion control module adjusts the left and right wheel motor speeds according to the differential speed command from the control processing module to correct tracking deviations; when the control processing module determines that the detection area has been reached, the motion control module receives a stop command, pauses the trolley's movement, and enters the precise aiming and defect detection stage.

[0030] Preferably, the motor drive receives PWM commands through a motor driver, adjusts the output voltage duty cycle, and controls the DC motor speed.

[0031] Preferably, the motion data feedback acquires the motor speed n (unit: rpm) through an encoder and calculates the vehicle speed v using a formula, where D is the wheel diameter.

[0032] Preferably, the differential speed correction adjusts the difference in PWM duty cycle ΔU between the left and right wheel motors based on the deviation e calculated by the control processing module: where K_diff is the differential speed coefficient, which is set by preset parameters.

[0033] Preferably, the gimbal attitude module: drives the gimbal to rotate to a preset initial angle according to the initialization command of the control processing module, and the MPU6050 attitude sensor collects the initial angle and stores it as a reference value; during the precise aiming and defect detection stage, it receives the precise aiming command from the control processing module, and the MPU6050 collects the current yaw angle θ_act and pitch angle φ_act in real time and feeds it back to the control processing module; then it receives the yaw angle control quantity U_θ and pitch angle control quantity U_φ output by the control processing module, and adjusts the gimbal angle through the stepper motor drive process until the aiming is stable, and then notifies the control processing module to start the detection marking module.

[0034] Preferably, the angle acquisition is performed by outputting triaxial acceleration and angular velocity data at a frequency of 50Hz through the MPU6050, and θ_act and φ_act are calculated by complementary filtering algorithm.

[0035] Preferably, the stepper motor drive receives U_θ and U_φ (in degrees) through a stepper motor driver and converts them into the number of stepping pulses N_step: , where α is the step angle of the stepper motor.

[0036] Preferably, the detection and marking module: after the gimbal attitude module completes aiming and stabilization, it receives the shooting instruction from the control processing module. The industrial camera continuously captures three images of the equipment surface and transmits them to the control processing module. After the control processing module performs image preprocessing and defect identification, if a defect is found, it sends a marking instruction to the detection and marking module. The laser marker starts marking the defect location for a preset time. This module is linked with the data storage module, and the defect images and marking records are stored uniformly by the control processing module.

[0037] Preferably, the image capture is performed by an industrial camera capturing a 1280×720 resolution image under a trigger signal, and then transmitting the image to the control processing module via an SPI interface.

[0038] Preferably, the laser marker receives a TTL level signal through a laser marker and emits light continuously for 0.5 seconds (the duration is set by preset parameters).

[0039] Preferably, the data storage module receives the inspection data packaged by the control processing module after inspection, including defect images, defect locations, inspection time, equipment operating status, and fault records, and writes it to the SD card for process storage as a FAT32 format file; after storage, it supports the control processing module to upload data summaries via the wireless communication module.

[0040] Preferably, the SD card write process control module writes data packets to the SD card via the SPI interface, and the file naming rule is "detection time_device ID.bin".

[0041] Preferably, the wireless communication module receives preset parameters and instructions sent by an external terminal during the system initialization phase and feeds back a system ready signal to the external terminal; after detection is completed, it receives a data upload instruction from the control processing module and sends a data summary to the external terminal. This module works in conjunction with the alarm module to synchronously upload fault details in the event of a fault.

[0042] Preferably, the data transmission is based on the TCP / IP protocol, with the external terminal acting as a server and the wireless communication module acting as a client, periodically sending heartbeat packets to maintain the connection.

[0043] Preferably, the alarm module receives the alarm command from the control processing module when the control processing module detects an initialization failure, module communication interruption, or track loss, and issues a warning signal through a buzzer; at the same time, it works with the wireless communication module to upload the fault type to an external terminal.

[0044] Preferably, the alarm trigger receives a PWM signal (frequency 2kHz, duty cycle 50%) via a buzzer and sounds continuously until the control processing module sends a stop signal.

