A method and system for trajectory planning of a bridge crane

Abstract

The present application relates to a kind of bridge crane trajectory planning method and system, the motion signal of main beam and trolley is collected by electromagnetic field intensity detection sensor array, after being denoised, feature is extracted and is decomposed into component signal group, establish its associated relationship with joint movement state, realize high-precision dynamic positioning, construct time series model based on position data analysis displacement and speed characteristics, generate the multi-dimensional constraint condition set containing speed limit and position boundary, iteratively adjust the final trajectory scheme output by optimization calculation framework, using dynamic feedback mechanism, electromagnetic field intensity signal and planning data are compared in real time, when exceeding deviation threshold, re-adjust parameter, generate coordinated motion instruction drive crane operation.The present application is based on electromagnetic field intensity detection and realizes accurate positioning, effectively solve the cumulative error and environmental interference problem of traditional method, significantly improve the stability, accuracy and self-adapting ability of crane movement.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of bridge crane trajectory planning, in particular to a bridge crane trajectory planning method and system. BACKGROUND

[0002] The bridge crane is a heavy lifting equipment widely used in industrial production and logistics transportation, mainly composed of a main beam, a trolley, a lifting mechanism, etc., and can realize precise positioning and hoisting of goods in three-dimensional space. Due to its complex structure, large load and significant motion inertia, how to plan an efficient, smooth and safe motion trajectory has always been a technical difficulty in actual operation. The traditional bridge crane trajectory planning method mainly relies on preset path or open-loop control based on simple kinematic model, which is difficult to adapt to the precise control demand in dynamic environment.

[0003] In the prior art, the path planning of the bridge crane usually adopts positioning methods based on encoders or laser ranging, but these methods have certain limitations. For example, the encoder is easily affected by mechanical wear and cumulative error, resulting in a decrease in positioning accuracy over time; while laser ranging has high accuracy, but is easily affected by dust, vibration and other disturbances in complex industrial environments, affecting data reliability. In addition, existing methods often rely on offline planning and lack real-time dynamic adjustment capability, resulting in problems such as vibration, overshoot or response lag of the crane during operation, affecting work efficiency and safety. SUMMARY

[0004] Therefore, the technical problem to be solved by the present application is to overcome the problems existing in the path planning of the bridge crane in the prior art, and to provide a bridge crane trajectory planning method and system based on electromagnetic field strength detection for positioning the bridge crane, thereby realizing path planning of the bridge crane, effectively overcoming the cumulative error and environmental interference problems of traditional positioning methods, and significantly improving the stability, accuracy and adaptive ability of the crane motion.

[0005] To solve the above technical problems, the present application provides a bridge crane trajectory planning method, comprising:

[0006] By means of an electromagnetic field strength detection sensor array, electromagnetic field strength signal data during the motion of the main beam and the trolley is collected, the electromagnetic field strength signal data is denoised to obtain a purified electromagnetic field strength signal data set;

[0007] According to the purified electromagnetic field strength signal data set, features are extracted and decomposed into a plurality of component signal groups, and the association between the component signal groups and the joint motion state is determined;

[0008] According to the association between the component signal groups and the joint motion state, the dynamic position coordinates of the main beam and the trolley are mapped to obtain high-precision dynamic position data records;

[0009] According to the high-precision dynamic position data record, a time series-based position change model is constructed, the displacement and speed characteristics of the main girder and the trolley at different time points are analyzed, and the real-time motion state description of each joint is determined;

[0010] According to the real-time motion state description, a multi-dimensional constraint condition set containing speed limit and position boundary is generated, and a trajectory planning basic parameter set suitable for complex scenes is obtained;

[0011] According to the trajectory planning basic parameter set, a preset optimization calculation framework is applied for iterative adjustment, and final trajectory planning scheme data is output;

[0012] According to the final trajectory planning scheme data, a dynamic feedback mechanism is constructed, the real-time collected electromagnetic field intensity signal is compared with the trajectory planning scheme data, if the comparison result exceeds the preset deviation threshold, the trajectory planning basic parameter set is re-adjusted, and the updated trajectory planning data is obtained;

[0013] According to the updated trajectory planning data, a coordinated motion instruction set of each joint is generated, the crane is driven to execute trajectory motion through the instruction set, and the state stability in the motion process is determined.

[0014] In an embodiment of the present application, through an electromagnetic field intensity detection sensor array, electromagnetic field intensity signal data in the motion process of the main girder and the trolley is collected, the electromagnetic field intensity signal data is denoised to obtain a purified electromagnetic field intensity signal data set, including:

[0015] A plurality of electromagnetic field intensity detection sensors are arranged along the track to form a sensor array covering the motion path of the main girder and the trolley, and electromagnetic field intensity signal data is collected in real time when the main girder and the trolley move to obtain an initial electromagnetic field intensity signal set;

[0016] For the initial electromagnetic field intensity signal set, a preset filtering mechanism is used to process the signal and remove noise interference to generate a purified electromagnetic field intensity signal data set;

[0017] The purified electromagnetic field intensity signal data set is processed in segments, and the continuous signal data is divided into a plurality of time segment signal subsets, so that more detailed data calibration can be performed on each time segment signal subset in the future;

[0018] For the divided signal subsets, data smoothing is performed one by one to further eliminate small fluctuations in the signal, and an optimized electromagnetic field intensity signal data set is obtained for subsequent motion state monitoring of the main girder and the trolley.

