Automatic calibration method and system for angle of pull rope of port crane

Abstract

The invention discloses an automatic calibration method and system for the angle of a pull rope of a port crane, and the method comprises the steps: taking a current pull rope reference value as a center, and symmetrically generating a plurality of target sampling points according to a fixed step length; after each sampling point arrives, waiting for a preset time to arrive at a stable point location, and then carrying out data acquisition; continuously collecting multi-frame angle data at each stable point location, removing high-frequency noise through a filtering algorithm, and calculating a statistic as final angle data of the point location; performing linear fitting on the final angle data of the plurality of target sampling points by adopting an RANSAC algorithm, and identifying and eliminating abnormal point data; and automatically calculating new slope and intercept parameters and writing the parameters into a configuration file. According to the port crane pull rope angle automatic calibration method and system provided by the invention, symmetric sampling is performed through a plurality of target points, so that the calibration precision is greatly improved.

Description

Technical Field

[0001] This application relates to the field of port crane hoisting, and in particular to a method and system for automatically calibrating the angle of the cable of a port crane. Background Technology

[0002] In container handling operations, rubber-tired gantry cranes (RTGs) in ports require precise control of the spreader's attitude to ensure the safe and efficient gripping and placement of containers. The key to spreader attitude control lies in accurately obtaining the spreader's tilt angle. A linear mapping relationship exists between the rope sensor and the spreader angle, and the accuracy of this relationship directly affects the precision of attitude control. In existing technologies, the mapping relationship between the rope sensor and the spreader angle is typically established through manual measurement and calibration. However, this manual calibration process requires on-site operation by staff and usually takes 2-3 hours to complete, severely impacting the operational efficiency and maintenance costs of port equipment. In the port environment, equipment maintenance time windows are precious, and prolonged calibration operations can lead to equipment downtime and losses.

[0003] Therefore, it is necessary to provide an automatic calibration method and system for the rope angle of port cranes to solve the above problems. Summary of the Invention

[0004] This application provides a method and system for automatic calibration of the rope angle of a port crane, which greatly improves the calibration accuracy by symmetrically sampling multiple target points.

[0005] In a first aspect, this application provides an automatic calibration method for the cable angle of a port crane, the method comprising: Centered on the current rope reference value, multiple target sampling points are generated symmetrically with a fixed step size; After each sampling point is reached, wait for a preset time to reach a stable position before data collection. Multiple frames of angle data are continuously collected at each stable point. After removing high-frequency noise through a filtering algorithm, the statistics are calculated as the final angle data for that point. The RANSAC algorithm is used to perform linear fitting on the final angle data of the multiple target sampling points to identify and remove outlier data. Automatically calculate new slope and intercept parameters and write them to the configuration file.

[0006] Preferably, the method further includes: acquiring operating condition information and calibrating the operating conditions with and without containers respectively; for the operating condition with containers, confirming that the spreader has grabbed the first layer of containers and hovered at a predetermined height; for the operating condition without containers, confirming that the spreader is in an unloaded state and hovering at the same height.

[0007] Preferably, there are 7 target sampling points.

[0008] Preferably, the step of linearly fitting the final angle data of the multiple target sampling points using the RANSAC algorithm specifically includes: Randomly select 2 sets of data from 7 sets of data, and calculate the equation of the straight line passing through these two points: θ = k × Rope + b; Substitute the remaining 5 sets of data into the linear equation and calculate the fitting error for each set of data. The fitting error is the absolute value of the difference between the actual angle value and the predicted angle value. The number of data points with a statistical error less than a preset threshold ε is denoted as the number of internal points. Repeat the above steps M times, recording the number of interior points and the corresponding line parameters for each iteration; Select the iteration result with the most interior points, and use all interior points from this iteration to perform least squares fitting again to obtain the final slope k_new and intercept b_new.

[0009] Preferably, M is greater than or equal to 100 and less than or equal to 500.

[0010] Secondly, this application also provides an automatic calibration system for the cable angle of a port crane, the system comprising: The target sampling point generation module is used to generate multiple target sampling points symmetrically with the current rope reference value as the center and a fixed step size. The data acquisition module is used to wait for a preset time after each sampling point is reached to reach a stable position before collecting data. The final angle data calculation module is used to continuously collect multiple frames of angle data at each stable point, and calculate the statistics as the final angle data of that point after removing high-frequency noise through a filtering algorithm. The linear fitting module is used to perform linear fitting on the final angle data of the multiple target sampling points using the RANSAC algorithm, and to identify and remove outlier data. The parameter auto-update module is used to automatically calculate new slope and intercept parameters and write them to the configuration file.

