Crown block positioning control method and related equipment

Abstract

The invention discloses a crown block positioning control method and related equipment, and relates to the technical field of industrial automation control, and the method comprises the steps: obtaining multi-source sensing data of a target crown block; carrying out fusion processing on the ultra-wideband positioning signal, the encoder displacement increment and the inertial measurement unit data based on a robust Kalman filtering algorithm to obtain a first positioning result; performing point cloud registration processing on the laser radar scanning data to obtain a second positioning result; determining a pose offset compensation amount through a preset kinetic model; and controlling the target crown block according to the first positioning result, the second positioning result and the pose offset compensation amount. Through the multi-sensor fusion and dynamic compensation technology, high-precision and anti-interference crown block positioning and control suitable for complex working conditions can be achieved.

Description

Technical Field

[0001] The present application relates to the field of industrial automation control technology, and more specifically, to an overhead crane positioning control method and related equipment. Background Art

[0002] With the rapid development of modern industrial logistics and heavy equipment manufacturing, bridge cranes (overhead cranes), as core equipment for material handling, are responsible for high-precision, high-load lifting tasks in ports, metallurgical workshops, large warehouses, and other scenarios. The positioning accuracy of overhead cranes directly affects material docking efficiency, equipment collision safety, and production cycle stability. Real-time, precise positioning and control of overhead cranes within three dimensions has become a key industry requirement, especially in complex electromagnetic environments, with multiple obstacles and dynamic load fluctuations.

[0003] In related technologies, overhead crane positioning often relies on a single sensing technology, such as encoder-based displacement feedback, ultra-wideband (UWB) base station positioning, or dead reckoning using an inertial measurement unit (IMU). However, encoders are susceptible to slippage and wear, leading to cumulative errors; UWB signals are susceptible to multipath interference in areas with dense metal; and inertial measurement units struggle to maintain accuracy over the long term due to their drift characteristics. The limitations of a single data source make positioning results prone to deviations or even failure. Consequently, related technologies suffer from the technical problem of low overhead crane positioning accuracy. Summary of the Invention

[0004] The Summary of the Invention section of this application introduces a series of simplified concepts that will be further described in detail in the Detailed Description of the Invention section. The Summary of the Invention section of this application is not intended to limit the key features and essential technical features of the claimed technical solution, nor is it intended to determine the scope of protection of the claimed technical solution.

[0005] The overhead crane positioning control method and related equipment provided in this application can achieve high-precision, interference-resistant, and adaptable overhead crane positioning and control under complex working conditions through multi-sensor fusion and dynamic compensation technology.

[0006] In the first aspect, the present application provides a method for positioning and controlling an overhead crane, comprising: obtaining multi-source sensor data of a target overhead crane, wherein the multi-source sensor data includes an ultra-wideband positioning signal, an encoder displacement increment, an inertial measurement unit data, and a lidar scanning data; fusing the ultra-wideband positioning signal, the encoder displacement increment, and the inertial measurement unit data based on an anti-error Kalman filtering algorithm to obtain a first positioning result; performing point cloud registration processing on the lidar scanning data to obtain a second positioning result; determining a posture offset compensation amount through a preset dynamic model, wherein the preset dynamic model is constructed based on the hook swing amplitude and load mass of the target overhead crane; and controlling the target overhead crane according to the first positioning result, the second positioning result, and the posture offset compensation amount.

[0007] In some embodiments, the ultra-wideband positioning signal, the encoder displacement increment and the inertial measurement unit data are fused based on the robust Kalman filter algorithm to obtain a first positioning result, including: detecting and eliminating abnormal measurement values ​​in the ultra-wideband positioning signal through the Mahalanobis distance threshold method to obtain a corrected ultra-wideband positioning signal; determining a state estimation covariance matrix based on the distribution characteristics of the abnormal measurement values; performing displacement integration on the encoder displacement increment according to the overhead crane kinematic model to generate an overhead crane relative displacement prediction value; inputting the angular velocity component and the acceleration component in the inertial measurement unit data into a pre-calibrated attitude solution module to obtain the overhead crane body attitude angle; performing multi-source data weighted fusion on the relative displacement prediction value, the overhead crane body attitude angle and the corrected ultra-wideband positioning signal based on the state estimation covariance matrix to obtain a state prediction vector; and performing robust Kalman filter update processing on the state prediction vector to obtain the first positioning result.

[0008] In some embodiments, the point cloud registration processing of the lidar scanning data to obtain a second positioning result includes: extracting geometric distribution characteristics of preset two-dimensional code markers in the lidar scanning data, wherein the two-dimensional code markers are continuously distributed along the track at a preset interval, and the geometric characteristics of each of the two-dimensional code markers are pre-associated and stored with the two-dimensional position coordinates; determining the absolute coordinate offset of the target overhead crane based on the registration and matching of the geometric distribution characteristics with a preset marker database; and correcting the cumulative error of the first positioning result according to the absolute coordinate offset to generate the second positioning result.

[0009] In some embodiments, the determination of the posture offset compensation amount through a preset dynamic model includes: based on the swing amplitude of the hook, calculating the product of the square of the swing amplitude and a preset swing compensation coefficient to generate a dynamic swing compensation amount; inputting the load mass data into a nonlinear inertia compensation model to output a load inertia compensation parameter; and determining the posture offset compensation amount based on a weighted fusion result of the dynamic swing compensation amount and the load inertia compensation parameter.

[0010] In some embodiments, inputting the load mass data into a nonlinear inertia compensation model and outputting load inertia compensation parameters include: detecting whether the load mass exceeds a preset mass threshold and generating a load mode judgment signal; when the load mode judgment signal is a light load mode, generating an initial inertia compensation parameter based on the product of the load mass and a preset linear compensation coefficient; when the load mode judgment signal is a heavy load mode, generating the initial inertia compensation parameter based on the product of the square value of the load mass and a preset nonlinear compensation coefficient; performing directional coupling calculation on the initial inertia compensation parameter according to the overhead crane operation acceleration direction vector, and outputting the load inertia compensation parameter.

[0011] In some embodiments, the target overhead crane is controlled based on the first positioning result, the second positioning result and the posture offset compensation, including: performing confidence assessment on the first positioning result and the second positioning result to generate fused positioning coordinates; performing vector superposition on the fused positioning coordinates based on the posture offset compensation to generate compensated target coordinates; converting the compensated target coordinates into control parameters in the overhead crane track coordinate system to generate travel speed and direction adjustment instructions; based on the travel speed and direction adjustment instructions, adjusting the proportional integral coefficient of the PID controller according to the real-time load swing spectrum characteristics of the target overhead crane, and performing overhead crane motion trajectory control on the target overhead crane.

[0012] In some embodiments, the confidence assessment of the first positioning result and the second positioning result to generate fused positioning coordinates includes: obtaining current environmental noise parameters, wherein the environmental noise parameters include dust concentration, electromagnetic interference intensity, and laser reflectivity attenuation coefficient; determining a first confidence weight of the ultra-wideband positioning signal based on the dust concentration and the electromagnetic interference intensity; determining a second confidence weight of the lidar scanning data based on the product of the laser reflectivity attenuation coefficient and a preset QR code matching success rate; and using an adaptive weighted fusion algorithm to perform weighted averaging on the first positioning result and the second positioning result based on the first confidence weight and the second confidence weight to obtain the fused positioning coordinates.