[0045] Preferably, a method for detecting surface defects in industrial equipment includes the following steps:

[0046] The system initializes, performs a self-test, and receives preset parameters;

[0047] Tracking and aiming coordinated movement: The control processing module generates PWM drive commands to control the movement of the movement control module based on the trajectory deviation value fed back by the tracking perception module through the tracking PID control algorithm, and at the same time determines whether the detection area has been reached based on the movement distance;

[0048] Precise aiming and defect detection: After reaching the detection area, the control processing module generates an angle control quantity through the gimbal dual PID control algorithm to control the gimbal attitude module to aim, and then triggers the detection marking module to capture images and perform defect identification and marking;

[0049] Data storage and uploading: The detection data is stored in the data storage module, and the data summary is uploaded through the wireless communication module.

[0050] The technical effects and advantages of this invention are as follows:

[0051] 1. This technical solution effectively overcomes the limitations of fixed inspection stations through a tracking-type self-aiming inspection mechanism. The collaborative work of the motion control module and the gimbal attitude module enables the inspection system to adaptively follow the surface contour of the equipment, achieving all-round coverage inspection of curved equipment and irregular structures. The real-time trajectory tracking capability of the tracking perception module ensures the accuracy of the inspection path, fundamentally eliminating the problem of blind spots in the inspection.

[0052] 2. The system adopts an active aiming and stable control strategy to significantly improve detection accuracy. The dual PID control algorithm of the gimbal attitude module ensures the stability of the aiming process and avoids image blurring caused by external vibration. The instant triggering mechanism and precise positioning function of the detection mark module solve the technical problem of mark position offset. At the same time, the overall system design has good flexibility and can quickly adapt to the detection needs of different models of products.

[0053] 3. The intelligent decision-making and multi-module collaborative mechanism of the control and processing module endow the system with strong adaptive capabilities. Through real-time data acquisition and processing, the system can dynamically adjust detection parameters and motion trajectory to effectively cope with various sudden working conditions. The closed-loop control and status feedback between modules ensure the reliability and stability of the system operation, greatly improving its applicability in complex industrial environments. Attached Figure Description

[0054] Figure 1 This is a system block diagram of a tracking-based self-aiming detection system for surface defects in industrial equipment.

[0055] Figure 2 This is a flowchart illustrating the specific implementation of a tracking-based self-aiming detection system for surface defects in industrial equipment. Detailed Implementation

[0056] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0057] refer to Figures 1-2 The control system shown is a tracking-type self-aiming detection system for surface defects in industrial equipment, comprising:

[0058] Control Processing Module: First, the system performs a self-test, confirming the initialization status by checking the communication connectivity of the tracking sensing module, motion control module, and gimbal attitude module; if communication is interrupted, the alarm module is triggered; then, it receives preset parameters sent by the external terminal through the wireless communication module, including the detection trajectory type, detection area interval, target angle, laser marking duration, and PID control parameters; during the tracking, aiming, and coordinated movement phase, the control processing module calculates the PWM drive command based on the trajectory deviation data fed back by the tracking sensing module using the tracking PID control algorithm and transmits it to the motion control module; simultaneously, it determines whether the preset detection area has been reached based on the vehicle's movement distance fed back by the encoder; after reaching the detection area, the control... The control processing module sends a stop command to the motion control module and initiates the precision aiming and defect detection phase. During this phase, the control processing module receives the current angle data from the gimbal attitude module, calculates the yaw and pitch control values ​​using the gimbal's dual PID control algorithm, and drives the gimbal attitude module to adjust its angle. After aiming stabilizes, the module sends a shooting command to the detection marking module and performs image preprocessing on the acquired image. After identifying defects, it controls the laser marker to mark the defects and records their location information. After detection is completed, the control processing module determines whether all areas have been detected. If not, it returns to the tracking movement phase; if so, it stores the detected data in the data storage module and uploads the data summary via the wireless communication module.

[0059] After the control processing module is powered on, the system self-test sends handshake signals to the tracking perception module, motion control module, and gimbal attitude module in sequence. If no response is received within a set time, it is determined that the communication is interrupted, and the alarm module startup process is executed.