[0011] Preferably, the method further includes: acquiring operating condition information and calibrating the operating conditions with and without containers respectively; for the operating condition with containers, confirming that the spreader has grabbed the first layer of containers and hovered at a predetermined height; for the operating condition without containers, confirming that the spreader is in an unloaded state and hovering at the same height.

[0012] Preferably, there are 7 target sampling points.

[0013] Preferably, the step of linearly fitting the final angle data of the multiple target sampling points using the RANSAC algorithm specifically includes: Randomly select 2 sets of data from 7 sets of data, and calculate the equation of the straight line passing through these two points: θ = k × Rope + b; Substitute the remaining 5 sets of data into the linear equation and calculate the fitting error for each set of data. The fitting error is the absolute value of the difference between the actual angle value and the predicted angle value. The number of data points with a statistical error less than a preset threshold ε is denoted as the number of internal points. Repeat the above steps M times, recording the number of interior points and the corresponding line parameters for each iteration; Select the iteration result with the most interior points, and use all interior points from this iteration to perform least squares fitting again to obtain the final slope k_new and intercept b_new.

[0014] Preferably, M is greater than or equal to 100 and less than or equal to 500.

[0015] This application offers the following advantages over existing technologies: It provides an automatic calibration method and system for the angle of a port crane's guy rope. The method includes: symmetrically generating multiple target sampling points centered on the current guy rope reference value using a fixed step size; waiting a preset time after each sampling point is reached to reach a stable point before data acquisition; continuously acquiring multiple frames of angle data at each stable point, removing high-frequency noise using a filtering algorithm, and calculating statistics as the final angle data for that point; using the RANSAC algorithm to linearly fit the final angle data of the multiple target sampling points, identifying and removing outlier data; automatically calculating new slope and intercept parameters and writing them into a configuration file; and significantly improving calibration accuracy through symmetrical sampling of multiple target points. Furthermore, the RANSAC algorithm is used to linearly fit the final angle data of the multiple target sampling points, and it can automatically identify and remove outlier data points, ensuring that the calibration results are not affected by a single outlier. Even in the harsh environment of the port, the calibration success rate can still be maintained above 98%, and the reliability is improved by about 50%. Furthermore, by calibrating separately for the cases with and without the case, and maintaining independent mapping parameters for different load states, the attitude control accuracy under each case is ensured. Compared with single-case calibration, the average accuracy under each case is improved by about 25%. Attached Figure Description

[0016] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0017] Figure 1 This is a flowchart illustrating an automatic calibration method for the cable angle of a port crane according to an embodiment of this application. Figure 2This is a schematic diagram of the structure of an automatic calibration system for the rope angle of a port crane in an embodiment of this application.

[0018] The accompanying drawings have illustrated specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to specific embodiments. Detailed Implementation

[0019] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0020] To address the aforementioned issues, the embodiments provided in this application offer an automatic calibration method and system for the rope angle of a port crane, which significantly improves calibration accuracy through symmetrical sampling of multiple target points.

[0021] Figure 1 This is a flowchart illustrating an automatic calibration method for the cable angle of a port crane according to an embodiment of this application. Now refer to... Figure 1 This invention provides an automatic calibration method for the pull rope angle of a port crane, the method comprising: Step S101: Using the current rope reference value as the center, generate multiple target sampling points symmetrically according to a fixed step size; Step S102: After each sampling point is reached, wait for a preset time to reach a stable position before collecting data; Step S103: Collect multiple frames of angle data continuously at each stable point, remove high-frequency noise using a filtering algorithm, and calculate the statistics as the final angle data for that point. Step S014: Use the RANSAC (Random Sample Consensus) algorithm to perform linear fitting on the final angle data of the multiple target sampling points, and identify and remove outlier data. Step S105: Automatically calculate the new slope and intercept parameters and write them to the configuration file.

[0022] Specifically, based on the preset offset step size Step, seven target rope positions are generated, as follows: Target_positions = [Base - 3×Step, Base - 2×Step, Base - 1×Step,Base, Base + 1×Step, Base + 2×Step, Base + 3×Step] Where Base is the baseline value, which is the real-time value of the current draw rope sensor. The offset step size Step is determined based on the range and resolution of the draw rope sensor, with an optimal range of 5%-10% of the total draw rope range.