[0019] In an embodiment of the present application, according to the purified electromagnetic field intensity signal data set, features are extracted and decomposed into a plurality of component signal groups, the association between the component signal groups and the articulation state is determined, including:

[0020] From the purified electromagnetic field intensity signal data set, time series features and spatial distribution features are extracted, and for the nonlinear changes in the signal caused by the multi-degree-of-freedom structure, the purified electromagnetic field intensity signal data set is split into a plurality of independent component signal groups by using a preset signal decomposition method, and the signal components related to different motion dimensions are preliminarily separated;

[0021] For the component signal groups, by comparing the fluctuation characteristics of each component signal group with the articulation motion patterns recorded in the database, the correspondence between each component signal group and a specific articulation state is determined, and a mapping table of component signals and motion states is formed;

[0022] According to the mapping table of component signals and motion states, the change rule of each component signal group in the time dimension is extracted, and for the nonlinear characteristics presented in the change rule, a preset signal processing tool is used for smoothing processing to generate a smoothed component signal feature set, which is used for further verification of the association relationship;

[0023] Using the smoothed component signal feature set, combined with the mapping table of articulation states, by comparing the signal changes in the feature set with the dynamic changes of the motion state one by one, the association between the component signal groups and the articulation state is refined to ensure the accuracy and reliability of the association, and the feature decomposition and state association of the purified electromagnetic field intensity signal data set are completed.

[0024] In an embodiment of the present application, according to the association between the component signal groups and the articulation state, the dynamic position coordinates of the girder and the trolley are mapped to obtain high-precision dynamic position data records, including:

[0025] According to the association between the component signal groups and the articulation state, combined with a preset electromagnetic field intensity change and position coordinate mapping relationship database, a nonlinear regression analysis method is used to map the component signal groups to the dynamic position coordinates of the girder and the trolley to generate preliminary position data records;

[0026] For the preliminary position data records, Kalman filtering is applied to smooth the dynamic position coordinates to eliminate random errors in the coordinate data, and high-precision dynamic position data records are obtained;

[0027] According to the high-precision dynamic position data records, a position change model based on time series is constructed to calculate the displacement and speed of the girder and the trolley at different time points to generate dynamic position data records.

[0028] In an embodiment of the present application, according to the high-precision dynamic position data record, a time-series-based position change model is constructed, the displacement and speed characteristics of the girder and the trolley at different time points are analyzed, and the real-time motion state description of each joint is determined, including:

[0029] From the high-precision dynamic position data record, the position coordinate information of the girder and the trolley at multiple time points is obtained, which is sorted into a time-series-based position data set. The position data is arranged in chronological order to form a continuous time position sequence;

[0030] For the continuous time position sequence, the displacement between adjacent time points of the girder and the trolley is calculated, and the speed characteristics of each time point are further determined by the ratio of displacement to time interval, and a characteristic curve reflecting the displacement and speed change with time is constructed;

[0031] From the characteristic curve, the displacement change trend and speed change trend of the girder and the trolley at different time points are extracted, and the motion state of each joint at the corresponding time point is judged whether it meets the expectation in combination with the preset joint motion range parameters, to form a preliminary motion state description;

[0032] For the preliminary motion state description, the real-time motion state of each joint is refined and labeled in combination with the time-series-based position data set, and the specific motion posture of the girder and the trolley at each time point is determined, and finally the time-series-based position change model construction and the real-time motion state description of each joint are completed.

[0033] In an embodiment of the present application, according to the real-time motion state description, a multi-dimensional constraint condition set containing speed limit and position boundary is generated, and a trajectory planning basic parameter set suitable for complex scenes is obtained, including:

[0034] From the real-time motion state description, the current speed and position data of the crane are obtained, and a constraint range containing the upper and lower limits of the speed is constructed for these data, and the position boundary range is determined to form an initial multi-dimensional constraint condition set;

[0035] The initial multi-dimensional constraint condition set is adjusted for scene adaptation, and the speed constraint and position boundary are refined in combination with the obstacle distribution and work area restriction in the complex scene to form an adapted constraint condition set;

[0036] Key parameters are extracted from the adapted constraint condition set to construct a trajectory planning basic parameter set suitable for complex scenes, ensuring that the parameter set can cover the multi-dimensional constraint requirements of speed limit and position boundary;

[0037] Data integration is performed on the trajectory planning basic parameter set, which is compared with the real-time motion state description. If it is found that the parameter set deviates from the actual state beyond the preset threshold, the parameter set is locally adjusted to ensure that the finally generated parameter set is suitable for the trajectory planning requirements of complex scenes.