[0013] In the second aspect, the present application also provides an overhead crane positioning control device, comprising: a data acquisition unit for acquiring multi-source sensor data of a target overhead crane, wherein the multi-source sensor data includes an ultra-wideband positioning signal, an encoder displacement increment, an inertial measurement unit data and a lidar scanning data; a first positioning unit for fusing the ultra-wideband positioning signal, the encoder displacement increment and the inertial measurement unit data based on an anti-error Kalman filtering algorithm to obtain a first positioning result; a second positioning unit for performing point cloud registration processing on the lidar scanning data to obtain a second positioning result; an offset compensation unit for determining a posture offset compensation amount through a preset dynamic model, wherein the preset dynamic model is constructed based on the hook swing amplitude and load mass of the target overhead crane; an overhead crane control unit for controlling the target overhead crane according to the first positioning result, the second positioning result and the posture offset compensation amount.

[0014] In a third aspect, the present application further provides an electronic device comprising: a memory and a processor, wherein the processor is configured to implement the steps of the overhead crane positioning control method described in the first aspect when executing a computer program stored in the memory.

[0015] In a fourth aspect, the present application further provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the overhead crane positioning control method described in the first aspect.

[0016] In a fifth aspect, the present application also provides a computer program product, including a computer program or computer executable instructions, which, when executed by a processor, implements the overhead crane positioning control method provided in an embodiment of the present application.

[0017] In summary, the sensors of the present application are redundant with each other and have complementary advantages, which can effectively make up for the defect that a single sensor is susceptible to environmental interference, and through the anti-error Kalman filtering algorithm, the ultra-wideband positioning signal, the encoder displacement increment and the inertial measurement unit data are fused, which can automatically eliminate outliers and significantly reduce error propagation. Point cloud matching is used for the absolute position correction of the lidar, which can eliminate the cumulative error caused by long-term operation; accurately calculating the posture offset compensation caused by the swing can effectively compensate for the actual position deviation in dynamic operation, and can improve the accuracy and stability under working conditions such as variable load and shaking; the first positioning result (fusion estimate), the second positioning result (laser correction value) and the posture compensation are comprehensively calculated to achieve more accurate control command output. In summary, the overhead crane positioning control method provided by the present application can achieve high-precision, anti-interference, and adaptable to complex working conditions overhead crane positioning and control through multi-sensor fusion and dynamic compensation technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present description. The same reference symbols are used throughout the drawings to represent the same components. In the drawings:

[0019] Figure 1 A flowchart of a method for controlling positioning of an overhead crane provided in an embodiment of the present application;

[0020] Figure 2 A schematic diagram of the structure of an overhead crane positioning control device provided in an embodiment of the present application;

[0021] Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0022] Terms in the specification, claims, and drawings of this application, such as "first," "second," "third," "fourth," and the like (if any), are used to distinguish between similar objects, rather than to describe a particular order or precedence. Therefore, it is understood that these terms can be used interchangeably where appropriate, so that the embodiments described can be implemented in a different order, unless otherwise specified in the drawings or descriptions. In addition, the terms "is" and "has" and any variations thereof in this application are intended to cover all possible constituent elements on a non-exclusive basis. For example, a process, method, system, product, or apparatus that includes several steps or units is not necessarily limited to the steps or units that are explicitly listed, but may also include other steps or units that are not explicitly listed, or steps or units that are inherent to the process, method, product, or apparatus.

[0023] In this application, a "module" or "unit" refers to a computer program or part of a computer program that has a specific function and works in conjunction with other related parts to achieve a predetermined goal. These modules or units can be implemented by software, hardware (such as processing circuits or memories), or a combination of the two. One or more processors or memories can implement one or more modules or units. At the same time, each module or unit can also be part of a larger module or unit.

[0024] The technical solutions in this application will be described in detail below in conjunction with the accompanying drawings in the embodiments. It should be noted that the embodiments described are only part of this application, not all embodiments. In the following description, the "some embodiments" mentioned are only a subset of all possible embodiments, which may be the same or different subsets, and different embodiments can be combined with each other without conflict.

[0025] Figure 1This is a flow chart of a method for controlling the positioning of an overhead crane provided in an embodiment of the present application. Figure 1 The method for controlling the positioning of an overhead traveling crane provided in the embodiment of the present application may include the following steps 101 to 105:

[0026] Step 101: Acquire multi-source sensor data of a target overhead crane, wherein the multi-source sensor data may include ultra-wideband positioning signals, encoder displacement increments, inertial measurement unit data, and lidar scanning data;

[0027] In some examples, the target overhead crane is a specific traveling device that is currently being located and controlled, and can be a bridge, gantry or suspended crane system in scenarios such as factory workshops, warehousing and logistics, and steel metallurgy. Multi-source sensing data refers to a data set derived from multiple different types of sensors, which is used to fully perceive the spatial position, posture, speed, motion trend and other status information of the target overhead crane. It can include ultra-wideband positioning signals, encoder displacement increments, inertial measurement unit data and lidar scanning data. Ultra-wideband is a low-power, high-time-resolution wireless positioning technology. The ultra-wideband positioning signal is transmitted by an ultra-wideband base station group deployed on the top of the workshop. The initial two-dimensional coordinates of the target overhead crane are solved by the time difference of arrival (TDOA) algorithm. The distance between ultra-wideband base stations is 15-25 meters, supporting sub-meter positioning. The encoder displacement increment is used to record the displacement change of the overhead crane on the track. The output is the change in position, not the absolute value. It is output in real time by the incremental encoder installed on the overhead crane drive wheel axle, recording the linear displacement increment of the target overhead crane along the track, and resolving The accuracy is ±0.01 meters; the inertial measurement unit is composed of a gyroscope (for measuring angular velocity) and an accelerometer (for measuring linear acceleration), which is used to calculate the attitude changes of the overhead crane in space (such as roll, pitch, yaw angle, etc.). The inertial measurement unit data includes the three-axis acceleration, angular velocity and attitude angle of the target overhead crane body, which can be collected by the inertial measurement unit at a sampling rate of 200Hz; the lidar scanning data is the high-precision point cloud data of the environment around the overhead crane obtained by the lidar by emitting laser beams and receiving reflected light, which can be used for absolute positioning and obstacle detection.

[0028] For example, data can be periodically collected from sensors installed at various locations on the target overhead crane: the ultra-wideband base station feeds back position coordinates, the encoder records the travel distance, the inertial measurement unit measures posture changes, and the lidar scans the track environment in real time and identifies QR code markers; through wireless or wired communication modules, these raw data are transmitted to the central control system or edge computing node in real time, providing key support for subsequent high-precision data fusion and overhead crane trajectory control.

[0029] By implementing step 101, a variety of sensor data are obtained, and the position information, attitude changes and motion status of the overhead crane can be fully obtained, which is conducive to the subsequent redundancy and complementarity between sensors. Even if some sensors are interfered with by the environment (such as electromagnetic interference or occlusion), the method can still be kept running, providing a solid foundation for subsequent high-precision fusion positioning.

[0030] Step 102: fusing the ultra-wideband positioning signal, the encoder displacement increment, and the inertial measurement unit data based on a robust Kalman filter algorithm to obtain a first positioning result;

[0031] In some examples, the Robust Kalman Filter is a data fusion algorithm developed based on the traditional Kalman filter that has the ability to suppress outliers. It can maintain estimation accuracy and stability when noise, interference, drift or sudden changes occur in the sensor data. The first positioning result is based on the ultra-wideband positioning signal, encoder displacement increment and inertial measurement unit data. After being processed by the robust Kalman filter, the output is a high-confidence overhead crane position and attitude estimate, which may include the spatial position (x, y, z) and attitude angle (roll, pitch, yaw) of the target overhead crane. The fusion processing process refers to the use of algorithms to match data from different sensors with physical models according to time, weight them, and correct errors to generate a unified estimation result that is more accurate and stable than that of a single sensor.

