An adaptive visual recognition and correction method for intelligent industrial robots

Abstract

The application discloses a kind of adaptive visual identification and correction methods for intelligent industrial robot, it is related to industrial automation and intelligent control technical field, the method includes S1, based on the real-time estimation result of production line object movement speed, the frame rate of visual acquisition module is regulated, S2, image data in the process of object movement is collected and initial positioning information is generated, S3, the deviation of initial positioning information and time delay parameter is analyzed, and position correction instruction is generated, S4, according to position correction instruction adjustment mechanical arm action to complete high-precision positioning operation;The adaptive visual identification and correction method for intelligent industrial robot is regulated according to the visual acquisition frame rate of object movement speed on production line to solve the positioning precision decline problem caused by time delay.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial automation and intelligent control, and particularly relates to an adaptive visual identification and correction method for an intelligent industrial robot. BACKGROUND

[0002] In the existing industrial automation production line, intelligent industrial robots are widely used in object identification, grabbing and assembly tasks. Among them, the visual identification system is a key component to realize the precise operation of the industrial robot. The existing industrial robot visual system usually adopts a fixed frame rate image acquisition method, which detects and locates the objects on the production line through image processing algorithms.

[0003] However, due to the dynamic changes in the motion speed of the objects on the production line, the visual system with fixed acquisition frame rate is difficult to adapt to the visual information acquisition requirements at different speeds. When the object speed is fast, the image acquisition delay and processing time delay will significantly increase, resulting in inaccurate position information obtained by the industrial robot, thereby affecting the positioning accuracy and overall production efficiency of the work. At present, some visual systems try to reduce the response time delay by improving the hardware performance or optimizing the image processing algorithm, but there are still the following shortcomings: on the one hand, the system cost increases significantly due to hardware upgrade, and it cannot adapt to the acquisition requirements in different speed scenarios in real time; on the other hand, although algorithm optimization can reduce the processing time delay to a certain extent, due to the constant acquisition frame rate, the system still cannot dynamically adjust to cope with high-speed moving objects, ultimately leading to decreased robot positioning accuracy and increased error. Especially in complex working conditions or high-speed production lines, this delay problem is more prominent. SUMMARY

[0004] The purpose of the present application is to provide an adaptive visual identification and correction method for an intelligent industrial robot, which adjusts the visual acquisition frame rate according to the motion speed of the objects on the production line to solve the problem of decreased positioning accuracy caused by time delay.

[0005] To achieve the above purpose, the present application provides the following technical solution: an adaptive visual identification and correction method for an intelligent industrial robot, the method comprising:

[0006] S1, based on the real-time estimation result of the motion speed of the objects on the production line, adjusting the frame rate of the visual acquisition module;

[0007] S2, acquiring image data during the motion of the object and generating initial positioning information, including selecting a reference object of a known size in the image, measuring the pixel distance of the reference object in the image, measuring the pixel distance of the target object in the image, calculating the physical distance of the target object in the actual space, and taking the calculated actual distance of the target object as the initial positioning information of the robot system;

[0008] S3, analyzing the deviation of the initial positioning information and the time delay parameter, generating a position correction instruction, including determining the basic processing time generated according to different image data, analyzing the time delay deviation of the initial positioning information, calculating the total time delay value under the current image data, generating a position correction instruction according to the total time delay value under the current image data, and ensuring the operation precision of the robot arm;

[0009] S4, adjusting the arm action according to the position correction instruction to complete the high-precision positioning operation, including determining the action point and the reaction force of the target according to the position of the target object and the correction instruction, determining the action force required by the arm, and controlling the arm to adjust according to the calculation result.

[0010] Preferably, the S1 includes determining the motion speed change rate of the target object on the production line, and calculating the image sampling frequency of the vision acquisition module over time.

[0011] Preferably, the specific steps of determining the motion speed change rate of the target object on the production line in the S1 include: detecting and recording the speed of the moving object on the production line at the current time through a vision recognition system or a sensor; calling the object motion speed at the previous time from the historical data of the system; confirming the time interval between the current time and the previous time; subtracting the speed value at the previous time from the speed value at the current time to obtain the speed change amount of the object within this time interval; and dividing the speed change amount by the time interval between the previous time and the current time to obtain the speed change rate per unit time.

[0012] Preferably, the specific steps of calculating the image sampling frequency of the vision acquisition module in the S1 include determining the time elapsed from the start of sampling to the current time according to the speed change rate per unit time, multiplying the speed increase rate by the current estimated time to obtain the sampling frequency of the vision acquisition module, and using the calculated sampling frame rate as a control parameter to adjust the sampling speed of the vision acquisition module in real time, so as to ensure that the image acquisition and the motion speed of the object are kept synchronized.

[0013] Preferably, the specific steps of calculating the physical distance of the target object in the actual space in the S2 include first taking the actual distance value of the reference object in reality; then, calculating the ratio of the measured distance of the target object in the image to the measured distance of the reference object in the image; and finally, multiplying the actual distance value of the reference object by the above ratio to obtain the physical distance of the target object in the actual space.

[0014] Preferably, the specific steps of calculating the total time delay value under the current image data in S3 include: evaluating the complexity of the current image according to the processed image data, the complexity of the current image being quantitatively analyzed according to the number of target objects, background complexity, and lighting condition factors; determining the basic processing time under an ideal state, the basic processing time being the shortest time required for image acquisition, recognition, and output of positioning information under the simplest image data condition; determining the delay increment caused by the increase of image complexity according to the value of the image complexity, the delay increment being the product of the image complexity and the unit complexity delay increment; adding the basic processing time and the delay increment caused by the increase of image complexity to obtain the total time delay value under the current image data; and taking the calculated total time delay value as the time delay compensation parameter to adjust the positioning information and the motion control instruction according to the compensation parameter.