[0038] In an embodiment of the present application, according to the trajectory planning basic parameter set, a preset optimization calculation framework is applied for iterative adjustment, and final trajectory planning scheme data is output, including:

[0039] Initial parameters are obtained from the preset trajectory planning basic parameter set, including the motion range and speed constraint of the crane joint, a preliminary geometric path of the target trajectory is constructed based on the initial parameters, the position and angle of the path node are adjusted through iterative calculation, and it is judged whether the adjusted path meets the preset convergence condition. If it meets, a preliminary trajectory planning scheme data is generated;

[0040] Based on the preliminary trajectory planning scheme data, the coordinates and speed information of the path node are extracted, a path smoothness evaluation function is constructed, the path node is smoothed through the least square method, and smoothed trajectory planning scheme data is generated;

[0041] For the smoothed trajectory planning scheme data, the time sequence distribution of each path node is calculated, the coordinated motion time sequence of each joint is generated combined with the dynamics constraint of the crane, and the continuity requirement of the time sequence is determined;

[0042] According to the coordinated motion time sequence, the motion instruction set of each joint is generated, the instruction set includes the control parameters of joint angle and speed, which is used to drive the crane to execute trajectory motion, and the final trajectory planning scheme data is output.

[0043] In an embodiment of the present application, according to the final trajectory planning scheme data, a dynamic feedback mechanism is constructed, the real-time collected electromagnetic field intensity signal is compared with the trajectory planning scheme data, and if the comparison result exceeds the preset deviation threshold, the trajectory planning basic parameter set is re-adjusted to obtain updated trajectory planning data, including:

[0044] Signal intensity data is obtained from the real-time collected electromagnetic field intensity signal, an electromagnetic field intensity distribution data set containing time stamp and position coordinates is constructed, the electromagnetic field intensity distribution data set is standardized to generate a standardized electromagnetic field intensity distribution matrix;

[0045] The standardized electromagnetic field intensity distribution matrix is compared with the final trajectory planning scheme data, the deviation value of the two in the position coordinates is calculated, and a deviation distribution vector is generated;

[0046] For the deviation distribution vector, it is judged whether it exceeds the preset deviation threshold, and if it exceeds, the speed and direction parameters in the trajectory planning basic parameter set are adjusted according to the value of the deviation distribution vector to generate an adjusted parameter set;

[0047] According to the adjusted parameter set, the preset optimization calculation framework is applied to iteratively update the trajectory planning data to generate updated trajectory planning data.

[0048] In an embodiment of the present application, according to the updated trajectory planning data, a coordinated motion instruction set for each joint is generated, and the crane is driven to perform trajectory motion through the instruction set to determine the state stability in the motion process, including:

[0049] From the updated trajectory planning data, the expected position and speed parameters of each joint are obtained, and the motion trajectory of each joint is discretized based on time series to generate an initial motion instruction set containing time stamps, position coordinates and speed vectors;

[0050] According to the initial motion instruction set, a constraint condition set containing speed limits and position boundaries is constructed for the dynamics of each joint of the crane to generate a coordinated motion instruction set, ensuring the synchronization and smoothness of each joint in the motion process;

[0051] The control system of the crane is driven by the coordinated motion instruction set, the displacement and speed data of the main beam and the trolley are collected in real time, a position change model based on time series is constructed, and the real-time motion state of each joint is recorded;

[0052] From the position change model, the displacement and speed characteristics of the main beam and the trolley are extracted, and the constraint condition set is combined to judge the state stability in the motion process to generate a trajectory planning basic parameter set suitable for complex scenes.

[0053] To solve the above technical problems, the present application also provides a bridge crane trajectory planning system for implementing the above method, comprising:

[0054] The sensor array module detects the electromagnetic field intensity signal data in the motion process of the main beam and the trolley through the electromagnetic field intensity detection sensor array, and obtains a purified electromagnetic field intensity signal data set by denoising the electromagnetic field intensity signal data;

[0055] The feature extraction module extracts features and decomposes them into a plurality of component signal groups according to the purified electromagnetic field intensity signal data set, and determines the association between the component signal groups and the joint motion state;

[0056] The position mapping module maps the dynamic position coordinates of the main beam and the trolley according to the association between the component signal groups and the joint motion state to obtain high-precision dynamic position data records;

[0057] A modeling analysis module constructs a time series-based position change model according to the high-precision dynamic position data record, analyzes displacement and speed characteristics of the main girder and the trolley at different time points, and determines a real-time motion state description of each joint.

[0058] A constraint generation module generates a multi-dimensional constraint condition set containing speed limits and position boundaries according to the real-time motion state description, and obtains a trajectory planning basic parameter set suitable for complex scenes.

[0059] An optimization adjustment module applies a preset optimization calculation framework for iterative adjustment according to the trajectory planning basic parameter set, and outputs final trajectory planning scheme data.