[0032] For example, in the overhead crane system of a steel plant, the ultra-wideband signal will jump due to metal obstruction and high temperature interference, and the inertial measurement unit is prone to cumulative errors over time; through the anti-error Kalman filter, the ultra-wideband abnormal values ​​​​can be automatically identified and their weights can be reduced. At the same time, the encoder and inertial measurement unit can be used to predict the current displacement and posture. The final output of the first positioning result is both absolutely referenceable and retains high dynamic response capabilities, which can lay a high-precision data foundation for the subsequent path planning and trajectory control of the overhead crane.

[0033] Through the implementation of step 102, the robust Kalman filter is used to enhance the ability to identify and suppress outliers, automatically eliminate abnormal data, and effectively reduce error propagation; the encoder and inertial measurement unit provide high-frequency, continuous relative displacement and attitude information, while the ultra-wideband provides a low-frequency, global positioning reference. The fusion of the three can achieve dynamic, real-time and accurate positioning estimation; the first positioning result obtained has high short-term dynamic response capability and stability, and is suitable for high-speed or complex motion scenarios.

[0034] Step 103: performing point cloud registration processing on the laser radar scanning data to obtain a second positioning result;

[0035] In some examples, point cloud registration processing refers to spatially aligning the current scan point cloud data obtained from the LiDAR with an existing reference map or a known landmark point cloud to calculate the position and attitude of the target overhead crane relative to the reference coordinate system. The point cloud registration processing process can be based on the optimal rigid transformation between the nearest neighbor iterative calculation point clouds, or it can be based on the alignment of point clouds based on a probability density function. It can also extract the geometric features of the QR code in the laser scan and match them one-to-one with the positions in the pre-stored database for higher accuracy. The second positioning result refers to the absolute pose estimate of the target overhead crane output after the LiDAR scan data is point cloud registered with the reference environment (such as a QR code map), which can include position coordinates (x, y, z) and attitude angles (roll, pitch, yaw).

[0036] For example, in high-precision industrial environments (such as heavy machinery plants or automated warehouses), lidar combined with QR code markers can achieve millimeter-level absolute positioning; for example, when the target overhead crane runs along the track, QR code reflectors are arranged on the side walls of the track at certain intervals. The lidar scanning data extracts the geometric distribution features of these QR codes in real time and matches them with the positions in the known database, which can achieve precise positioning and calibration of the current point.

[0037] By implementing step 103, using lidar in combination with environmental QR code markers to perform point cloud matching, accurate global position correction can be achieved, effectively solving the cumulative error (drift) problem caused by the inertial measurement unit and encoder, and enhancing the positioning accuracy of the control method during long-term operation.

[0038] Step 104: determining a posture offset compensation amount using a preset dynamic model, wherein the preset dynamic model is constructed based on the hook swing amplitude and load mass of the target overhead crane;

[0039] In some examples, the preset dynamic model is a mathematical model established in advance based on the physical structure and motion characteristics of the target overhead crane. It is used to describe the dynamic response law of the position and attitude offset caused by swing and inertia during the operation of the load. It can be used to calculate the deviation between the actual effective position of the target overhead crane and the position measured by the sensor, and provide compensation to improve the overall positioning accuracy. The hook swing amplitude refers to the swing angle or swing displacement amplitude caused by factors such as inertia, braking or direction change during the operation of the target overhead crane hoisting the load. The angular velocity and acceleration can be obtained in real time through the inertial measurement unit installed on the hook, and the real-time swing angle can be calculated using trigonometric functions or Kalman filtering algorithms. The load mass is the actual weight of the object currently being hoisted by the target overhead crane, which will directly affect the swing characteristics, inertial response and position offset. It can be obtained by a weighing sensor or a load detection device embedded in the crane, or preset by the operating system, such as entering it in the work task list.

[0040] For example, the inertial measurement unit data at the hook can be read in real time to identify the swing angle and frequency, and then combined with the current load mass provided by the weighing module, it is input into the preset nonlinear dynamic model to calculate the current posture offset; for example, in the heavy load, rapid braking or steering stage, the model can output a lateral + longitudinal coordinate correction value, that is, the posture offset compensation value.

[0041] By implementing step 104, the dynamic factors in the operation of the overhead crane (such as hook swing and load change) are modeled as key variables affecting the posture, which can effectively compensate for the actual offset error and improve the response accuracy of the control method under non-ideal working conditions such as variable load, shaking, acceleration / deceleration, etc. The obtained posture offset compensation plays a dynamic error correction role in subsequent control instructions, which can improve the consistency of positioning and trajectory control.

[0042] Step 105: Control the target overhead travelling crane according to the first positioning result, the second positioning result, and the posture offset compensation amount;

[0043] In some examples, the final positioning result after the fusion and compensation of the first positioning result, the second positioning result, and the posture offset compensation amount can be converted into a control instruction (such as position, speed, acceleration or PID parameters) to drive the target overhead crane to accurately move, turn, brake or position to a specified working point.

[0044] By implementing step 105, the first positioning result, the second positioning result and the posture offset compensation are integrated to form a multi-source fusion control input, ensuring that the control instructions are both accurate and robust, and enabling the overhead crane to achieve high-quality operation with high path accuracy, stable control response and small error in complex environments.

[0045] In summary, the sensors in the embodiments of the present application are redundant with each other and have complementary advantages, which can effectively make up for the defect that a single sensor is susceptible to environmental interference, and through the anti-error Kalman filtering algorithm, the ultra-wideband positioning signal, the encoder displacement increment and the inertial measurement unit data are fused, which can automatically eliminate outliers and significantly reduce error propagation. Point cloud matching is used for the absolute position correction of the lidar, which can eliminate the cumulative error caused by long-term operation; accurately calculating the posture offset compensation caused by the swing can effectively compensate for the actual position deviation in dynamic operation, and can improve the accuracy and stability under working conditions such as variable load and shaking; the first positioning result (fusion estimate), the second positioning result (laser correction value) and the posture compensation are comprehensively calculated to achieve more accurate control command output. In summary, the overhead crane positioning control method provided in the embodiments of the present application can achieve high-precision, anti-interference, and adaptable to complex working conditions overhead crane positioning and control through multi-sensor fusion and dynamic compensation technology.

[0046] In some embodiments, the aforementioned step 102 may include: detecting and eliminating abnormal measurement values ​​in the ultra-wideband positioning signal through the Mahalanobis distance threshold method to obtain a corrected ultra-wideband positioning signal; determining a state estimation covariance matrix based on the distribution characteristics of the abnormal measurement values; performing displacement integration on the encoder displacement increment according to the overhead crane kinematic model to generate an overhead crane relative displacement prediction value; inputting the angular velocity component and the acceleration component in the inertial measurement unit data into a pre-calibrated attitude solution module to obtain the overhead crane body attitude angle; performing multi-source data weighted fusion on the relative displacement prediction value, the overhead crane body attitude angle and the corrected ultra-wideband positioning signal based on the state estimation covariance matrix to obtain a state prediction vector; performing robust Kalman filtering update processing on the state prediction vector to obtain a first positioning result.