[0060] The parameter configuration receives preset parameters in JSON format sent by an external terminal through the wireless communication module, parses them, and stores them in an internal register.

[0061] The tracking PID control algorithm receives the trajectory deviation value e(t) calculated by the tracking sensing module, and calculates the PWM drive command using the PID formula: Where K_p, K_i, and K_d are preset parameters, t is a time variable, and U(t) is the PWM duty cycle output to the motion control module.

[0062] The movement distance is calculated based on the number of pulses N fed back by the encoder in the movement control module, using the formula... Calculate the distance S that the trolley travels, where D is the wheel diameter and P is the number of encoder pulses per revolution.

[0063] The gimbal dual PID control algorithm includes:

[0064] Calculation of yaw angle control parameters:

[0065]

[0066] Pitch angle control value calculation:

[0067]

[0068] Where θ_target and φ_target are preset target angles, θ_act and φ_act are the current angles fed back by the MPU6050, and K_p1, K_i1, K_d1 and K_p2, K_i2, K_d2 are PID parameters.

[0069] The image preprocessing includes:

[0070] Gaussian filtering noise reduction: Convolution operations are performed on images captured by an industrial camera, using a Gaussian kernel matrix G:

[0071] Where σ is the filter intensity parameter, and x and y are pixel coordinates, histogram equalization enhancement: maps the image grayscale distribution to a uniform distribution, and transforms it through the cumulative distribution function CDF. Where r_k is the original gray level, s_k is the converted gray level, n_j is the number of pixels at gray level j, and N is the total number of pixels.

[0072] The defect identification is achieved by calculating the grayscale gradient of the preprocessed image. If the gradient value of a continuous region exceeds the threshold Th, it is determined to be a defect region.

[0073] The data storage and uploading process packages defect images, locations, detection times, and device status data into binary files, writes them to the SD card of the data storage module, and simultaneously uploads data summaries to an external terminal via a wireless communication module.

[0074] Tracking Sensing Module: Based on the initialization command of the control processing module, the grayscale sensor is activated to continuously collect grayscale data of the preset trajectory on the surface of the industrial equipment. The deviation value between the actual position of the trajectory and the target position is obtained through the trajectory deviation calculation process and fed back to the control processing module in real time to provide input data for the tracking PID control algorithm. This module works in conjunction with the motion control module. When the control processing module outputs the PWM drive command, the tracking sensing module continuously monitors the trajectory changes and dynamically updates the deviation value to support the differential correction of the motion control module.

[0075] The grayscale data acquisition is achieved by scanning the trajectory area at a sampling frequency of 100Hz using a grayscale sensor, outputting a 0-5V analog voltage signal, which is then converted into an 8-bit grayscale value G by an ADC.

[0076] The trajectory deviation calculation is performed by setting the grayscale value G_target corresponding to the center of the target trajectory, and the difference e between the actual collected grayscale value G_actual and G_target is the deviation value: G_target is set by a preset parameter.

[0077] The motion control module receives PWM drive commands from the control processing module and controls the DC motor speed through the motor drive process. The encoder collects the motor speed in real time and converts it into data on the vehicle's moving distance and speed, which is then fed back to the control processing module. During the tracking and aiming coordinated movement phase, the motion control module adjusts the speed of the left and right wheel motors according to the differential speed command from the control processing module to correct tracking deviations. When the control processing module determines that the detection area has been reached, the motion control module receives a stop command, pauses the vehicle's movement, and transitions to the precise aiming and defect detection phase.

[0078] The motor drive receives PWM commands through a motor driver, adjusts the output voltage duty cycle, and controls the speed of the DC motor.

[0079] The motion data feedback is obtained by acquiring the motor speed n (unit: rpm) through an encoder, and then processed by the formula... Calculate the speed v of the trolley, where D is the wheel diameter.

[0080] The differential speed correction adjusts the difference in PWM duty cycle ΔU between the left and right wheel motors based on the deviation e calculated by the control processing module. Where K_diff is the differential coefficient, which is set by preset parameters.