[0060] A feedback adjustment module constructs a dynamic feedback mechanism according to the final trajectory planning scheme data, compares the real-time collected electromagnetic field intensity signal with the trajectory planning scheme data, and if the comparison result exceeds a preset deviation threshold, triggers re-adjustment of the trajectory planning basic parameter set to obtain updated trajectory planning data.

[0061] A motion control module generates a coordinated motion instruction set of each joint according to the updated trajectory planning data, drives the crane to perform trajectory motion through the instruction set, and determines the state stability in the motion process.

[0062] The above technical solution of the present application has the following advantages compared with the prior art:

[0063] The present application proposes a bridge crane trajectory planning method, which real-time collects electromagnetic field intensity signal data of the main girder and the trolley through a sensor array arranged along the track, and performs denoising processing to improve signal quality; then, through feature extraction and signal decomposition technology, an association between component signal group and joint motion state is established, so as to accurately map the dynamic position coordinates of the main girder and the trolley, and form a high-precision dynamic position data record; this process effectively overcomes the error accumulation and environmental interference problem of traditional encoder or laser ranging, and improves the real-time and robustness of positioning.

[0064] Further, the method constructs a position change model based on high-precision dynamic position data, analyzes displacement and speed characteristics, adjusts motion state description in real time, and generates a multi-dimensional constraint condition set, providing an optimization basis for trajectory planning, through a dynamic feedback mechanism, which can continuously adjust the coordinated motion instruction to ensure the stability and accuracy of the crane during operation; compared with the traditional method, the technical solution realizes the transformation from static planning to dynamic optimization, can adaptively adjust the trajectory, reduces vibration and overshoot phenomenon, and improves the operation efficiency.

[0065] From the technical principle, the beneficial effects of the present application mainly reflect in the following aspects: first, the electromagnetic field intensity signal collection and denoising processing based on the sensor array improves the data reliability and reduces the influence of environmental interference; second, through signal decomposition and dynamic position mapping, high-precision real-time positioning is realized, avoiding the cumulative error problem of traditional methods; finally, the introduction of dynamic feedback mechanism enables real-time optimization of trajectory planning, thereby improving the motion stability and control accuracy of the crane, suitable for high-demand industrial application scenarios. BRIEF DESCRIPTION OF DRAWINGS

[0066] In order to make the content of the present application easier to be clearly understood, the following further detailed description of the present application is made according to the specific embodiments of the present application and in conjunction with the drawings, in which:

[0067] Figure 1 is a step flow chart of the bridge crane trajectory planning method of the present application;

[0068] Figure 2 is a step flow chart of the electromagnetic field intensity signal data acquisition and preprocessing method of the present application;

[0069] Figure 3 is a step flow chart of the electromagnetic field intensity signal feature decomposition and motion state correlation method of the present application;

[0070] Figure 4 is a step flow chart of the dynamic position coordinate mapping and optimization method of the present application;

[0071] Figure 5 is a step flow chart of the real-time motion state modeling and analysis method of the present application;

[0072] Figure 6 is a step flow chart of the multi-dimensional constraint condition generation and optimization method of the present application;

[0073] Figure 7 is a step flow chart of the trajectory planning optimization and generation method of the present application;

[0074] Figure 8 is a step flow chart of the dynamic feedback and trajectory adjustment method of the present application;

[0075] Figure 9 is a step flow chart of the coordinated motion control and stability determination method of the present application;

[0076] Figure 10 is a structural framework diagram of the bridge crane trajectory planning system of the present application. DETAILED DESCRIPTION

[0077] The application will be further described below in conjunction with the drawings and specific embodiments, so that those skilled in the art can better understand the application and implement it, but the embodiments are not limiting to the application.

[0078] Referring to Figure 1 As shown in the figure, the application discloses a bridge crane trajectory planning method, comprising the following steps:

[0079] S10, through the electromagnetic field intensity detection sensor array, collect the electromagnetic field intensity signal data in the process of the main beam and the trolley movement, and denoise the electromagnetic field intensity signal data to obtain the purified electromagnetic field intensity signal data set.

[0080] Referring to Figure 2 As shown in the figure, specifically comprising the following steps:

[0081] S101, a plurality of electromagnetic field intensity detection sensors are arranged along the track to form a sensor array covering the movement path of the main beam and the trolley, and real-time collection of electromagnetic field intensity signal data is performed when the main beam and the trolley move to obtain an initial electromagnetic field intensity signal set.

[0082] In a possible implementation, when a plurality of electromagnetic field intensity detection sensors are arranged along the track, a uniform arrangement mode can be selected, for example, a sensor is placed on each length of the track to ensure that the array covers the entire movement path of the main beam and the trolley. This can bring the beneficial effect of comprehensively capturing signal changes, because the electromagnetic field intensity fluctuates due to different positions during movement. Real-time collection of the initial electromagnetic field intensity signal set helps to provide a complete data basis for subsequent processing.