[0047] In some examples, the Mahalanobis distance thresholding method is a statistical method used to detect outliers in multivariate data. It measures the Mahalanobis distance between an observation and the overall data distribution to determine whether the value deviates from the normal range. For example, if the Mahalanobis distance of a certain ultra-wideband ranging value is greater than a set threshold, it is considered an outlier. In ultra-wideband ranging, outliers are data that deviate from the true value due to factors such as occlusion, reflection, and electromagnetic interference. This can be determined using the Mahalanobis distance or by combining time series trends to determine the jump value. Correcting the ultra-wideband positioning signal involves removing or replacing outliers from the ultra-wideband data set for subsequent fusion processing to improve positioning reliability. Outliers can be replaced using interpolation, forward and backward mean, or a prediction model. The state estimation covariance matrix describes the joint distribution of the estimated errors of the system state variables. It is part of the Kalman filter core and can be dynamically adjusted based on the frequency of outliers, the magnitude of measurement errors, or an empirical model. For example, when the outlier rate is high, the covariance can be increased to enhance noise immunity. The kinematic model of the overhead crane is a mathematical model that describes the motion patterns of the overhead crane (such as speed, acceleration, and direction). A differential equation or state transition model can be established based on the structural parameters of the overhead crane (such as wheelbase and motor response time). The predicted relative displacement of the overhead crane is the distance the crane has moved or the change in coordinates calculated by integrating the encoder displacement increments through the kinematic model. The encoder displacement increments can be accumulated and combined with the direction to calculate the displacement. The pre-calibrated attitude solution module is a fusion processing module for the angular velocity and acceleration of the inertial measurement unit. It is used to solve the attitude angle of the overhead crane body. This can be achieved through filtering algorithms such as complementary filtering and extended Kalman filtering. The attitude angle of the overhead crane body refers to the rotational angle state of the target overhead crane in space and can be expressed using Euler angles (roll, pitch, yaw). The state prediction vector is a prediction of the current state of the target overhead crane, including a vector composed of variables such as position, speed, and attitude. It can be generated by weighted fusion of the relative displacement prediction value, the overhead crane body attitude angle, and the corrected ultra-wideband positioning signal according to the state covariance; the state prediction vector is used as the initial estimate, combined with the new round of observation data to perform robust Kalman filtering update, and the corrected final estimate is output, that is, the first positioning result.

[0048] For example, the Mahalanobis distance algorithm can be used to eliminate abnormal points in ultra-wideband ranging, and the corrected data can be combined with the encoder displacement and inertial measurement unit attitude angle data. The displacement is predicted through the kinematic model, and the three types of data are dynamically weighted and fused by the covariance matrix to obtain a state prediction vector; the state prediction vector is further updated and corrected under the framework of the robust Kalman filter to obtain the first highly reliable and real-time first positioning result, laying the foundation for subsequent positioning fusion and control command generation.

[0049] Through the implementation of the above embodiments, the introduction of robust Kalman filtering and the use of the Mahalanobis distance anomaly elimination method can effectively identify and eliminate outliers before data fusion, reducing the impact of sensor noise on system accuracy; at the same time, the dynamic weight fusion and state estimation covariance adjustment of ultra-wideband, encoder and inertial measurement unit data can improve the robustness and accuracy of the overall state estimation, and ensure the immunity of the positioning results to dynamic interference.

[0050] In some embodiments, the aforementioned step 103 may include: extracting the geometric distribution characteristics of preset QR code markers in the lidar scanning data, wherein the QR code markers are continuously distributed along the track at a preset interval, and the geometric characteristics of each QR code marker are pre-associated and stored with the two-dimensional position coordinates; based on the geometric distribution characteristics, aligning and matching with the preset marker database, determining the absolute coordinate offset of the target overhead crane; correcting the cumulative error of the first positioning result according to the absolute coordinate offset, and generating a second positioning result.

[0051] In some examples, the geometric distribution characteristics of the preset QR code markers refer to the shapes of the QR codes distributed around the factory tracks and recognized by the LiDAR and their corresponding geometric relationships, such as arrangement spacing, angles, side lengths, etc. Edge feature extraction algorithms such as RANSAC and Hough transform are used to detect the geometric distribution characteristics of the preset QR code markers in the LiDAR scan data. A QR code marker is a two-dimensional pattern mark with a unique code that can be recognized by a LiDAR. Each QR code not only has visual information (pattern number), but also can locate its relative coordinates through the geometric structure in the point cloud. For example, an industrial QR code can be attached to a track column, wall or equipment bracket in a size of 10×10 cm and made of reflective film for clearer identification by the LiDAR. The preset spacing is a fixed spacing (such as every 5 meters) at which the QR code markers are evenly placed beside the track to facilitate the construction of a coordinate system. This regular distribution can be used to accurately measure the movement of the overhead crane along the track. The geometric features of each QR code marker and its physical installation coordinates are pre-registered in a preset marker database. This database stores all QR code numbers and their actual two-dimensional positions, sizes, orientations, and other information, providing a reference for point cloud registration and global positioning. The absolute coordinate offset refers to the deviation between the current position perceived by the lidar and the first positioning result, representing the error introduced in the first positioning result due to sensor drift. For example, in the first positioning result: the overhead crane position is (52.3m, 1.0m), the lidar identifies QR code marker #11, and its position in the preset marker database is (50.0m, 0.0m). The offset is Δx = -2.3m, Δy = -1.0m. The first positioning result is compensated by the aforementioned offset to eliminate the long-term integration error of the inertial measurement unit or encoder, and the second positioning result can be obtained.

[0052] For example, when the overhead crane is in operation, the laser radar scans the side of the track every second, extracts several QR code point cloud contours, and matches their numbers with the preset database after geometric recognition. It is detected that the currently recognized QR code should be at 50.0m, but the first positioning fusion data shows that the overhead crane is 52.3m. At this time, the system position estimation is corrected through point cloud registration, and the first positioning result is recalibrated to generate a second positioning result for subsequent path planning and control execution.

[0053] Through the implementation of the above embodiments, the absolute positioning correction of the laser radar is realized by using QR code marker registration point cloud matching, which has the advantages of strong resistance to environmental changes and high calibration accuracy. It can eliminate inertial cumulative errors and filter drift during long-term operation of the overhead crane. The matching accuracy can be further improved through database geometric matching, thereby providing a reliable global positioning correction benchmark for the entire method.

[0054] In some embodiments, the aforementioned step 104 may include: based on the swing amplitude of the hook, calculating the product of the square value of the swing amplitude and the preset swing compensation coefficient to generate a dynamic swing compensation amount; inputting the load mass data into the nonlinear inertia compensation model to output the load inertia compensation parameters; and determining the posture offset compensation amount based on the weighted fusion result of the dynamic swing compensation amount and the load inertia compensation parameters.