[0015] Preferably, the specific steps of determining the force required to be applied by the mechanical arm in S4 include: determining the counterforce generated by the target object on the mechanical arm; measuring the actual distance of the target object from the force receiving point to the support point; measuring the actual distance of the mechanical arm from the position of applying the force to the fixed support point; multiplying the counterforce generated by the target object by the action distance of the object to obtain the moment generated by the target object; dividing the moment of the target object by the action distance of the mechanical arm to obtain the force required to be applied by the mechanical arm; and adjusting the output of the mechanical arm according to the force calculated above.

[0016] Preferably, S1 further includes estimating the frame rate of the visual acquisition module, specifically, calculating the frame rate by performing a ratio operation on the maximum motion speed of the object on the production line and the actual motion speed of the object currently measured according to a preset proportionality coefficient; then, detecting the initial position of the object on the production line through a sensor; then, inputting the measured object speed and current position information into the control module to calculate the adjustment amplitude of the frame rate; if the adjustment amplitude exceeds a preset threshold, triggering a real-time frame rate regulation mechanism to dynamically adjust the frame rate of the visual acquisition module; if the adjustment amplitude does not exceed the threshold, maintaining the original frame rate unchanged.

[0017] Preferably, S1 further includes first estimating the acceleration of the object on the production line by using the Kalman filtering method; then, updating the parameters in the object motion speed model according to the estimated acceleration value; then, introducing an acceleration change constraint condition to limit the difference between the updated acceleration value and the acceleration value at the previous moment; finally, combining the updated acceleration value to calculate the average motion rate in the target region, and setting the frame rate of the visual acquisition module according to the average rate.

[0018] Preferably, the updating of the parameter in the object motion speed model according to the estimated acceleration value in S1 comprises calculating a new acceleration value by weighting the acceleration value at the previous moment and the current measured object motion speed, and a smoothing factor is used in the weighting calculation, and the value range of the smoothing factor is 70%-95%.

[0019] From the above technical solutions, the present application has the following beneficial effects:

[0020] The adaptive visual recognition and correction method for intelligent industrial robots breaks through the technical limitation of fixed acquisition frame rate of the existing industrial robot vision system, and can dynamically adjust the image sampling frequency of the vision acquisition module according to the dynamic change of the object motion speed on the production line. Through this method, the vision acquisition system and the object motion state are synchronously responded, the real-time performance of image acquisition and the accuracy of data processing are greatly improved, the positioning accuracy and operation stability of the robot system are significantly improved, the image acquisition delay and processing lag problem caused by high-speed motion of the target object are effectively solved, and the industrial robot can continuously obtain clear and accurate image information under different speed conditions. Compared with the existing technology which relies on fixed frame rate and high-performance hardware, the present application does not need to significantly increase the hardware cost, reduces the system upgrade and operation and maintenance cost, has higher economy and practicability, improves the operation precision and reaction speed of the industrial robot in complex environment, effectively suppresses the positioning error caused by image acquisition and processing delay, significantly reduces the operation deviation rate, improves the application performance of the robot system on the high-speed dynamic production line, and can adapt to different types of industrial production lines and various working scenes. The dynamic adjustment strategy not only improves the production efficiency, but also enhances the adaptability of the system to complex working conditions and multi-target operation environment, significantly improves the automatic operation level and market competitiveness of the intelligent industrial robot. BRIEF DESCRIPTION OF DRAWINGS

[0021] Figure 1 The method flowchart of the present application. DETAILED DESCRIPTION

[0022] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the present application.

[0023] As Figure 1 shown, the present application provides a technical solution: an adaptive visual recognition and correction method for intelligent industrial robots, the method comprising:

[0024] S1, based on the real-time estimation result of the production line object motion speed, the frame rate of the vision acquisition module is regulated;

[0025] S2, collect image data in the process of object motion and generate initial positioning information, including selecting a reference object with a known size in the image, measuring the pixel distance of the reference object in the image, measuring the pixel distance of the target object in the image, calculating the physical distance of the target object in the actual space, and taking the calculated actual distance of the target object as the initial positioning information of the robot system;

[0026] S3, analyze the deviation of the initial positioning information and the time delay parameter, generate position correction instructions, including determining the basic processing time generated according to different image data, analyzing the time delay deviation of the initial positioning information, calculating the total time delay value under the current image data, and generating the position correction instruction according to the total time delay value under the current image data, to ensure the operation precision of the robot arm;

[0027] S4, adjust the mechanical arm action according to the position correction instruction to complete the high-precision positioning operation, including determining the action point and the reaction force of the target according to the position of the target object and the correction instruction, determining the action force required by the mechanical arm, and controlling the mechanical arm to adjust according to the calculation result.