[0083] Exemplarily, if the total length of the track is a certain distance, the sensor array collects signals through wireless transmission to avoid missing key points, thereby improving the reliability of data collection.

[0084] S102, for the initial electromagnetic field intensity signal set, a preset filtering mechanism is used to process the signal to remove noise interference, and a purified electromagnetic field intensity signal data set is generated.

[0085] Specifically, the initial electromagnetic field intensity signal set is processed using a preset filtering mechanism, for example, low-pass filtering is used to remove high-frequency noise interference. This mechanism filters out environmental interference such as electromagnetic wave scattering by setting a frequency threshold, generates a purified electromagnetic field intensity signal data set, and this can bring the beneficial effect of improving signal purity. Because noise distorts the true electromagnetic field intensity value, the purified data more accurately reflects the actual movement state of the main beam and the trolley, supporting the accuracy of subsequent analysis.

[0086] S103, segment the purified electromagnetic field intensity signal dataset, divide the continuous signal data into multiple time segment signal subsets, so as to perform subsequent fine data calibration for each time segment signal subset.

[0087] For example, the data is divided into fixed length segments according to the movement speed. Such segmentation helps to calibrate each subset independently and avoid overall data deviation. This can bring the beneficial effect of fine processing, because different time segments may be affected by local interference, and segmentation can be more targeted for optimization to ensure the stability of the entire dataset.

[0088] S104, for the divided signal subsets, perform data smoothing processing one by one, further eliminate the micro fluctuations in the signal, and obtain the final optimized electromagnetic field intensity signal dataset, which is used for subsequent monitoring of the movement state of the girder and the trolley.

[0089] For example, the moving average method is used to eliminate the micro fluctuations. This processing is to smooth the curve by calculating the average value of adjacent data points to obtain the final optimized electromagnetic field intensity signal dataset. This can bring the beneficial effect of enhancing the smoothness of the signal, because the micro fluctuations may cause position inference error. After optimization, the dataset can provide more reliable input when used for monitoring the movement state of the girder and the trolley, supporting the accuracy improvement of the overall system.

[0090] From multiple aspects, the above arrangement and collection process and filtering mechanism support each other. For example, the comprehensive coverage of the sensor array ensures that the initial signal is rich, and the filtering purifies these signals to avoid noise affecting the accuracy of the segmentation processing. At the same time, the combination of segmentation and smoothing refines the data from the time dimension, and together supports the optimization of the final dataset. For example, in a high-speed motion scene, smoothing can reduce fluctuations caused by vibration, and segmentation allows adjustment for specific segments. Overall, these steps are progressive, forming a complete data processing chain, which is beneficial to achieve the goal of high-precision electromagnetic field intensity signal acquisition.

[0091] S20, according to the purified electromagnetic field intensity signal dataset, extract features and decompose into multiple component signal groups, and determine the association relationship between the component signal groups and the joint motion state.

[0092] Referring to Figure 3 , specifically comprising the following steps:

[0093] S201, from the purified electromagnetic field intensity signal dataset, extract time series features and spatial distribution features, for the signal nonlinear change caused by the multi-degree-of-freedom structure, adopt a preset signal decomposition method to split the purified electromagnetic field intensity signal dataset into multiple independent component signal groups, and preliminarily separate the signal components related to different motion dimensions.

[0094] Specifically, the process of extracting time sequence features and spatial distribution features from the purified electromagnetic field intensity signal dataset can be achieved by analyzing the intensity fluctuations of the signal at different time points and the distribution patterns of the spatial positions. This helps to identify the nonlinear changes caused by the multi-degree-of-freedom structure, thereby providing basic data support for subsequent decomposition and improving the accuracy of signal processing.

[0095] For example, separating the signal components related to the girder motion can effectively reduce interference and ensure that each component signal group focuses on a specific motion dimension, thereby bringing more reliable preliminary separation results.

[0096] S202, for the component signal group, combining the pre-established motion state association database, by comparing the fluctuation characteristics of each component signal group with the joint motion mode recorded in the database, the corresponding relationship between each component signal group and the specific joint motion state is determined, and a mapping table of component signals and motion states is formed.

[0097] Specifically, the corresponding relationship is determined by matching the fluctuation characteristics and the joint motion mode to form the mapping table. This can verify the association of signals and states from multiple aspects, such as comparing signal peak values and joint rotation modes in the trolley motion scenario, which can support the accuracy of the association and help avoid misjudgment and enhance the robustness of the system.

[0098] It should be noted that this comparison can also support each other from the time and space directions, such as the continuous fluctuations in time corresponding to the position changes in space, thereby forming a consistent mapping result together and bringing a more comprehensive state recognition effect.

[0099] S203, according to the mapping table of the component signal and the motion state, the change rule of each component signal group in the time dimension is extracted, and for the nonlinear characteristics presented in the change rule, a preset signal processing tool is used for smoothing processing to generate a smoothed component signal feature set, which is used for further verification of the subsequent association relationship.