[0055] In some examples, the preset swing compensation coefficient is an empirically or fitted coefficient related to the hook length, rope stiffness, etc., and is used to reflect the degree of influence of the hook movement on the overall positioning error of the overhead crane under a specific swing amplitude; the dynamic swing compensation amount is an offset correction value calculated based on the swing amplitude generated by the hook of the target overhead crane during operation, and can be used to estimate the positioning error caused by the swing. The nonlinear inertia compensation model is a model used to describe the posture drift of the overhead crane body caused by the inertial effect under different load masses. It uses a nonlinear function to characterize the complex relationship between mass change and inertial influence; the nonlinear inertia compensation model can be obtained by fitting historical operation data, or it can be trained using methods such as neural networks and nonlinear least squares. The load inertia compensation parameter is the output value of the nonlinear inertia compensation model, reflecting the amount of inertial drift that should be compensated during the movement of the overhead crane due to acceleration / deceleration under the current load mass. The pose offset compensation is the vector correction value used to adjust the positioning result of the overhead crane. It combines the dual effects of hook swing and load inertia, which can improve the accuracy of the overhead crane pose estimation. The compensation information from the two sources can be integrated using a weighted fusion method.

[0056] For example, during the operation of the target overhead crane, the inertial measurement unit monitors the hook swing angle in real time, and the system calculates the dynamic swing compensation amount; at the same time, the load identification system generates the current mass value according to the lifting weight, and inputs it into the pre-trained nonlinear inertia model to output the load inertia compensation parameters; the two are weightedly fused to obtain the posture offset compensation amount, which is used to correct the positioning results of the overhead crane in real time, especially under high-speed movement or heavy-load conditions, which can significantly improve the control accuracy and.

[0057] Through the implementation of the above embodiments, a dynamic compensation model based on hook swing and load mass is constructed, which can dynamically estimate the position error of the overhead crane, and thus solve the positioning drift problem caused by the hook swing; it realizes a more realistic modeling of the actual operating state of the overhead crane, which can improve the dynamic accuracy and control stability of the method, and is particularly suitable for scenarios with frequent load changes or severe shaking during the lifting process.

[0058] In some embodiments, the aforementioned inputting of load mass data into the nonlinear inertia compensation model and outputting of load inertia compensation parameters may include: detecting whether the load mass exceeds a preset mass threshold and generating a load mode judgment signal; when the load mode judgment signal is a light load mode, generating initial inertia compensation parameters based on the product of the load mass and a preset linear compensation coefficient; when the load mode judgment signal is a heavy load mode, generating initial inertia compensation parameters based on the product of the square value of the load mass and a preset nonlinear compensation coefficient; performing directional coupling calculation on the initial inertia compensation parameters according to the overhead crane operation acceleration direction vector, and outputting the load inertia compensation parameters.

[0059] In some examples, the preset mass threshold is a pre-set mass limit used to distinguish between the two operating modes of "light load" and "heavy load", and can be set according to the rated load and safety specifications of the overhead crane; for example, if the maximum rated load of the target overhead crane is 10t, the preset mass threshold can be set to 5t; when the hoisting mass exceeds 5t, it is judged as "heavy load". The load mode judgment signal can be a binary signal, such as a Boolean value or an enumeration value, which can be used to indicate whether the current mode is light load or heavy load. The light load mode is an operating mode that indicates that the current load mass is lower than the preset mass threshold. The hanging object has little effect on the inertia of the overhead crane. A linear compensation strategy is adopted to quickly simplify the calculation and respond quickly; it can be calculated by formula P init =k lin ×m to calculate the initial inertia compensation parameters, where P init is the initial inertia compensation parameter, m is the current load mass, k lin It is a preset linear compensation coefficient that has been calibrated in advance. The heavy load mode indicates that the current load mass is higher than or equal to the preset mass threshold. The inertia effect of large mass is significant, and a higher order nonlinear compensation is required. The compensation parameter can be set to be proportional to the square of the mass to enhance the correction of nonlinear effects under large loads. The formula P can be used to calculate the inertia effect of large mass.init =k nonlin ×m 2 Calculate the initial inertia compensation parameters, where P init is the initial inertia compensation parameter, m is the current load mass, k nonlin is a nonlinear coefficient used in heavy-load mode. The overhead crane's acceleration direction vector is a unit vector describing the direction of acceleration or deceleration of the target crane on the track. It can be calculated from the acceleration command of the drive motor or the linear acceleration direction measured by the inertial measurement unit. For example, if the target crane is accelerating forward, the direction vector is [1, 0]; if it is decelerating to the left, it may be [0, -1]. The initial inertia compensation parameter in scalar form is decomposed along the acceleration direction to obtain the vector compensation value in the global coordinate system, namely the load inertia compensation parameter, which can be directly added to the positioning error correction.

[0060] For example, in a heavy-load lifting operation, the current load is detected to be 6t (exceeding the 5t threshold), the load mode is determined to be "heavy load", and the nonlinear model is called to calculate the initial inertia compensation of 0.0288m; at this time, the target overhead crane is in an accelerating state to the right front, and the acceleration direction vector is [0.707, 0.707]; after directional coupling, the final compensation vector is [0.020, 0.020]m. Superimposing this compensation amount on the fusion positioning result can effectively correct the positioning offset caused by heavy-load inertia and ensure the accurate execution of the subsequent travel path.

[0061] Through the implementation of the above embodiment, the use of linear or nonlinear inertia compensation models is switched according to different load modes, and coupled calculations are performed in combination with the running acceleration direction, which can more accurately reflect the actual impact of the load mass on the movement of the overhead crane. This adaptive compensation strategy can improve the intelligence level and response sensitivity of the control method, and is particularly suitable for coping with complex load change conditions.

[0062] In some embodiments, the aforementioned step 105 may include: performing confidence assessment on the first positioning result and the second positioning result to generate fused positioning coordinates; performing vector superposition on the fused positioning coordinates based on the posture offset compensation amount to generate compensated target coordinates; converting the compensated target coordinates into control parameters in the overhead crane track coordinate system to generate travel speed and direction adjustment instructions; based on the travel speed and direction adjustment instructions, adjusting the proportional integral coefficient of the PID controller according to the real-time load swing spectrum characteristics of the target overhead crane, and executing overhead crane motion trajectory control on the target overhead crane.

[0063] In some examples, confidence assessment is a comprehensive evaluation used to measure the reliability and accuracy of the first and second positioning results. It can be calculated using methods such as covariance matrix and residual error. By comparing the covariance, error variance, or Mahalanobis distance of the two positioning sources, the credibility of each can be assessed and a fusion weight assigned. The fused positioning coordinates are the coordinate positions obtained by fusing the first and second positioning results based on the confidence assessment results. They have the advantages of both dynamic response and environmental mapping and can be implemented using weighted averaging, Bayesian filtering, multi-model fusion, and other methods. For example, if the first positioning coordinates are (2.0, 3.0) and the second positioning is (2.2, 2.9), confidence-weighted fusion yields (2.1, 2.95). The compensated target coordinates are the actual control target position obtained by superimposing the pose offset compensation on the fused positioning coordinates to correct for positional deviations caused by load swing. For example, if the fused positioning coordinates are (2.1, 2.95) and the offset is (0.02, -0.01), the compensated coordinates are (2.12, 2.94). The overhead crane track coordinate system is a local coordinate system based on the track installation direction of the target overhead crane. It is typically two-dimensional (X is the main rail direction, Y is the crossbeam direction) and is used for driving and turning guidance. It can be derived from encoders and gyroscopes mounted on the track, or from LiDAR map projection correction. Control parameters are the conversion of the compensated target coordinates into parameters understandable by the motion control system, such as velocity vector, heading angle, acceleration threshold, etc. Speed ​​and direction control commands are a combination of commands used to guide the overhead crane to execute precise movements. They can be output by the controller to the drive system. The control parameters are used as input to process speed planning and deceleration segments to generate a continuously controllable signal, namely the speed and direction control command. The real-time load swing spectrum is the frequency distribution of the sway generated by the hook or load during operation. This can be obtained by performing a Fast Fourier Transform (FFT) on the acceleration / angular velocity signals acquired in real time by the inertial measurement unit sensor installed on the hook. The proportional-integral coefficients of the PID controller include a proportional coefficient (Kp) for adjusting the response speed and an integral coefficient (Ki) for adjusting the steady-state error. In the event of severe swing, the proportional coefficient should be reduced to improve stability, while the integral coefficient should be increased to suppress cumulative error. For example, during stable operation with a light load, Kp = 2.0 and Ki = 0.5 are used. If the load swings violently, Kp = 1.2 and Ki = 1.0 are adjusted. By adjusting the speed, direction, and controller parameters, the target overhead crane can be moved smoothly along the predetermined path, avoiding resonance, drift, or offset.