[0081] Gimbal Attitude Module: Based on the initialization command from the control processing module, the gimbal is driven to rotate to a preset initial angle. The MPU6050 attitude sensor collects the initial angle and stores it as a reference value. During the precision aiming and defect detection phase, the module receives the precision aiming command from the control processing module. The MPU6050 collects the current yaw angle θ_act and pitch angle φ_act in real time and feeds it back to the control processing module. Subsequently, the module receives the yaw angle control quantity U_θ and pitch angle control quantity U_φ output from the control processing module. The gimbal angle is adjusted through the stepper motor drive process until the aiming is stable. Then, the control processing module is notified to start the detection marking module.

[0082] The angle acquisition is achieved by outputting triaxial acceleration and angular velocity data at a frequency of 50Hz through the MPU6050, and θ_act and φ_act are calculated by complementary filtering algorithm.

[0083] The stepper motor driver receives U_θ and U_φ (in degrees) through a stepper motor driver and converts them into the number of stepping pulses N_step: , where α is the step angle of the stepper motor.

[0084] Detection and marking module: After the gimbal attitude module completes aiming and stabilization, it receives the shooting instruction from the control processing module. The industrial camera continuously captures three images of the equipment surface and transmits them to the control processing module. After the control processing module performs image preprocessing and defect identification, if a defect is found, it sends a marking instruction to the detection and marking module. The laser marker starts marking the defect location for a preset time. This module is linked with the data storage module, and the defect images and marking records are stored uniformly by the control processing module.

[0085] The image capture is performed by an industrial camera capturing a 1280×720 resolution image under a trigger signal, and then transmitting the image to the control processing module via an SPI interface.

[0086] The laser marker receives a TTL level signal through a laser marker and emits light continuously for 0.5 seconds (duration is set by preset parameters).

[0087] Data storage module: Receives the inspection data packaged by the control processing module after inspection, including defect images, defect locations, inspection time, equipment operating status, and fault records, and writes it to the SD card for process storage as a FAT32 format file; after storage, it supports the control processing module to upload data summaries via the wireless communication module.

[0088] The SD card write process control module writes data packets to the SD card via the SPI interface, and the file naming rule is "detection time_device ID.bin".

[0089] Wireless communication module: During the system initialization phase, it receives preset parameters and instructions sent by the external terminal and feeds back a system ready signal to the external terminal; after the detection is completed, it receives the data upload instruction from the control processing module and sends the data summary to the external terminal. This module works in conjunction with the alarm module to synchronously upload fault details when a fault occurs.

[0090] The data transmission is based on the TCP / IP protocol, with the external terminal acting as the server and the wireless communication module acting as the client, periodically sending heartbeat packets to maintain the connection.

[0091] Alarm module: When the control processing module detects initialization failure, module communication interruption, or track loss, it receives the alarm command from the control processing module and emits a warning signal through a buzzer; at the same time, it works with the wireless communication module to upload the fault type to an external terminal.

[0092] The alarm is triggered by receiving a PWM signal (frequency 2kHz, duty cycle 50%) via a buzzer, which continues to sound until the control processing module sends a stop signal.

[0093] A method for detecting surface defects in industrial equipment includes the following steps:

[0094] The system initializes, performs a self-test, and receives preset parameters;

[0095] Tracking and aiming coordinated movement: The control processing module generates PWM drive commands to control the movement of the movement control module based on the trajectory deviation value fed back by the tracking perception module through the tracking PID control algorithm, and at the same time determines whether the detection area has been reached based on the movement distance;

[0096] Precise aiming and defect detection: After reaching the detection area, the control processing module generates an angle control quantity through the gimbal dual PID control algorithm to control the gimbal attitude module to aim, and then triggers the detection marking module to capture images and perform defect identification and marking;

[0097] Data storage and uploading: The detection data is stored in the data storage module, and the data summary is uploaded through the wireless communication module.

[0098] Secondly: The accompanying drawings of the embodiments disclosed in this invention only involve the structures involved in the embodiments disclosed in this invention. Other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of this invention can be combined with each other.