[0028] The core of the embodiment is to realize high-precision identification and real-time positioning control of target objects in a dynamic environment through adaptive visual recognition and correction technology. First, in step S1, real-time estimation is performed based on the movement speed of the objects on the production line. The system obtains the movement parameters of the objects in real time through the speed sensor installed on the production line or through image sequence analysis technology. According to the detected movement speed, the frame rate of the visual acquisition module is dynamically adjusted to ensure that clear and effective image data can still be obtained under different speed conditions. For example, when the object movement speed is high, the acquisition frame rate is increased to avoid image blurring and ensure the integrity and timeliness of the image information. In step S2, the visual acquisition module obtains continuous image data, and identifies the reference object set in the image through image processing algorithms. The reference object has a known physical size, and by measuring its pixel distance in the image, the system can calculate the proportion coefficient of pixels and actual distance. Then, the pixel size of the target object in the image is detected and measured, and the position coordinates and size information of the target object in the actual space are calculated based on the above-mentioned proportion coefficient. The calculation process uses a computer vision-based geometric correction model to ensure the accuracy and consistency of the target object positioning data. In step S3, the system performs comprehensive analysis based on the initial positioning information and the time delay parameters introduced in the image acquisition, processing and transmission process. By statistically analyzing the basic processing time caused by different image data amounts, the system establishes a dynamic time delay model to calculate the total time delay value of the current acquisition image in real time. The time delay value includes image acquisition delay, processing delay and network transmission delay. The actual position of the target object is predicted and corrected using the time delay value to generate accurate correction instructions. The correction instructions are issued to the robot arm execution unit through the control system. In step S4, the robot control system adjusts the action of the mechanical arm according to the correction instructions. First, the action point position of the target object is determined, and the reaction force parameters in the operation process are analyzed according to the physical model. The system further calculates the required action force size and direction of the mechanical arm according to the material properties, shape and force requirement of the target object. The control unit drives the mechanical arm to perform fine adjustment operation according to the calculation result. The adjustment process can realize real-time feedback correction through closed-loop control to ensure the force control accuracy and dynamic response ability in the operation process, and improve the stability and flexibility of the mechanical arm in complex operation environment. In addition, to further improve the robustness and adaptability of the system, the embodiment also introduces multi-sensor information fusion technology to realize comprehensive perception and dynamic adjustment of the target object state by combining the data of force sensors, visual sensors and position sensors. The overall architecture design ensures the visual recognition accuracy and mechanical operation accuracy of the robot system on the high-speed dynamic production line, and improves the intelligent operation ability of the industrial robot under different complex working conditions.

[0029] The embodiment improves the accuracy and timeliness of image acquisition through real-time estimation of the object motion speed of the production line and dynamic regulation of the frame rate of the visual acquisition module. The pixel proportion model is established by using the reference object to realize high-precision positioning of the target object in the actual space. Through comprehensive analysis of the time delay of different data amounts and automatic correction of the time delay deviation, the positioning error caused by system time delay is greatly reduced, and the operation accuracy and stability of the industrial robot mechanical arm are significantly improved. In addition, based on the action point and reaction force analysis, the precise control of the mechanical arm action is realized, further enhancing the adaptability and robustness of the robot system, and improving the automation level and application breadth of the intelligent industrial robot.

[0030] S1 includes determining the motion speed variation rate of the target object on the production line and calculating the image sampling frequency of the visual acquisition module over time. The embodiment further optimizes the frame rate regulation mechanism of the visual acquisition module in S1. Specifically, by detecting and analyzing the motion speed variation rate of the target object on the production line, i.e. the variation amplitude of the target object speed per unit time, the image sampling frequency of the visual acquisition module is dynamically calculated. The system obtains the speed variation rate of the target object by installing a speed sensor in the production line or robot work area, or using a visual recognition algorithm to process consecutive image frames. The system predicts the motion trend of the target object at future time based on the real-time obtained speed variation rate data. In the process of time dimension advancement, the calculation of sampling frequency uses dynamic adjustment algorithm, such as prediction control theory or adaptive scheduling algorithm, so that the visual acquisition module can automatically adjust the sampling frequency according to different speed variation. When the speed variation rate of the target object is large, i.e. the object motion state is unstable or frequently accelerated or decelerated, the system increases the image sampling frequency to ensure that more real-time image information is collected, reducing the recognition error caused by information loss or blur; on the contrary, when the speed variation rate is small and the object motion state is stable, the image sampling frequency is appropriately reduced to save system resources and reduce data processing pressure. This adaptive regulation method enhances the response ability of the system to the dynamic motion of the target object and improves the efficiency and accuracy of the overall image processing.

[0031] By introducing real-time detection and analysis of the motion speed variation rate of the target object, the embodiment can more accurately reflect the dynamic motion state of the target object, avoiding the problems of insufficient or redundant information collection caused by fixed sampling frequency. Adaptive adjustment of image sampling frequency not only improves the ability of the system to identify and locate the target in high-speed dynamic environment, but also effectively reduces the computational load and energy consumption in the image acquisition and processing process, optimizes the allocation of system resources, and further improves the working efficiency and stability of the robot system. In addition, this method can effectively improve the adaptability of the robot to the changes of the production line, and broaden its application scenarios and practical value.

[0032] The specific steps for determining the rate of change of the movement speed of the target object on the production line in S1 include: detecting and recording the speed of the moving object at the current time on the production line in real time through a visual recognition system or a sensor; calling the speed of the object movement at the previous time from the historical data of the system; confirming the time interval between the current time and the previous time; subtracting the speed value at the previous time from the speed value at the current time to obtain the speed change amount of the object in this time interval; and dividing the speed change amount by the time interval between the previous time and the current time to obtain the rate of change of the speed per unit time.

[0033] The specific formula for determining the rate of change of the movement speed of the target object on the production line is:

[0034] wherein r represents the rate of change of the movement speed of the target object, v t represents the speed of the object movement at the current time, v t-Δt represents the speed of the object movement at the previous time, and Δt represents the time interval between the current time and the previous time.