[0064] For example, during a path adjustment process, there is a certain deviation between the first positioning result and the second positioning result. The intermediate coordinates (3.2, 4.8) are obtained through confidence-weighted fusion; then, combined with the offset compensation vector (0.03, -0.02), the compensated target coordinates (3.23, 4.78) are calculated; the trajectory controller converts this coordinate into an instruction of "forward 0.75m / s, turn right 5°", and detects that the current load swing frequency reaches 0.5Hz, and automatically adjusts the PID control parameters to Kp=1.4, Ki=0.8, to achieve softer speed changes to suppress swing and ensure that the lifting trajectory is stable and accurate.

[0065] Through the implementation of the above embodiment, the first positioning result and the second positioning result are integrated and combined with posture compensation to realize a high-precision dynamic fusion coordinate system, and then the compensation result under the coordinate system is mapped into the control instruction of the overhead crane, which can ensure that the control output is highly consistent with the actual state; in addition, adjusting the PID controller parameters based on the real-time hook swing spectrum can improve the control algorithm's adaptability to actual disturbances, thereby achieving smoother and more refined trajectory control.

[0066] In some embodiments, the aforementioned confidence assessment of the first positioning result and the second positioning result to generate fused positioning coordinates may include: obtaining current environmental noise parameters, where the environmental noise parameters may include dust concentration, electromagnetic interference intensity and laser reflectivity attenuation coefficient; determining a first confidence weight of the ultra-wideband positioning signal based on the dust concentration and the electromagnetic interference intensity; determining a second confidence weight of the lidar scanning data based on the product of the laser reflectivity attenuation coefficient and the preset QR code matching success rate; and using an adaptive weighted fusion algorithm to perform weighted averaging on the first positioning result and the second positioning result based on the first confidence weight and the second confidence weight to obtain fused positioning coordinates.

[0067] In some examples, the current environmental noise parameter refers to a set of environmental interference factors that affect the accuracy of the sensor, which may include dust concentration, electromagnetic interference intensity, and laser reflectivity attenuation coefficient; dust concentration can be collected using a laser dust sensor, electromagnetic interference intensity can be collected using an electromagnetic interference (EMI) detection module or a radio frequency analyzer, and laser reflectivity attenuation can be automatically estimated by the lidar system through the reflected light intensity. Dust concentration represents the density of suspended particles in the air, and the unit can be μg / m 3, directly impacting UWB communications and lidar echoes. Electromagnetic interference intensity (EMI) represents the strength of electromagnetic background interference, measured in dBμV / m. Interference sources include high-voltage motors and frequency converters. For example, near welding areas, EMI intensity can reach 90 dBμV / m, significantly interfering with UWB signals. The laser reflectivity attenuation coefficient (LRC) is the ratio of the LIDAR transmitted signal to the return signal strength. It assesses signal attenuation caused by the reflective surface material or obstructions. For example, when dust obscures a QR code marker, the reflectivity drops to 30%, impacting point cloud matching. The first confidence weight represents the reliability of the UWB positioning results in the current environment; higher values ​​indicate more reliable data. The second confidence weight represents the reliability of the lidar registration data. It is calculated based on optical properties and a preset QR code matching success rate. The preset QR code matching success rate reflects the probability of successfully identifying the QR code marker in the current environment and is typically fed back by the radar point cloud matching algorithm. For example, in a relatively clean environment, the matching success rate reaches 95%, while under obstruction or light interference, it drops to 60%. The adaptive weighted fusion algorithm is an algorithm that dynamically adjusts the weighting coefficient according to the current environmental confidence of the sensor and fuses the multi-source positioning results.

[0068] Through the implementation of the above embodiments, environmental perception parameters are introduced to dynamically evaluate sensor confidence, and an adaptive weighted fusion algorithm is used to integrate ultra-wideband and lidar results. The weights can be dynamically adjusted under different environmental conditions to implement an intelligent fusion strategy driven by perception credibility, which can improve the environmental adaptability and accuracy robustness of fusion positioning.

[0069] In some embodiments, the step of determining the preset QR code matching success rate may include: obtaining the geometric feature integrity parameters of the QR code marker in the lidar scanning data, wherein the geometric feature integrity parameters include edge clarity, corner point missing rate and reflection intensity uniformity; calculating the initial matching success rate based on the ratio of the number of historical matching successes to the total number of scans within a preset time window; performing dynamic attenuation correction on the initial matching success rate according to the deviation between the current dust concentration and the preset concentration threshold to generate an environmental attenuation factor; determining the preset QR code matching success rate based on the product of the geometric feature integrity parameters and the environmental attenuation factor, wherein when the corner point missing rate exceeds the preset missing threshold, the preset QR code matching success rate is forcibly set to the lowest confidence level.

[0070] In some examples, edge clarity is determined by extracting the QR code outline through Hough transform or Canny edge detection algorithm and calculating the ratio of continuous edge pixels; the corner point missing rate is determined by using Harris corner point detection algorithm to count the missing ratio of the four positioning corner points of the QR code. If one corner point is missing, the missing rate is 25%; the reflection intensity uniformity can be determined by analyzing the standard deviation of the reflection value of the point cloud in the QR code area. The initial matching success rate is the number of times the lidar scans the QR code (such as 8 times) within a preset time window (such as the past 5 seconds), of which the number of successful matches (such as 6 times), then the initial matching success rate is 6 / 8 = 75%. The environmental attenuation factor can be calculated based on the current dust concentration (such as 120μg / m 3 ) and the preset threshold (such as 50μg / m 3 ), for example, if the deviation is (120-50) / 50=1.4, the corresponding dynamic attenuation factor is 1 / (1+1.4)=0.42; if the corner point missing rate exceeds the preset missing threshold (e.g., 30%), regardless of other parameters, the preset QR code matching success rate is forcibly set to the lowest level (e.g., 10%);

[0071] Through the implementation of the above embodiments, the joint modeling of geometric integrity and environmental interference can avoid mismatching due to local feature loss or dust interference. The dust concentration affects the matching success rate weight in real time, which can reduce the confidence of lidar data in harsh environments. The forced degradation mechanism for corner point loss can effectively prevent positioning jumps caused by damaged or contaminated markers.

[0072] In some embodiments, after obtaining the fused positioning coordinates by weighted averaging the first positioning result and the second positioning result using an adaptive weighted fusion algorithm based on the first confidence weight and the second confidence weight, it may also include: deploying distributed environmental monitoring sensors at key nodes of the target overhead crane to collect environmental parameters in real time. The environmental parameters may include temperature gradient, air turbulence intensity and structural vibration spectrum; constructing an environmental interference factor mapping model to quantify the coupling relationship between environmental parameters and positioning errors into a compensation weight matrix; and performing spatiotemporal alignment and error back propagation correction on multi-source sensor data based on the weight matrix to generate an interference-resistant fusion positioning result.