[0100] It can be understood that this smoothing can remove noise mutations and make the change rule clearer, for example, when processing the girder vibration signal, the smoothed feature set is used to verify the association, which is beneficial to improve the reliability of subsequent refinement; this processing refines the nonlinear problem from multiple aspects, such as smoothing combined with historical data patterns, which can support the stability of the feature set and bring more accurate verification basis.

[0101] S204, using the smoothed component signal feature set, combining the mapping table of the joint motion state, by comparing the signal changes in the feature set with the dynamic changes of the motion state one by one, the association relationship between the component signal group and the joint motion state is refined to ensure the accuracy and reliability of the association relationship, and the feature decomposition and state association target of the purified electromagnetic field intensity signal dataset is completed.

[0102] For example, in a multi-degree-of-freedom structure, comparing the trolley displacement signal with the joint speed change can support each other from the speed and position sides to form a rigorous correlation logic, which is beneficial to complete the feature decomposition and state correlation objectives and improve the processing efficiency of the signal data set as a whole.

[0103] S30, according to the correlation between the component signal group and the joint motion state, mapping the dynamic position coordinates of the girder and the trolley to obtain high-precision dynamic position data records.

[0104] Referring to Figure 4 Specifically, the steps include the following steps:

[0105] S301, according to the correlation between the component signal group and the joint motion state, combining a pre-set electromagnetic field strength change and position coordinate mapping relationship database, using a nonlinear regression analysis method, mapping the component signal group to the dynamic position coordinates of the girder and the trolley to generate preliminary position data records.

[0106] Specifically, nonlinear regression analysis is a statistical method that fits a nonlinear function relationship between electromagnetic field strength and position, for example, using a polynomial function or a neural network model to establish a mapping, storing corresponding data of historical electromagnetic field strength and coordinates in the database, and optimizing parameters by least squares method to generate preliminary position data records. This can support the accuracy of position estimation from multiple sides and is beneficial to capture nonlinear dynamics in motion.

[0107] S302, for the preliminary position data records, applying Kalman filtering to smooth the dynamic position coordinates to eliminate random errors in the coordinate data to obtain high-precision dynamic position data records.

[0108] Specifically, Kalman filtering is a recursive estimation algorithm that combines prediction and measurement updates to eliminate random errors, for example, first predicting the current state, then correcting the prediction with the measurement, and iteratively calculating the state covariance to minimize the error to obtain high-precision dynamic position data records. This method supports each other from the perspectives of prediction accuracy and noise suppression, and is beneficial to provide more reliable position tracking, especially in a vibrating environment.

[0109] S303, according to the high-precision dynamic position data records, constructing a position change model based on time series to calculate the displacement and speed of the girder and the trolley at different time points to generate dynamic position data records.

[0110] For example, the displacement difference and the velocity vector are obtained by the difference method, and the model considers the autocorrelation of the time series to predict the trend, which can analyze the motion characteristics, is beneficial to real-time monitoring and optimization control, supports the overall system performance from the displacement stability and the speed consistency, and jointly forms the complete utilization of the position data.

[0111] S40, according to the high-precision dynamic position data record, a position change model based on time series is constructed, the displacement and speed characteristics of the main beam and the trolley at different time points are analyzed, and the real-time motion state description of each joint is determined.

[0112] Referring to Figure 5 , specifically comprising the following steps:

[0113] S401, from the high-precision dynamic position data record, the position coordinate information of the main beam and the trolley at multiple time points is obtained, and is arranged into a position data set based on time series, and the position data is arranged in time sequence, and a continuous time position sequence is formed.

[0114] Specifically, when obtaining the position coordinate information of the main beam and the trolley from the high-precision dynamic position data record, the scene of the crane in the actual hoisting operation can be considered, wherein the coordinates of the main beam can be collected in real time by a laser sensor, and the coordinates of the trolley are provided by an encoder. Arranging the position data set based on time series helps to capture the continuity of the motion, thereby providing a reliable basis for subsequent analysis. This processing method can improve the accuracy of data processing, because it ensures the integrity of the time sequence and avoids errors caused by data confusion.

[0115] S402, for the continuous time position sequence, the displacement between the main beam and the trolley at adjacent time points is calculated, the speed characteristics of each time point are further determined by the ratio of the displacement to the time interval, and the characteristic curve reflecting the change of displacement and speed with time is constructed.

[0116] Specifically, when calculating the displacement of the continuous time position sequence, for example, the displacement of the main beam from one time point to the next can be obtained by the coordinate difference, and the speed characteristics are obtained by dividing the displacement by the time interval. The role of constructing the characteristic curve is to intuitively reflect the change law, which not only facilitates the identification of abnormal fluctuations, but also provides data support for optimizing crane control, and the beneficial effect is to improve the accuracy of motion prediction.