[0035] The embodiment further refines the determination process of the rate of change of the movement speed of the target object on the production line, ensuring the accuracy and real-time nature of the sampling frequency adjustment. First, the visual recognition system or the external sensor (such as a laser speedometer, infrared sensor, or millimeter wave radar) installed on the production line continuously detects the moving target object on the production line, and records the movement speed data at the current time in real time. This speed data can be obtained through continuous frame image analysis (such as inter-frame displacement calculation) or sensor data output, and stored in the system database. Subsequently, the system automatically calls the historical speed data of the target object at the previous time from the database, forming a pair of speed sampling values at adjacent times. The system confirms the time interval between the current time and the previous time, i.e., the time sampling interval. By calculating the difference between the speed value at the current time and the speed value at the previous time, the speed change amount of the target object in this time interval is obtained. The system then divides this change amount by the time interval to calculate the instantaneous speed change rate of the target object. This real-time calculation process automatically runs in the system control unit with a fixed sampling period, ensuring continuous speed change rate data acquisition. The system dynamically adjusts the image sampling frequency of the visual acquisition module based on this data. When a significant increase in the speed change rate is detected, the system increases the image sampling frequency to capture more detailed motion state information; when the speed change rate is small, the sampling frequency is appropriately reduced to optimize the use of computing resources. In this way, the system realizes efficient visual acquisition control based on the dynamic behavior of the target object, ensuring the accuracy of subsequent positioning and operation.

[0036] The embodiment realizes fine-grained dynamic analysis of the motion state of the target object by introducing a speed change rate calculation method based on historical data comparison. Compared with the traditional single-time speed sampling method, the method can more accurately identify the acceleration, deceleration or uniform speed state of the object, thereby dynamically optimizing the image sampling frequency according to the actual requirements. The ability of the vision acquisition module to respond to system dynamics is improved, and the adaptability of the robot system to high-speed production lines and complex environments is significantly enhanced. The scheme effectively improves the accuracy of target recognition and positioning, reduces the error rate, and optimizes the overall resource allocation and energy consumption control of the system. In addition, the method has good scalability and universality, and is suitable for various industrial production line environments, which has significant significance for improving the level of production automation.

[0037] The specific steps of calculating the image sampling frequency of the vision acquisition module in S1 include determining the time elapsed from the start of sampling to the current time according to the obtained speed change rate per unit time, multiplying the speed growth rate by the current estimated time to obtain the sampling frequency of the vision acquisition module, taking the calculated sampling frame rate as a control parameter, and adjusting the sampling speed of the vision acquisition module in real time to ensure that the image acquisition and the motion speed of the object are kept synchronized.

[0038] The specific formula for calculating the image sampling frequency of the vision acquisition module is A = r x t.

[0039] Wherein, A represents the image sampling frequency of the vision acquisition module, r represents the motion speed change rate of the target object, and t represents the time.

[0040] On the basis of the foregoing, the embodiment further refines the calculation method of the image sampling frequency of the visual acquisition module. The system first obtains the speed change rate of the target object on the production line through a visual recognition system or an external sensor, that is, the speed increase or decrease rate per unit time. The speed change rate is obtained by continuously sampling the speed values of the target object at different time points and using a difference algorithm or a more complex prediction model. Subsequently, the system records the time value elapsed from the start of image sampling to the current time, that is, the sampling cumulative time T, in real time during the acquisition process. The current acquired speed change rate is multiplied by the time T to obtain the dynamic sampling frequency at the current time. This sampling frequency value represents the optimal sampling speed required by the image acquisition system under the changing motion state of the target object. The sampling frequency is transmitted to the visual acquisition module as a dynamic control parameter, and the sampling frame rate of the visual acquisition module is adjusted in real time by the control system. The control system automatically increases or decreases the image sampling rate of the visual acquisition module according to the calculation result, thereby ensuring that the time resolution of image acquisition can adapt to the real-time motion state of the target object. When the target object accelerates, the system increases the sampling frame rate to reduce the image sampling time interval, thereby preventing image blurring or positioning errors due to rapid motion of the target object; when the target object moves steadily or decelerates, the system reduces the sampling frequency to optimize the use efficiency of system resources. In addition, to ensure that the image acquisition is synchronized with the motion speed of the target object, the system also introduces a closed-loop feedback mechanism to dynamically optimize the sampling frequency based on the acquisition image quality and motion state feedback, so that the visual acquisition is always in an optimal state, thereby improving the overall recognition accuracy and data processing efficiency.

[0041] By dynamically calculating the sampling frequency based on the speed change rate and the cumulative time, the embodiment realizes the synchronous control of the visual acquisition module and the motion state of the target object. Compared with the traditional fixed sampling method, the system response sensitivity and adaptability can be significantly improved, and the integrity and clarity of image acquisition under different motion states of the target object can be ensured. This method optimizes the sampling control logic, reduces image acquisition delay and processing redundancy, and improves the overall performance and working efficiency of the system. In particular, in high-speed production lines or industrial scenes with frequent motion state changes, the embodiment effectively enhances the processing capability of the industrial robot system in complex dynamic environments, and significantly improves the accuracy and stability of automated operations.

[0042] The specific steps of calculating the physical distance of the target object in the actual space in S2 include first taking the actual distance value of the reference object in reality; then calculating the ratio of the measured distance of the target object in the image to the measured distance of the reference object in the image; and finally multiplying the actual distance value of the reference object by the above ratio to obtain the physical distance of the target object in the actual space.

[0043] The specific formula for calculating the physical distance of the target object in the actual space is:

[0044] wherein B1 represents the actual distance of the reference object, B2 represents the physical distance of the target object in the actual space, c1 represents the measured distance of the reference object in the image, and c2 represents the measured distance of the target object in the image.

[0045] This embodiment describes in detail the calculation method of the target object's actual space physical distance in the S2 step, aiming to improve the robot system's ability to accurately acquire the target object's position. First, the system needs to introduce a reference object of known size when collecting images. This reference object can be a standard calibration device with clear geometric parameters, such as a standard length ruler, a calibration block of fixed size, or a cylinder with a known diameter. By pre-recording the actual distance value of the reference object in the actual space, the system establishes the basic mapping relationship between the actual space and the image pixel coordinates. Second, the vision acquisition module detects the contour boundaries of the reference object and the target object in the image through image processing algorithms and measures their pixel distances in the image plane. The measured distance of the target object in the image and the measured distance of the reference object in the image are obtained. The system calculates the ratio of the two, which reflects the size ratio of the target object relative to the reference object. Finally, multiply the actual distance value of the known reference object by the above ratio to obtain the physical distance of the target object in the actual space, i.e. B2 = B1 * (c2 / c1). This calculation process simplifies the complex three-dimensional reconstruction or calibration process, and quickly realizes the measurement of the target object's physical distance through direct conversion based on relative proportion. The obtained distance value serves as an important part of the robot system's initial positioning information, providing accurate spatial data support for subsequent position correction and mechanical arm motion control.