[0073] In some examples, the environmental interference factor mapping model may include the following compensation mechanisms: in terms of temperature gradient compensation, when the driving wheel bearing temperature exceeds 80°C, the encoder displacement increment is corrected according to the thermal expansion correction factor λ=1+0.005×(T-80) to compensate for the integral error caused by thermal expansion and contraction of the wheel diameter; in terms of air turbulence intensity modeling, if the ultrasonic anemometer detects that the turbulent pulsation speed at the beam exceeds 2m / s, the inertial measurement unit attitude angle confidence will be reduced to 30% of the baseline value in an exponential decay manner, thereby reducing the weight of the attitude data in the fusion; in terms of structural vibration spectrum, when the piezoelectric sensor at the track joint detects a resonance peak that coincides with the natural frequency of the overhead crane (such as 25Hz±2Hz), the system introduces a virtual damping term in the Kalman filter process state prediction equation to suppress high-frequency jitter caused by resonance.

[0074] Through the implementation of the above embodiments, a physical environment coupling compensation mechanism is introduced, breaking through the limitations of traditional positioning systems on single sensor error correction and achieving a compensation accuracy improvement of more than 40%; a dynamic weight reconstruction mechanism is constructed to increase the ultra-wideband data utilization rate from 50% to 85% in complex environments such as strong magnetism and high dust; with the help of a virtual damping injection strategy, positioning jitter caused by high-frequency mechanical resonance can be effectively suppressed, improving the anti-interference robustness by more than 90%.

[0075] Furthermore, as an implementation of the aforementioned method embodiment, the present application also provides an overhead crane positioning control device for implementing the aforementioned method embodiment. This device embodiment corresponds to the aforementioned method embodiment. For ease of reading, this overhead crane positioning control device embodiment will no longer describe the details of the aforementioned method embodiment one by one, but it should be clear that the device in the embodiment of the present application can implement all the contents of the aforementioned method embodiment. Figure 2As shown, the overhead crane positioning control device 20 includes: a data acquisition unit 201, a first positioning unit 202, a second positioning unit 203, an offset compensation unit 204 and an overhead crane control unit 205, wherein the data acquisition unit 201 is used to acquire multi-source sensor data of the target overhead crane, wherein the aforementioned multi-source sensor data may include ultra-wideband positioning signals, encoder displacement increments, inertial measurement unit data and laser radar scanning data; the first positioning unit 202 is used to fuse the ultra-wideband positioning signals, encoder displacement increments and inertial measurement unit data based on the robust Kalman filtering algorithm to obtain a first positioning result; the second positioning unit 203 is used to perform point cloud registration processing on the laser radar scanning data to obtain a second positioning result; the offset compensation unit 204 is used to determine the posture offset compensation amount through a preset dynamic model, wherein the preset dynamic model is constructed based on the hook swing amplitude and load mass of the target overhead crane; the overhead crane control unit 205 is used to control the target overhead crane according to the first positioning result, the second positioning result and the posture offset compensation amount.

[0076] In some embodiments, the first positioning unit is further used to detect and eliminate abnormal measurement values ​​in the ultra-wideband positioning signal through the Mahalanobis distance threshold method to obtain a corrected ultra-wideband positioning signal; determine the state estimation covariance matrix based on the distribution characteristics of the abnormal measurement values; perform displacement integration on the encoder displacement increment according to the overhead crane kinematic model to generate a relative displacement prediction value of the overhead crane; input the angular velocity component and acceleration component in the inertial measurement unit data into the pre-calibrated attitude solution module to obtain the attitude angle of the overhead crane body; perform multi-source data weighted fusion of the relative displacement prediction value, the overhead crane body attitude angle and the corrected ultra-wideband positioning signal based on the state estimation covariance matrix to obtain a state prediction vector; perform robust Kalman filtering update processing on the state prediction vector to obtain a first positioning result.

[0077] In some embodiments, the second positioning unit is also used to extract the geometric distribution characteristics of preset QR code markers in the lidar scanning data, wherein the QR code markers are continuously distributed along the track at a preset interval, and the geometric characteristics of each QR code marker are pre-associated with the two-dimensional position coordinates and stored; based on the geometric distribution characteristics, alignment and matching are performed with the preset marker database to determine the absolute coordinate offset of the target overhead crane; the cumulative error of the first positioning result is corrected according to the absolute coordinate offset to generate a second positioning result.

[0078] In some embodiments, the offset compensation unit is also used to calculate the product of the square of the swing amplitude and the preset swing compensation coefficient based on the swing amplitude of the hook to generate a dynamic swing compensation amount; input the load mass data into the nonlinear inertia compensation model to output the load inertia compensation parameters; and determine the posture offset compensation amount based on the weighted fusion result of the dynamic swing compensation amount and the load inertia compensation parameters.

[0079] In some embodiments, the offset compensation unit is also used to detect whether the load mass exceeds a preset mass threshold and generate a load mode judgment signal; when the load mode judgment signal is a light load mode, the initial inertia compensation parameter is generated based on the product of the load mass and a preset linear compensation coefficient; when the load mode judgment signal is a heavy load mode, the initial inertia compensation parameter is generated based on the product of the square value of the load mass and a preset nonlinear compensation coefficient; the initial inertia compensation parameter is directionally coupled calculated according to the overhead crane operation acceleration direction vector, and the load inertia compensation parameter is output.

[0080] In some embodiments, the overhead crane control unit is further used to perform confidence assessment on the first positioning result and the second positioning result to generate fused positioning coordinates; perform vector superposition on the fused positioning coordinates based on the posture offset compensation amount to generate compensated target coordinates; convert the compensated target coordinates into control parameters in the overhead crane track coordinate system to generate travel speed and direction adjustment instructions; based on the travel speed and direction adjustment instructions, adjust the proportional integral coefficient of the PID controller according to the real-time load swing spectrum characteristics of the target overhead crane, and execute overhead crane motion trajectory control on the target overhead crane.

[0081] In some embodiments, the overhead crane control unit is also used to obtain current environmental noise parameters, where the environmental noise parameters include dust concentration, electromagnetic interference intensity and laser reflectivity attenuation coefficient; based on the dust concentration and electromagnetic interference intensity, a first confidence weight of the ultra-wideband positioning signal is determined; according to the product of the laser reflectivity attenuation coefficient and the preset QR code matching success rate, a second confidence weight of the lidar scanning data is determined; according to the first confidence weight and the second confidence weight, an adaptive weighted fusion algorithm is used to perform weighted averaging on the first positioning result and the second positioning result to obtain a fused positioning coordinate.

[0082] The present application also provides a computer-readable storage medium, which stores computer-executable instructions or computer programs. When the computer-executable instructions or computer programs are executed by a processor, the processor will be caused to execute any step of the overhead crane positioning control method provided in the present application.

[0083] In some embodiments, the computer-readable storage medium may be a random access memory (RAM), a read-only memory (ROM), a flash memory, a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM); or it may be various devices including one or any combination of the above memories.

[0084] In some embodiments, computer-executable instructions may be in the form of a program, software, software module, script, or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as a stand-alone program or as a module, component, subroutine, or other unit suitable for use in a computing environment.