[0023] The PLC-controlled drive unit of the rope-pulling mechanism adjusts the rope length to ensure the real-time reading of the rope-pulling sensor reaches the target position value. The drive process employs closed-loop control, continuously monitoring the deviation between the rope-pulling value and the target value to ensure accurate positioning.

[0024] After the rope-pulling mechanism reaches the target position, the PLC waits for a preset time, which can be dynamically adjusted according to the wind speed and the weight of the spreader in the actual port environment. After the preset time has elapsed, the PLC sends a signal to the IPC via the communication interface, notifying the IPC that the current position has stabilized and data acquisition can begin.

[0025] After receiving the signal, the IPC continuously collects angle data for a predetermined duration. Assuming the angle sensor's sampling frequency is fHz, a total of N = f × duration frames of angle data will be collected within this duration, denoted as... The preferred duration is 5 to 20 seconds.

[0026] In step S103, the IPC preprocesses the acquired multi-frame angle data: Low-pass filters are used to remove high-frequency noise, such as vibrations caused by minor wind gusts or electrical interference. Remove outlier frames that deviate significantly from the sequence median, such as data points that deviate from the sequence median by more than three standard deviations. Calculate the statistics of the effective data after filtering, preferably the arithmetic mean or median, as the final angle data θ_final for that point.

[0027] In practice, this also includes: obtaining working condition information and calibrating the working conditions with and without containers respectively; for the working conditions with containers, confirming that the spreader has grabbed the first layer of containers and hovered at the predetermined height; for the working conditions without containers, confirming that the spreader is in an unloaded state and hovering at the same height.

[0028] Specifically, for cases involving containers, the predetermined height can be, for example, 3.5 meters.

[0029] In practice, there are seven target sampling points.

[0030] In specific implementation, the linear fitting of the final angle data of the multiple target sampling points using the RANSAC algorithm specifically includes: Randomly select 2 sets of data from 7 sets of data, and calculate the equation of the straight line passing through these two points: θ = k × Rope + b; Substitute the remaining 5 sets of data into the linear equation and calculate the fitting error for each set of data. The fitting error is the absolute value of the difference between the actual angle value and the predicted angle value. The number of data points with a statistical error less than a preset threshold ε is denoted as the number of internal points. Repeat the above steps M times, recording the number of interior points and the corresponding line parameters for each iteration; Select the iteration result with the most interior points, and use all interior points from this iteration to perform least squares fitting again to obtain the final slope k_new and intercept b_new.

[0031] In specific implementation, M is greater than or equal to 100 and less than or equal to 500.

[0032] Specifically, the rationality of the calculated slope k_new and intercept b_new is verified: Check if the slope k_new is within a reasonable range, for example, the range of slopes estimated based on the physical model; Check if the goodness of fit meets the minimum requirements, for example, whether the R² value meets the minimum requirements, preferably R² > 0.95; If the verification passes, the slope k_new and intercept b_new are written to the configuration file for the corresponding working condition, and the rope-angle mapping relationship is updated; If verification fails, a calibration error will be indicated, and it will be recommended to re-execute the calibration process.

[0033] Figure 2 This is a schematic diagram of an automatic calibration system for the cable angle of a port crane, as described in an embodiment of this application. Now refer to... Figure 2 This invention also provides an automatic calibration system for the cable angle of a port crane, the system comprising: The target sampling point generation module 21 is used to generate multiple target sampling points symmetrically with the current rope reference value as the center and according to a fixed step size. The data acquisition module 22 is used to wait for a preset time after each sampling point is reached to reach a stable position before collecting data. The final angle data calculation module 23 is used to continuously collect multiple frames of angle data at each stable point, and calculate the statistics as the final angle data of that point after removing high-frequency noise through a filtering algorithm. The linear fitting module 24 is used to perform linear fitting on the final angle data of the multiple target sampling points using the RANSAC algorithm, and to identify and remove outlier data. The parameter automatic update module 25 is used to automatically calculate new slope and intercept parameters and write them to the configuration file.

[0034] In practice, this also includes: obtaining working condition information and calibrating the working conditions with and without containers respectively; for the working conditions with containers, confirming that the spreader has grabbed the first layer of containers and hovered at the predetermined height; for the working conditions without containers, confirming that the spreader is in an unloaded state and hovering at the same height.

[0035] In practice, there are seven target sampling points.