[0046] By adopting the method based on the proportion conversion of the reference object, this embodiment significantly improves the simplicity and practicality of the calculation of the target object's actual distance. Compared with traditional distance measurement methods based on complex camera parameter calibration or three-dimensional point cloud reconstruction, this method is more efficient and easy to implement, especially suitable for rapid deployment and instant calculation requirements in dynamic production line scenarios. This method does not require frequent camera calibration, reducing system maintenance costs and complexity. At the same time, by using the known reference object for accurate conversion, the accuracy of the target object's spatial positioning is improved, thereby enhancing the precision and reliability of the robot system in operations such as grabbing, assembly, and detection. This method has good stability and applicability, and can be widely applied in different industrial production environments.

[0047] The specific steps of calculating the total time delay value under the current image data in S3 include: evaluating the complexity of the current image according to the processed image data, the complexity of the current image being quantitatively analyzed according to the number of target objects, background complexity and light condition factors; determining the basic processing time under the ideal state, the basic processing time being the shortest time required for completing image acquisition, recognition and outputting positioning information under the simplest image data condition; determining the delay increment caused by the increase of image complexity according to the value of the image complexity, the delay increment caused by the increase of image complexity being the product of the image complexity and the unit complexity delay increment; adding the basic processing time and the delay increment caused by the increase of image complexity to obtain the total time delay value under the current image data; and taking the calculated total time delay value as the time delay compensation parameter and adjusting the positioning information and the motion control instruction according to the compensation parameter.

[0048] The specific formula of calculating the total time delay value under the current image data is D = g + h x E.

[0049] Wherein, D represents the total time delay value under the current image data, g represents the basic processing time, h represents the unit complexity delay increment, and E represents the complexity of the current image data.

[0050] The embodiment aims to improve the time delay compensation capability of the intelligent industrial robot vision recognition system. By dynamically calculating the total time delay value of image processing, the real-time and accuracy of robot positioning and operation are ensured. First, the system analyzes the complexity of each collected image data in step S3. The complexity evaluation considers the number of target objects, background complexity, and current lighting conditions. The more target objects, the longer the image analysis time. When the background complexity is high, the algorithm needs to perform more feature extraction and denoising processing. When the lighting conditions differ greatly, the system needs to increase image enhancement or dynamic exposure compensation operations, thereby increasing the overall processing load. The system quantifies the image complexity factors through a pre-set parameter model or an image feature evaluation model based on machine learning, forming a complexity evaluation value. Second, the system sets the basic processing time under ideal conditions based on test calibration stage or experimental results under ideal conditions. The basic processing time is the shortest processing time required from image acquisition, target recognition to output positioning information when the system processes the simplest image (such as single target object, uniform background, ideal lighting). Based on the image complexity, the system calculates the delay increment caused by the complexity of the image according to the unit complexity delay growth. The unit complexity delay growth can be determined by experiment or dynamically adjusted by adaptive algorithm according to the system processing capacity and algorithm efficiency. Finally, the basic processing time and delay increment are added to obtain the total time delay value under the current image data. The total time delay value is input to the robot control system as a time delay compensation parameter to correct the real-time positioning information and motion control instructions of the target object. By compensating for the time lag of image acquisition and processing, the system realizes accurate positioning and mechanical arm operation control of dynamic moving targets, reduces the positioning deviation caused by system time delay, and improves the real-time response capability and operation precision of the robot system.

[0051] Through quantitative evaluation of image complexity and delay increment calculation, the embodiment effectively solves the problem of processing time fluctuation caused by different image data complexity, ensuring the positioning and control of target objects by the robot system with high timeliness. Compared with traditional fixed delay estimation or static model compensation methods, this method introduces dynamic image analysis and time delay adaptive calculation mechanism, significantly improving the adaptability and robustness of the system in complex production environment. Further optimizing the real-time control and fine operation level of the robot arm on the dynamic production line, reducing the product defect rate caused by time delay error, improving the overall production efficiency and process stability.

[0052] The specific steps for determining the force that the robot arm needs to apply in S4 include: determining the counterforce generated by the target object on the robot arm; measuring the actual distance from the force receiving point of the target object to the support point; measuring the actual distance from the position where the force is applied to the fixed support point of the robot arm; multiplying the counterforce generated by the target object by the action distance of the object to obtain the moment generated by the target object; dividing the moment of the target object by the action distance of the robot arm to obtain the force that the robot arm needs to apply; and adjusting the output of the robot arm according to the force obtained through the above calculation.

[0053] The specific formula for determining the force that the robot arm needs to apply is:

[0054] Wherein, X1 represents the force that the robot arm needs to apply, X2 represents the counterforce generated by the target object, Y1 represents the actual distance from the force applying point of the robot arm to its fixed support point, and Y2 represents the actual distance from the force receiving point of the target object to the support point.