[0117] S403, from the characteristic curve, the displacement change trend and the speed change trend of the main beam and the trolley at different time points are extracted, the motion state of each joint at the corresponding time point is judged whether it conforms to the expectation combined with the preset joint motion range parameter, and a preliminary motion state description is formed.

[0118] When extracting displacement trend and speed trend from the characteristic curve, in the example of the crane crossing the obstacle, if the trend shows that the speed suddenly increases, combined with the preset range of joint motion parameters, it can be judged whether the state is as expected, so as to discover potential risks early and form a preliminary motion state description. This method supports the operation stability from the safety angle, because it converts abstract data into operable judgment basis.

[0119] S404, for the preliminary motion state description, combined with the time series position data set, the real-time motion state of each joint is refined and labeled, and the specific motion posture of the girder and the trolley at each time point is determined, and finally the time series position change model construction and real-time motion state description of each joint are completed.

[0120] Specifically, when refining and labeling the preliminary motion state description, for example, combined with the position data set, the inclination posture of the girder and the lateral movement posture of the trolley at a specific time point are labeled, and finally the position change model construction is completed. This refinement can bring more accurate real-time motion state description, and the beneficial effect is to enhance the adaptability of the crane in complex environment, and through mutual support of multiple aspects such as displacement, speed and posture, the comprehensiveness and consistency of the description are ensured.

[0121] In practical application, the continuity of the above process is reflected from the direct input of the obtained position data set to the displacement calculation, to the trend extraction and refinement labeling, forming a closed loop thinking chain, for example, when the data set shows that the girder displacement is abnormal, the speed characteristic curve will highlight the problem, the trend judgment will confirm the state, and the refinement labeling will finally determine the posture. This supports each other from multiple directions such as data continuity and state judgment, and improves the practical value of the model together, which is beneficial to reduce operation failure.

[0122] S50, according to the real-time motion state description, a multi-dimensional constraint condition set containing speed limit and position boundary is generated, and a trajectory planning basic parameter set suitable for complex scene is obtained.

[0123] Referring to Figure 6 , specifically comprising the following steps:

[0124] S501, obtain the current speed and position data of the crane from the real-time motion state description, construct a constraint range containing the upper and lower limits of the speed for these data, and determine the position boundary range at the same time, forming an initial multi-dimensional constraint condition set.

[0125] Specifically, the current speed and position data of the crane are obtained from the real-time motion state description, for example, in a construction site scenario, when the crane is moving the boom at a medium speed, the speed value is several meters per second and the position coordinates are a certain point, which ensures that the subsequent constraint construction is based on accurate real-time information, and is beneficial to improve the real-time performance and safety of trajectory planning. A constraint range containing upper and lower speed limits is constructed for these data, and a position boundary range is determined to form an initial multi-dimensional constraint condition set, for example, the upper speed limit is set as a threshold to avoid collision, the lower limit ensures smooth operation, and the position boundary is limited within the site fence. This process defines a multi-dimensional space by integrating data points, which is beneficial to prevent accidents caused by the crane exceeding the safety area.

[0126] S502, scene adaptation adjustment is performed on the initial multi-dimensional constraint condition set, and the velocity constraint and position boundary are refined in combination with the obstacle distribution and operation area restriction in the complex scene to form an adapted constraint condition set.

[0127] For example, in a port environment with multiple obstacles, the speed constraint is adjusted to bypass the stacked containers, and the position boundary is refined to avoid the ship berthing area. This makes the constraint more suitable for the actual environment, and is beneficial to improve the operation efficiency and obstacle avoidance ability of the crane in a crowded scene.

[0128] S503, key parameters are extracted from the adapted constraint condition set to construct a trajectory planning basic parameter set suitable for complex scenes, ensuring that the parameter set can cover the multi-dimensional constraint requirements of speed limits and position boundaries.

[0129] For example, the upper speed limit parameter and boundary coordinate parameter are extracted to construct a parameter set for planning the path. This extraction process involves selecting core constraint elements, which is beneficial to simplify planning calculations and maintain comprehensive coverage, thereby making the trajectory more reliable.

[0130] S504, data integration is performed on the trajectory planning basic parameter set, and it is compared with the real-time motion state description. If it is found that the parameter set deviates from the actual state by more than a preset threshold, the parameter set is adjusted locally to ensure that the final generated parameter set is suitable for the trajectory planning requirements of complex scenes.

[0131] For example, in an outdoor scene with wind interference, when the speed deviation is large, the upper limit value is adjusted locally to adapt to the changes. This dynamically optimizes the parameters, which is beneficial to maintain motion stability and adapt to sudden environmental factors.

[0132] It should be noted that the above process forms a closed loop from data acquisition to final adjustment, for example, the initial constraint construction depends on real-time data, the adaptive adjustment expands its applicability, the parameter extraction and integration further strengthens the effectiveness of the constraint, this logical chain ensures that each link supports each other, which is beneficial to the accuracy and robustness of the overall trajectory planning, for example, in a multi-machine cooperative factory, this method can coordinate the speed and position of multiple cranes to avoid interference.