[0036] In specific implementation, the linear fitting of the final angle data of the multiple target sampling points using the RANSAC algorithm specifically includes: Randomly select 2 sets of data from 7 sets of data, and calculate the equation of the straight line passing through these two points: θ = k × Rope + b; Substitute the remaining 5 sets of data into the linear equation and calculate the fitting error for each set of data. The fitting error is the absolute value of the difference between the actual angle value and the predicted angle value. The number of data points with a statistical error less than a preset threshold ε is denoted as the number of internal points. Repeat the above steps M times, recording the number of interior points and the corresponding line parameters for each iteration; Select the iteration result with the most interior points, and use all interior points from this iteration to perform least squares fitting again to obtain the final slope k_new and intercept b_new.

[0037] In specific implementation, M is greater than or equal to 100 and less than or equal to 500.

[0038] The following is a detailed explanation of the rope angle calibration for RTG equipment at a port under non-container conditions.

[0039] Step 1: Initialization and Operating Condition Selection By selecting the "No Box" calibration mode through the human-machine interface, the system confirms that the spreader is unloaded and hovers at the predetermined height of 3.5 meters. At this time, the real-time reading of the rope sensor is 2500 (the sensor range is 0-5000, and the unit is the internal count value of the sensor). The system reads this value as the base value Base = 2500.

[0040] Step 2: Generate a 7-point symmetric sampling sequence Based on the preset offset step size Step = 300, seven target rope positions are generated: Target_positions = [1600, 1900, 2200, 2500, 2800, 3100, 3400] The sampling sequence covers the range of pull rope values ​​from 1600 to 3400, which is about 36% of the sensor's range. This ensures sufficient coverage while avoiding exceeding the physical limits of the pull rope mechanism.

[0041] Step 3: Drive to the target location sequentially and collect data. For the first target location Rope = 1600, perform the following operations: a) The PLC-controlled drive unit of the rope pulling mechanism adjusts the rope length, causing the rope sensor reading to gradually decrease from 2500 to 1600, with a positioning accuracy of ±5. b) After positioning, the PLC delays for 60 seconds to ensure that the sway of the lifting device is completely attenuated; c) After the waiting period ends, the PLC sends a Move_Done signal to the IPC; d) After receiving the signal, the IPC continuously collects angle data for 10 seconds at a frequency of 10 Hz, obtaining a total of 100 frames of angle values; e) The IPC applies a low-pass filter (cutoff frequency 1 Hz) to 100 frames of data to remove high-frequency noise, and calculates the arithmetic mean of the filtered data. ; f) The IPC sends a Data_Ack signal to the PLC to confirm that the data acquisition at this point is complete.

[0042] Repeat the above steps to collect data at the remaining 6 target locations, ultimately obtaining 7 sets of rope-angle data pairs: (1600, -2.35°), (1900, -1.42°), (2200, -0.51°), (2500, 0.38°), (2800, 1.29°), (3100, 2.18°), (3400, 3.12°) Step 4: RANSAC Robust Linear Fitting IPC applied the RANSAC algorithm to perform linear fitting on 7 sets of data, setting the number of iterations M = 200 and the error threshold ε = 0.2°. The algorithm execution process is as follows: a) First iteration: Randomly select points (1600, -2.35°) and (3400, 3.12°), calculate the slope of the line k = (3.12 - (-2.35)) / (3400 - 1600) ≈ 0.00304, and the intercept b = -2.35 - 0.00304 × 1600 ≈ -7.21. Substitute the remaining 5 points into the line and calculate the fitting error. It is found that the error of all 5 points is less than 0.2°, and the number of inliers is 7. b) Iterations 2 through 200: Repeated random sampling and interior point statistics; c) Statistical analysis of 200 iterations revealed that 180 iterations had 7 interior points, while the remaining 20 iterations had 6 or 5 interior points due to random sampling of suboptimal pairs. d) Select any iteration with 7 interior points, such as the first iteration, and perform least squares fitting using all 7 points to obtain the final parameters: k_new = 0.00305, b_new = -7.23.

[0043] Step 5: Parameter Verification and Update Verify the calculated parameters: a) The slope k_new = 0.00305 is within a reasonable range (estimated range is 0.002 to 0.005); b) The goodness of fit R² = 0.9996, which meets the minimum requirement (R² > 0.95); c) If the verification is successful, the system will write k_new = 0.00305 and b_new = -7.23 to the configuration file for the non-boxed operating condition and update the mapping relationship.

[0044] Example 2: Robustness verification of the RANSAC algorithm under outlier data The simulation of strong wind interference during the data acquisition process caused abnormal deviations in the angle data of a certain sampling point, thus verifying the anti-interference capability of the RANSAC algorithm.