[0085] In some embodiments, computer-executable instructions may, but need not, correspond to a file in a file system, may be stored as part of a file that stores other programs or data, for example, in one or more scripts in a HyperText Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple coordinating files (for example, files storing one or more modules, subroutines, or code portions).

[0086] In some embodiments, computer-executable instructions may be deployed to be executed on one electronic device, or on multiple electronic devices located at one site, or on multiple electronic devices distributed across multiple sites and interconnected by a communication network.

[0087] like Figure 3 As shown, the present application also provides an electronic device 30, including a memory 310, a processor 320 and a computer program 311 stored in the memory 310 and executable on the processor. When the processor 320 executes the computer program 311, any step of the above-mentioned overhead crane positioning control method is implemented.

[0088] The present application also provides a computer program product, comprising a computer program or computer-executable instructions stored in a computer-readable storage medium. A processor of an electronic device reads the computer program or computer-executable instructions from the computer-readable storage medium and executes the computer program or computer-executable instructions, causing the electronic device to perform any step of the overhead crane positioning control method described above.

[0089] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for controlling positioning of an overhead crane, characterized in that: include: Acquiring multi-source sensor data of the target overhead crane, wherein the multi-source sensor data includes ultra-wideband positioning signals, encoder displacement increments, inertial measurement unit data, and lidar scanning data; fusing the ultra-wideband positioning signal, the encoder displacement increment, and the inertial measurement unit data based on a robust Kalman filter algorithm to obtain a first positioning result; Performing point cloud registration processing on the laser radar scanning data to obtain a second positioning result; Determining a posture offset compensation amount through a preset dynamic model, wherein the preset dynamic model is constructed based on the hook swing amplitude and load mass of the target overhead crane; The target overhead travelling crane is controlled according to the first positioning result, the second positioning result and the posture offset compensation amount.

2. The method for controlling the positioning of an overhead crane according to claim 1, wherein: The step of fusing the ultra-wideband positioning signal, the encoder displacement increment, and the inertial measurement unit data based on a robust Kalman filter algorithm to obtain a first positioning result includes: Detecting and eliminating abnormal measurement values ​​in the ultra-wideband positioning signal by using a Mahalanobis distance threshold method to obtain a corrected ultra-wideband positioning signal; determining a state estimation covariance matrix based on distribution characteristics of the abnormal measurement values; Performing displacement integration on the encoder displacement increment according to the overhead crane kinematic model to generate an overhead crane relative displacement prediction value; Inputting the angular velocity component and the acceleration component in the inertial measurement unit data into a pre-calibrated attitude solution module to obtain the attitude angle of the overhead crane body; Performing multi-source data weighted fusion on the relative displacement prediction value, the attitude angle of the overhead crane body, and the corrected ultra-wideband positioning signal based on the state estimation covariance matrix to obtain a state prediction vector; Perform robust Kalman filtering update processing on the state prediction vector to obtain the first positioning result.

3. The method for controlling positioning of an overhead crane according to claim 1, wherein: The performing point cloud registration processing on the laser radar scanning data to obtain a second positioning result includes: Extracting geometric distribution features of preset two-dimensional code markers in the laser radar scan data, wherein the two-dimensional code markers are continuously distributed along the track at a preset interval, and the geometric features of each two-dimensional code marker are pre-associated and stored with two-dimensional position coordinates; Based on the geometric distribution characteristics, the registration and matching is performed with a preset landmark database to determine the absolute coordinate offset of the target overhead traveling vehicle; The accumulated error of the first positioning result is corrected according to the absolute coordinate offset to generate the second positioning result.

4. The method for controlling positioning of an overhead crane according to claim 1, wherein: The determining of the posture offset compensation amount by using a preset dynamic model includes: Based on the hook swing amplitude, the product of the square value of the swing amplitude and a preset swing compensation coefficient is calculated to generate a dynamic swing compensation amount; Inputting the load mass data into a nonlinear inertia compensation model and outputting load inertia compensation parameters; The posture offset compensation amount is determined according to a weighted fusion result of the dynamic swing compensation amount and the load inertia compensation parameter.

5. The method for controlling positioning of an overhead crane according to claim 1, wherein: Inputting the load mass data into a nonlinear inertia compensation model and outputting load inertia compensation parameters includes: detecting whether the load mass exceeds a preset mass threshold and generating a load mode determination signal; When the load mode determination signal indicates a light load mode, generating an initial inertia compensation parameter based on a product of the load mass and a preset linear compensation coefficient; When the load mode determination signal indicates a heavy load mode, generating the initial inertia compensation parameter based on a product of a square value of the load mass and a preset nonlinear compensation coefficient; Directional coupling calculation is performed on the initial inertia compensation parameter according to the overhead crane running acceleration direction vector, and the load inertia compensation parameter is output.

6. The method for controlling positioning of an overhead crane according to claim 1, wherein: The controlling of the target overhead travelling crane according to the first positioning result, the second positioning result and the posture offset compensation amount includes: Performing a confidence evaluation on the first positioning result and the second positioning result to generate fused positioning coordinates; Performing vector superposition on the fused positioning coordinates based on the posture offset compensation amount to generate compensated target coordinates; Converting the compensated target coordinates into control parameters in the overhead crane track coordinate system to generate travel speed and direction adjustment instructions; Based on the travel speed and direction adjustment instructions, the proportional integral coefficient of the PID controller is adjusted according to the real-time load swing spectrum characteristics of the target overhead crane, and the overhead crane motion trajectory control is performed on the target overhead crane.

7. The method for controlling positioning of an overhead crane according to claim 6, wherein: The performing confidence evaluation on the first positioning result and the second positioning result to generate fused positioning coordinates includes: Acquiring current environmental noise parameters, wherein the environmental noise parameters include dust concentration, electromagnetic interference intensity, and laser reflectivity attenuation coefficient; Determining a first confidence weight of the ultra-wideband positioning signal based on the dust concentration and the electromagnetic interference intensity; Determining a second confidence weight of the laser radar scanning data according to a product of the laser reflectivity attenuation coefficient and a preset two-dimensional code matching success rate; According to the first confidence weight and the second confidence weight, an adaptive weighted fusion algorithm is used to perform weighted averaging on the first positioning result and the second positioning result to obtain the fused positioning coordinates.

8. A positioning control device for an overhead crane, characterized in that: include: A data acquisition unit, configured to acquire multi-source sensor data of a target overhead crane, wherein the multi-source sensor data includes ultra-wideband positioning signals, encoder displacement increments, inertial measurement unit data, and lidar scanning data; a first positioning unit, configured to fuse the ultra-wideband positioning signal, the encoder displacement increment, and the inertial measurement unit data based on a robust Kalman filter algorithm to obtain a first positioning result; A second positioning unit is used to perform point cloud registration processing on the laser radar scanning data to obtain a second positioning result; an offset compensation unit, configured to determine a posture offset compensation amount by using a preset dynamic model, wherein the preset dynamic model is constructed based on the hook swing amplitude and load mass of the target overhead crane; The overhead crane control unit is used to control the target overhead crane according to the first positioning result, the second positioning result and the posture offset compensation amount.

9. An electronic device comprising: A memory and a processor, wherein the processor is configured to implement the steps of the overhead crane positioning control method according to any one of claims 1 to 7 when executing the computer program stored in the memory.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the overhead crane positioning control method according to any one of claims 1 to 7 are implemented.

Images