[0099] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A control system for detecting surface defects in industrial equipment using a tracking-based self-aiming mechanism, characterized in that, include: The control processing module is used for system self-test and parameter configuration. During the tracking and aiming coordinated movement stage, it calculates the PWM drive command based on the trajectory deviation using the tracking PID control algorithm and determines the detection area based on encoder feedback. During the precise aiming and defect detection stage, it calculates the angle control quantity based on the current angle using the gimbal dual PID control algorithm, controls image acquisition and defect recognition, and triggers defect marking. The tracking perception module is connected to the control processing module and is used to collect grayscale data of the surface trajectory and obtain the trajectory deviation value through the trajectory deviation calculation process. A motion control module, connected to the control processing module, is used to drive the motor according to the PWM drive command and feed back motion data; The gimbal attitude module is connected to the control processing module and is used to collect the current angle and drive the gimbal to rotate according to the angle control amount. The detection marking module, connected to the control processing module, is used to acquire images of the equipment surface and laser mark the identified defects.

2. The control system for detecting surface defects in industrial equipment using a tracking-type self-aiming method according to claim 1, characterized in that: The formula for the line-following PID control algorithm executed by the control processing module is as follows: Where U(t) is the PWM drive command output to the motion control module, e(t) is the trajectory deviation value fed back by the tracking perception module, and K_p, Ki, and K_d are preset PID control parameters.

3. The control system for detecting surface defects in industrial equipment using a tracking-type self-aiming method according to claim 1, characterized in that: The control processing module executes the gimbal dual PID control algorithm to calculate the yaw angle control quantity U_θ and the pitch angle control quantity U_φ, respectively, with the following formulas: , Where θ_target and φ_target are preset target angles, θ_act and φ_act are the current angles fed back by the gimbal attitude module, and K_p1, K_i1, K_d1, K_p2, K_i2, and K_d2 are preset PID control parameters.

4. The control system for detecting surface defects in industrial equipment using a tracking-type self-aiming method according to claim 1, characterized in that: The control processing module performs an image preprocessing procedure on the acquired images, including: Gaussian filtering for noise reduction, using a Gaussian kernel matrix. Perform a convolution operation on the image, where σ is the filter intensity parameter; Histogram equalization enhancement is achieved through the cumulative distribution function. The function converts the image grayscale level, where r_k is the original grayscale level, s_k is the converted grayscale level, n_j is the number of pixels at grayscale level j, and N is the total number of pixels.

5. The control system for detecting surface defects in industrial equipment using a tracking-type self-aiming method according to claim 1, characterized in that: The motion control module adjusts the speeds of the left and right wheel motors according to the instructions of the control processing module using a differential speed correction process. The formula for calculating the difference ΔU is as follows: , where K_diff is the preset differential coefficient and e is the trajectory deviation value.

6. The control system for detecting surface defects in industrial equipment using a tracking-based self-aiming method according to claim 1, characterized in that, Also includes: A data storage module, connected to the control processing module, is used to store detection data; The wireless communication module is connected to the control processing module and is used to receive parameters and upload data with an external terminal.

7. A method for detecting surface defects in industrial equipment based on the system described in any one of claims 1-6, characterized in that, Includes the following steps: The system initializes, performs a self-test, and receives preset parameters; Tracking and aiming coordinated movement: The control processing module generates PWM drive commands to control the movement of the movement control module based on the trajectory deviation value fed back by the tracking sensing module through the tracking PID control algorithm, and at the same time determines whether the detection area has been reached based on the movement distance; Precise aiming and defect detection: After reaching the detection area, the control processing module generates an angle control quantity through the gimbal dual PID control algorithm to control the gimbal attitude module to aim, and then triggers the detection marking module to capture images and perform defect identification and marking; Data storage and uploading: The detection data is stored in the data storage module, and the data summary is uploaded through the wireless communication module.

8. The method according to claim 7, characterized in that, If the self-test fails during the system initialization step, the alarm module is triggered; if trajectory loss is detected during the tracking and aiming coordinated movement step, the alarm module is also triggered.

Images