[0045] Assuming that, based on Example 1, the actual angle value collected at the 5th sampling point (2800, 1.29°) is affected by a sudden strong wind, the angle value actually collected is... The data for the remaining 6 points are normal. At this point, the 7 sets of data become: (1600, -2.35°), (1900, -1.42°), (2200, -0.51°), (2500, 0.38°), (2800, 3.85°), (3100, 2.18°), (3400, 3.12°) If the traditional least squares method is used for direct fitting, the calculated slope k' = 0.00380, intercept b' = -8.45, and goodness of fit R² = 0.912 are obtained. This result is significantly affected by outliers, resulting in a decrease in accuracy.

[0046] The RANSAC algorithm was used for fitting. a) In 200 iterations, if an outlier (2800, 3.85°) is selected during random sampling, the number of interior points in that iteration is usually 2-4 (because the line fitted based on the outlier deviates significantly from other normal points); b) If no outliers are selected during random sampling, the number of interior points in this iteration is 6 (the 6 normal points match each other); c) Analyze the results of 200 iterations, select the iteration with the most interior points (6), and refit using these 6 normal points to obtain k_new = 0.00305, b_new = -7.23, R² = 0.9998; d) The system identified the anomaly (2800, 3.85°) and marked it in the log. The final calibration result was almost identical to that of Example 1 and was not affected by the anomaly.

[0047] The RANSAC algorithm can accurately identify and remove outliers even when there is one outlier among the seven points, ensuring the reliability of the calibration results.

[0048] Example 3: Independent Calibration under Boxed Operating Conditions The calibration of the RTG under the box condition is explained, and the effectiveness of the dual-condition adaptive strategy is verified.

[0049] By selecting the "with container" calibration mode through the interface, the system confirms that the spreader has grabbed a standard 20-foot container (weighing approximately 15 tons) and is hovering at a height of 3.5 meters. At this time, the pull rope sensor reading is Base = 2650 (the base value is slightly higher than that in the unloaded condition due to the increased load).

[0050] The calibration was performed following the same procedure as in Example 1, generating 7 target locations: Target_positions = [1750, 2050, 2350, 2650, 2950, ​​3250, 3550] After data collection was completed, 7 sets of data were obtained: (1750, -2.18°), (2050, -1.31°), (2350, -0.45°), (2650, 0.42°), (2950, ​​1.28°), (3250, 2.15°), (3550, 3.05°) The RANSAC algorithm was used for fitting to obtain the mapping parameters for the belt box condition: k_belt box = 0.00291, b_belt box = -7.30, R² = 0.9994.

[0051] Comparing the parameters of Example 1 without the casing (k_without casing = 0.00305, b_without casing = -7.23), a significant difference was found: a) Slope difference: (0.00305 - 0.00291) / 0.00305 ≈ 4.6%, indicating that the load change did indeed affect the linear relationship between the rope and the angle; b) Intercept difference: |-7.23 - (-7.30)| = 0.07°, the difference is relatively small.

[0052] The parameters k_with container and b_with container are written into a separate configuration file for the container-carrying condition, and managed separately from the parameters for the non-container-carrying condition. In actual operation, the parameters for the corresponding condition are automatically loaded based on whether the spreader has grabbed the container, ensuring the attitude control accuracy under each condition.

[0053] The necessity and effectiveness of the dual-condition adaptive strategy were verified. By maintaining independent parameters for different load states, the system can adapt to various operating scenarios and improve overall control performance.

[0054] In summary, the present application provides an automatic calibration method and system for the angle of a port crane's guy rope. The method includes: symmetrically generating multiple target sampling points centered on the current guy rope reference value according to a fixed step size; waiting a preset time after each sampling point is reached to reach a stable point before data acquisition; continuously acquiring multiple frames of angle data at each stable point, removing high-frequency noise using a filtering algorithm, and calculating statistics as the final angle data for that point; using the RANSAC algorithm to linearly fit the final angle data of the multiple target sampling points, identifying and removing outlier data; automatically calculating new slope and intercept parameters and writing them into a configuration file; and significantly improving calibration accuracy through symmetrical sampling of multiple target points. Furthermore, the RANSAC algorithm is used to linearly fit the final angle data of the multiple target sampling points, and it can automatically identify and remove outlier data points, ensuring that the calibration results are not affected by a single outlier. Even in the harsh environment of the port, the calibration success rate can still be maintained above 98%, and the reliability is improved by about 50%. Furthermore, by calibrating separately for the cases with and without the case, and maintaining independent mapping parameters for different load states, the attitude control accuracy under each case is ensured. Compared with single-case calibration, the average accuracy under each case is improved by about 25%.