[0055] The embodiment details the calculation method of the mechanical arm force applied in the S4 step, ensuring the stability and reliability of the mechanical arm when performing high-precision positioning operations. First, the system determines the counterforce generated by the target object during contact with the mechanical arm through sensors or based on physical modeling analysis. This counterforce is the reaction force generated by the object on the mechanical arm due to external force, which can be measured by force sensors, strain gauges or dynamic simulation systems. Subsequently, the system measures the actual distance between the target object from its force point (i.e. the point of external force) to the support point (i.e. the support position of the object fixing or contact surface). This distance reflects the lever arm length of the moment of the target object, which is an important parameter for calculating the moment generated by the target object. Further measure the actual distance of the mechanical arm from its force application position (such as mechanical fingers, clamps, etc.) to its fixed support point (usually the base or joint axis of the mechanical arm), which determines the lever arm length of the moment generated by the mechanical arm. After obtaining the above data, the system multiplies the counterforce generated by the target object by its action distance to calculate the moment of the target object to the outside world. In order to maintain the balance of system moment and realize accurate control, the mechanical arm needs to apply a corresponding force. This force is calculated by dividing the moment generated by the target object by the action distance of the mechanical arm. According to the calculated force, dynamically adjust the output control parameters of the mechanical arm, drive the servo motor or hydraulic actuator to apply the corresponding force, realize the stable clamping or operation of the target object, and ensure the attitude stability and operation precision of the mechanical arm in complex dynamic environment. The whole calculation and adjustment process is realized through closed-loop control system for real-time feedback, ensuring accurate force and rapid response, and improving the overall operation performance of the robot system. By using the torque balance principle to calculate the force of the mechanical arm, the embodiment effectively improves the force control precision and dynamic stability of the intelligent industrial robot in high-precision operation. Compared with the traditional method of setting force value based on experience or static model, this method introduces real-time data measurement and dynamic calculation mechanism, which can automatically adjust the force according to the changes of target object and operating environment, reduce the error and operation failure rate. Improve the flexibility and force control ability of the system, so that the mechanical arm has stronger adaptability and safety when handling target objects of different materials, shapes and weights. Further optimize the production process, improve the comprehensive performance in application scenarios such as automatic assembly, precision operation and safe interaction.

[0056] S1 further comprises estimating the frame rate of the visual acquisition module. Specifically, according to a preset proportion coefficient, the maximum movement speed of the object on the production line is compared with the actual movement speed of the object currently measured, and the frame rate is calculated. Then, the initial position of the object on the production line is detected by a sensor. Next, the measured object speed and the current position information are input into the control module to calculate the adjustment range of the frame rate. If the adjustment range exceeds the preset threshold, the real-time frame rate regulation mechanism is triggered to dynamically adjust the frame rate of the visual acquisition module. If the adjustment range does not exceed the threshold, the original frame rate is maintained unchanged.

[0057] The embodiment further improves the estimation and regulation mechanism of the frame rate of the visual acquisition module on the basis of S1. First, the system detects the current actual speed V1 of the moving object on the production line in real time through a sensor or a visual recognition system. According to the maximum movement speed V max of the object set by the system, a preset proportion coefficient K1 is used to calculate the ratio R = V max / V1 of the two, to preliminarily estimate the sampling frame rate F required by the visual acquisition module. The ratio reflects the proportion of the current object movement state relative to the maximum processing capacity of the system, to ensure that the image acquisition frequency of the visual system meets the changes in the movement speed of the object. Subsequently, the system obtains the initial position information P0 of the target object through a sensor (such as a laser range finder, an infrared sensor, a visual sensor, etc.) installed on the production line or the robot operation area. This information is input into the control module together with the real-time measured speed data V1, and the control module calculates the adjustment range AF of the sampling frame rate F compared with the current set value based on a kinematic model or a prediction algorithm. The adjustment range AF is calculated based on the current detected movement speed change and the spatial displacement change of the object, and the subsequent movement trend is predicted through dynamic analysis to determine whether real-time adjustment of the sampling frame rate is needed. If the calculated adjustment range AF exceeds the preset threshold T (i.e., AF > T), the real-time frame rate regulation mechanism is triggered. The system dynamically adjusts the sampling speed of the visual acquisition module according to the newly calculated sampling frame rate F, increases or decreases the image acquisition frequency, and ensures that the image acquisition is always synchronized with the movement state of the target object, avoiding image blur, recognition errors or information loss. If the adjustment range AF does not exceed the preset threshold T (i.e., AF ≤ T), the current sampling frame rate is maintained unchanged to reduce system resource consumption and control complexity. The real-time regulation process is implemented through a closed-loop control system, which continuously optimizes the sampling frame rate in combination with the movement state feedback to ensure efficient response of the system to the movement state of the target object on the production line and stability of the image acquisition quality.

[0058] By calculating the visual acquisition frame rate based on the ratio of the maximum motion speed of the target object to the actual motion speed, the embodiment realizes dynamic optimization control of the sampling frequency, effectively improving the sensitivity and response speed of the system to the state changes of the moving object on the production line. The introduction of adjustment amplitude judgment and threshold control mechanism avoids the problem of frequent adjustment of the sampling frame rate due to slight fluctuations, reduces the system control complexity and resource occupation, and improves the system stability and energy management efficiency. The overall scheme enhances the image acquisition capability and data processing efficiency of the robot system in a high-speed dynamic operation environment, improves the automation control level and operation reliability of the intelligent industrial robot.

[0059] S1 also includes first using Kalman filtering method to estimate the acceleration of the object on the production line; then, updating the parameters in the object motion speed model according to the estimated acceleration value; then, introducing an acceleration change constraint condition to limit the difference between the updated acceleration value and the acceleration value at the previous moment; finally, combining the updated acceleration value, calculating the average motion rate in the target region, and setting the frame rate of the visual acquisition module according to the average rate.