[0055] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the claims.

[0056] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A method for automatically calibrating the angle of a port crane's guy rope, characterized in that, The method includes: Centered on the current rope reference value, multiple target sampling points are generated symmetrically with a fixed step size; After each sampling point is reached, wait for a preset time to reach a stable position before data collection. Multiple frames of angle data are continuously collected at each stable point. After removing high-frequency noise through a filtering algorithm, the statistics are calculated as the final angle data for that point. The RANSAC algorithm is used to perform linear fitting on the final angle data of the multiple target sampling points to identify and remove outlier data. Automatically calculate new slope and intercept parameters and write them to the configuration file.

2. The automatic calibration method for the angle of the cable of a port crane according to claim 1, characterized in that, It also includes: acquiring operating condition information and calibrating for operating conditions with and without containers; for operating conditions with containers, confirming that the spreader has grabbed the first layer of containers and hovered at the predetermined height; for operating conditions without containers, confirming that the spreader is in an unloaded state and hovered at the same height.

3. The automatic calibration method for the angle of the cable of a port crane according to claim 1, characterized in that, There are 7 target sampling points.

4. The automatic calibration method for the angle of the cable of a port crane according to claim 3, characterized in that, The step of using the RANSAC algorithm to perform linear fitting on the final angle data of the multiple target sampling points specifically includes: Randomly select 2 sets of data from 7 sets of data, and calculate the equation of the straight line passing through these two points: θ = k × Rope + b; Substitute the remaining 5 sets of data into the linear equation and calculate the fitting error for each set of data. The fitting error is the absolute value of the difference between the actual angle value and the predicted angle value. The number of data points with a statistical error less than a preset threshold ε is denoted as the number of internal points. Repeat the above steps M times, recording the number of interior points and the corresponding line parameters for each iteration; Select the iteration result with the most interior points, and use all interior points from this iteration to perform least squares fitting again to obtain the final slope k_new and intercept b_new.

5. The automatic calibration method for the angle of the cable of a port crane according to claim 4, characterized in that, M is greater than or equal to 100 and less than or equal to 500.

6. An automatic calibration system for the angle of a port crane's guy rope, characterized in that, The system includes: The target sampling point generation module is used to generate multiple target sampling points symmetrically with the current rope reference value as the center and a fixed step size. The data acquisition module is used to wait for a preset time after each sampling point is reached to reach a stable position before collecting data. The final angle data calculation module is used to continuously collect multiple frames of angle data at each stable point, and calculate the statistics as the final angle data of that point after removing high-frequency noise through a filtering algorithm. The linear fitting module is used to perform linear fitting on the final angle data of the multiple target sampling points using the RANSAC algorithm, and to identify and remove outlier data. The parameter auto-update module is used to automatically calculate new slope and intercept parameters and write them to the configuration file.

7. The automatic calibration system for the cable angle of a port crane according to claim 6, characterized in that, It also includes: acquiring operating condition information and calibrating for operating conditions with and without containers; for operating conditions with containers, confirming that the spreader has grabbed the first layer of containers and hovered at the predetermined height; for operating conditions without containers, confirming that the spreader is in an unloaded state and hovered at the same height.

8. The automatic calibration system for the cable angle of a port crane according to claim 6, characterized in that, There are 7 target sampling points.

9. The automatic calibration system for the cable angle of a port crane according to claim 8, characterized in that, The step of using the RANSAC algorithm to perform linear fitting on the final angle data of the multiple target sampling points specifically includes: Randomly select 2 sets of data from 7 sets of data, and calculate the equation of the straight line passing through these two points: θ = k × Rope + b; Substitute the remaining 5 sets of data into the linear equation and calculate the fitting error for each set of data. The fitting error is the absolute value of the difference between the actual angle value and the predicted angle value. The number of data points with a statistical error less than a preset threshold ε is denoted as the number of internal points. Repeat the above steps M times, recording the number of interior points and the corresponding line parameters for each iteration; Select the iteration result with the most interior points, and use all interior points from this iteration to perform least squares fitting again to obtain the final slope k_new and intercept b_new.

10. The automatic calibration system for the cable angle of a port crane according to claim 9, characterized in that, M is greater than or equal to 100 and less than or equal to 500.

Images