[0060] On the basis of S1 step, the embodiment introduces Kalman filtering method to estimate and dynamically correct the acceleration of the moving object on the production line with high precision, further optimizing the frame rate regulation mechanism of the visual acquisition module. First, the system collects the motion state data of the target object through sensors or visual recognition system, including position, speed and other basic information. Then, the Kalman filtering algorithm is used to dynamically process the collected data to eliminate interference such as sensing error, system noise, etc., and obtain smooth and accurate acceleration estimation value Z1. The Kalman filter is based on state space model and measurement noise model to recursively predict and update the state of the target object's acceleration, ensuring the stability and robustness of the estimation data. The system uses the estimated acceleration value Z1 to update the related parameters in the object motion speed model in real time, including speed prediction value, acceleration state quantity, etc., to improve the description accuracy of the motion model to the dynamic behavior of the target object. After the model is updated, the system introduces an acceleration change constraint condition to limit the difference ΔZ between the current estimated acceleration Z1 and the acceleration Z0 at the previous moment. The constraint condition usually uses absolute value limitation or dynamic window filtering strategy to ensure that the acceleration change amplitude is within a reasonable range, avoid misjudgment or control abnormality caused by sudden disturbance or abnormal data, and improve the robustness and control stability of the system. Finally, combining the updated acceleration value, the system calculates the average motion rate V avg of the object in the target region. The average rate is calculated by comprehensively analyzing the acceleration and speed data of the target object in the continuous period, reflecting the overall motion trend of the target object. The system sets the frame rate of the visual acquisition module according to V avgThe sampling frame rate F of the visual acquisition module is set or dynamically adjusted to ensure that the image acquisition frequency matches the actual motion state of the target object, improving the timeliness and effectiveness of image acquisition. The processing flow effectively improves the accuracy and real-time performance of the motion state estimation of the object on the production line through recursive prediction and real-time feedback mechanism, ensuring the synchronization and response capability of the visual acquisition module in dynamic environment.

[0061] By using Kalman filtering method to dynamically estimate the acceleration of the target object, the embodiment significantly improves the perception accuracy and prediction ability of the system for the state change of the moving object on the production line. Compared with the traditional static speed model, Kalman filtering can effectively eliminate system noise and sensing error, improving the stability and reliability of motion parameter estimation. The introduction of acceleration change constraint condition further guarantees the suppression ability of the system to abnormal data, avoiding false triggering and false control problems. According to the average motion rate in the target area calculated from the acceleration value, the sampling frame rate of the visual acquisition module is adjusted, realizing the precise control of the sampling frequency, ensuring the image acquisition quality and the optimization of system resource use. This method enhances the adaptability and control accuracy of intelligent industrial robot system to high-speed dynamic production environment, effectively improves the automation efficiency and system stability.

[0062] In S1, according to the estimated acceleration value, the parameters in the object motion speed model are updated, including calculating the new acceleration value by weighting the acceleration value at the previous moment and the measured object motion speed at the current moment. A smoothing factor is used in the weighting calculation, and the value range of the smoothing factor is 70%-95%.

[0063] This embodiment further improves the dynamic updating method of the object motion speed model parameters in S1. Based on Kalman filtering, the system introduces a weighted smoothing calculation mechanism for updating the acceleration value of the target object to improve the stability and response sensitivity of state estimation. First, the system obtains the actual motion speed V1 of the target object measured at the current moment through Kalman filtering algorithm or other sensing data analysis method, and combines the acceleration estimation value Z0 at the previous moment. The system uses weighted calculation to integrate the difference between the previous moment acceleration Z0 and the measured motion speed V1 at the current moment to calculate the new acceleration value Z1. The weighted calculation formula is: Z1=α×Z0+(1-α)×ΔV / Δt;

[0064] Wherein, a is a smoothing factor, the value range is 70% to 95%, and is flexibly set according to application requirements and motion state complexity. Delta V is the difference between the current measured speed V1 and the previous speed V0, and delta t is the time interval. The introduction of the smoothing factor realizes the effective balance of the historical acceleration value and the current speed change. By increasing the value of a, the system can enhance the maintenance of the historical state and reduce the dramatic fluctuations caused by noise or abnormal data. When the value of a is reduced, the system enhances the sensitivity to the current state change and enhances the rapid response ability to dynamic environmental changes. The updated acceleration value Z1 is used to real-time correct the related parameters in the object motion speed model, including the predicted speed, position and other state information, and provides a high reliability data basis for the subsequent target area average rate calculation, visual acquisition module frame rate setting and other control links. Through the smoothing weighted update, the system ensures that it still has stable and accurate acceleration estimation ability in the noise environment, effectively improving the system's tracking ability of the high-speed dynamic change object motion state.

[0065] By introducing the weighted calculation and smoothing factor mechanism, the embodiment significantly improves the smoothness and robustness of the acceleration estimation value. Compared with the traditional method of directly using the current speed difference to calculate the acceleration, this method can effectively filter out the abnormal influence caused by high-frequency noise and data mutation, and avoid frequent adjustment of the sampling frame rate or control strategy due to short-time data abnormality. The flexible setting range of the smoothing factor (70%-95%) ensures that the system can realize the adaptive balance of sensitivity and stability according to the specific application scene, improving the overall response performance of the system. This optimization mechanism improves the control precision and stability of the intelligent industrial robot system in complex production environment, especially suitable for high-speed dynamic operation or application scenarios with high requirements for position and speed control precision, further enhancing the reliability of system automation control and data processing.

[0066] Although embodiments of the present application have been shown and described, it is to be understood that various modifications, substitutions, replacements and changes can be made to these embodiments without departing from the principles and spirit of the present application, and the scope of the present application is defined by the appended claims and their equivalents.

Claims

1. An adaptive visual recognition and correction method for an intelligent industrial robot, characterized in that: The method comprises: S1. Based on the real-time estimation results of the object movement speed on the production line, the frame rate of the visual acquisition module is adjusted; S2. Collecting image data of the object during its motion and generating initial positioning information, including selecting a reference object of known size in the image, measuring the pixel distance of the reference object in the image, measuring the pixel distance of the target object in the image, calculating the physical distance of the target object in real space, and using the calculated actual distance of the target object as the initial positioning information of the robot system; S3. Analyze the deviation between the initial positioning information and the delay parameter and generate a position correction instruction, including determining the basic processing time generated according to different image data volumes, analyzing the delay deviation of the initial positioning information, calculating the total delay value under the current image data, and generating a position correction instruction based on the total delay value under the current image data to ensure the operation accuracy of the robot arm; S4. Adjust the robot arm's motion according to the position correction instruction to complete the high-precision positioning operation, including determining the target's point of action and reaction force according to the position of the target object and the correction instruction, determining the force that the robot arm needs to apply, and controlling the robot arm to adjust according to the calculated results.

2. The adaptive visual recognition and correction method for an intelligent industrial robot according to claim 1, characterized in that: The S1 includes determining the rate of change of the moving speed of the target object on the production line, and calculating the image sampling frequency of the vision acquisition module as time progresses.

3. The adaptive visual recognition and correction method for an intelligent industrial robot according to claim 2, characterized in that: The specific steps of determining the rate of change of the moving speed of the target object on the production line in S1 include: detecting and recording the speed of the moving object on the production line at the current moment in real time through a visual recognition system or sensor; retrieving the object's moving speed at the previous moment from the system's historical data; confirming the time interval between the current moment and the previous moment; subtracting the speed value at the previous moment from the speed value at the current moment to obtain the speed change of the object during this period; and dividing the speed change by the time interval between the previous and next moments to obtain the speed change rate per unit time.

4. The adaptive visual recognition and correction method for an intelligent industrial robot according to claim 3, characterized in that: The specific steps of calculating the image sampling frequency of the visual acquisition module in S1 include determining the time elapsed from the start of sampling to the current moment based on the speed change rate per unit time, multiplying the speed growth rate by the current estimated time to obtain the sampling frequency of the visual acquisition module, using the calculated sampling frame rate as a control parameter, and adjusting the sampling speed of the visual acquisition module in real time to ensure that image acquisition is synchronized with the movement speed of the object.

5. The adaptive visual recognition and correction method for an intelligent industrial robot according to claim 1, characterized in that: The specific steps of calculating the physical distance of the target object in the real space in S2 include first, obtaining the actual distance value of the reference object in reality; then, calculating the ratio of the measured distance of the target object in the image to the measured distance of the reference object in the image; and finally, multiplying the actual distance value of the reference object by the above ratio to obtain the physical distance of the target object in the real space.

6. The adaptive visual recognition and correction method for an intelligent industrial robot according to claim 1, characterized in that: The specific steps of calculating the total delay value under the current image data in S3 include: evaluating the complexity of the current image based on the processed image data, and quantitatively analyzing the complexity of the current image based on the number of target objects, background complexity, and lighting conditions; determining the basic processing time under ideal conditions, and the basic processing time is the shortest time required to complete image acquisition, recognition and output positioning information under the simplest image data conditions; determining the delay increment caused by the increase in image complexity based on the numerical value of the image complexity, and the delay increment caused by the increase in image complexity is the product of the image complexity and the unit complexity delay growth; adding the basic processing time and the delay increment caused by the increase in image complexity to obtain the total delay value under the current image data; using the calculated total delay value as a delay compensation parameter, and adjusting the positioning information and motion control instructions according to the compensation parameter.

7. The adaptive visual recognition and correction method for an intelligent industrial robot according to claim 1, characterized in that: The specific steps of determining the force that the robotic arm needs to apply in S4 include: determining the reaction force generated by the target object on the robotic arm; measuring the actual distance between the target object from its force-receiving point to the support point; measuring the actual distance between the robotic arm from the position where the force is applied to its fixed support point; multiplying the reaction force generated by the target object by the action distance of the object to obtain the torque generated by the target object; dividing the torque of the target object by the action distance of the robotic arm to obtain the force that the robotic arm needs to apply; and adjusting the output of the robotic arm according to the force obtained by the above calculation.

8. The adaptive visual recognition and correction method for an intelligent industrial robot according to claim 1, characterized in that: The S1 further includes estimating the frame rate of the visual acquisition module, specifically, performing a ratio operation on the maximum motion speed of the object on the production line and the current measured actual motion speed of the object according to a preset proportional coefficient to calculate the frame rate; Then, sensors detect the initial position of objects on the production line. The measured object speed and current position information are then fed into the control module to calculate the frame rate adjustment. If the adjustment exceeds a preset threshold, the real-time frame rate control mechanism is triggered to dynamically adjust the frame rate of the visual acquisition module. If the adjustment does not exceed the threshold, the original frame rate is maintained.

9. The adaptive visual recognition and correction method for an intelligent industrial robot according to claim 1, characterized in that: The S1 also includes first using the Kalman filter method to estimate the acceleration of the object on the production line; then, based on the estimated acceleration value, updating the parameters in the object motion velocity model; then, introducing the acceleration change constraint condition to limit the difference between the updated acceleration value and the acceleration value at the previous moment; finally, combining the updated acceleration value, calculating the average motion rate in the target area, and setting the frame rate of the visual acquisition module according to the average rate.

10. The adaptive visual recognition and correction method for an intelligent industrial robot according to claim 9, characterized in that: In said S1, based on the estimated acceleration value, the parameters in the object motion velocity model are updated, including obtaining a new acceleration value by weighted calculation of the acceleration value at the previous moment and the currently measured object motion velocity, and a smoothing factor is used in the weighted calculation, and the value range of the smoothing factor is 70%-95%.

Images