[0133] S60, according to the trajectory planning basic parameter set, applying a preset optimization calculation framework for iterative adjustment, outputting final trajectory planning scheme data.

[0134] Referring to Figure 7 The specific steps include the following steps:

[0135] S601, obtain initial parameters from a preset trajectory planning basic parameter set, including the motion range and speed constraint of the crane joint, construct a preliminary geometric path of the target trajectory based on the initial parameters, adjust the position and angle of the path node through iterative calculation, judge whether the adjusted path meets the preset convergence condition, if yes, generate preliminary trajectory planning scheme data.

[0136] Specifically, the process of obtaining initial parameters from a preset trajectory planning basic parameter set can be understood as first identifying the motion range and speed constraint of the crane joint, these parameters are used as basic input to ensure that the trajectory construction avoids joint overload or speed overrun, which can improve the safety of the trajectory, because if these constraints are ignored, the crane may shake or fail during movement.

[0137] Specifically, a straight line or curve segment connecting the starting point and the ending point is formed to form a path framework, then the position and angle of the path node are adjusted through iterative calculation, for example, the distance between nodes is gradually optimized during adjustment to make the path closer to the ideal curve, and it is judged whether the adjusted path meets the preset convergence condition, if yes, preliminary trajectory planning scheme data is generated, the effect of this iterative adjustment is to gradually approach the optimal path, which improves the efficiency of the trajectory and avoids energy waste caused by directly using a rough path.

[0138] S602, based on the preliminary trajectory planning scheme data, extract the coordinates and speed information of the path node, construct a path smoothness evaluation function, and smooth the path node through the least square method to generate smoothed trajectory planning scheme data.

[0139] Specifically, after extracting the coordinate and speed information of the path nodes based on the preliminary trajectory planning scheme data, the process of constructing the path smoothness evaluation function is actually to take the coordinate and speed as variables, and the evaluation function measures the curvature change between nodes. For example, if the path has sharp turns, the function value will be higher. Through least squares method, the path nodes are smoothed, and the specific method is to fit a curve with the smallest error to generate smoothed trajectory planning scheme data. The beneficial effect of this is to reduce crane vibration and ensure smooth movement, supporting the reliability of long-time operation.

[0140] S603, for the smoothed trajectory planning scheme data, calculate the time sequence distribution of each path node, combine the dynamics constraints of the crane to generate the coordinated motion time sequence of each joint, and determine that the time sequence meets the continuity requirement.

[0141] Specifically, for the smoothed trajectory planning scheme data, calculate the time sequence distribution of each path node, combine the dynamics constraints of the crane such as acceleration limit, generate the coordinated motion time sequence of each joint, and determine that the time sequence meets the continuity requirement, for example, ensure the continuity of velocity between adjacent nodes when distributing time, avoid sudden changes, which can enhance coordination because discontinuous time sequence may cause joint jamming and affect overall execution efficiency.

[0142] S604, according to the coordinated motion time sequence, generate the motion instruction set of each joint, the instruction set including the control parameters of joint angle and speed, used to drive the crane to execute trajectory motion, output the final trajectory planning scheme data.

[0143] Specifically, according to the coordinated motion time sequence, the process of generating the motion instruction set of each joint includes converting the time sequence into control parameters of joint angle and speed, which is used to drive the crane to execute trajectory motion, and finally output the final trajectory planning scheme data. The role of generating the instruction set is to convert abstract planning into executable commands, which brings improvement in actual motion accuracy, for example, in complex environment, the instruction set can support obstacle avoidance operation and support multi-task scene application.

[0144] It should be noted that the above steps support each other, for example, the adjustment of the preliminary path provides a basis for subsequent smoothing, and the smoothing data is input for time sequence calculation, ensuring the continuity of the whole process from parameter acquisition to instruction output, and achieving the goal of trajectory optimization together.

[0145] S70, according to the final trajectory planning scheme data, construct a dynamic feedback mechanism, compare the real-time collected electromagnetic field intensity signal with the trajectory planning scheme data, if the comparison result exceeds the preset deviation threshold, trigger the re-adjustment of the trajectory planning basic parameter set, get the updated trajectory planning data;

[0146] Referring to Figure 8 The method comprises the following steps in detail:

[0147] S701, acquire signal intensity data from real-time collected electromagnetic field strength signals, construct an electromagnetic field strength distribution dataset containing time stamps and position coordinates, and perform standardization processing on the electromagnetic field strength distribution dataset to generate a standardized electromagnetic field strength distribution matrix.

[0148] Specifically, during the process of real-time collection of electromagnetic field strength signals, key indicators such as radio wave intensity values captured by sensors are extracted from signal intensity data, which form an electromagnetic field strength distribution dataset in combination with time stamps and position coordinates. This construction method helps to capture the changes of signals in the time and space dimensions, thereby providing a complete basis for subsequent standardization processing.

[0149] It should be noted that the standardization processing involves mapping the intensity values in the dataset to a unified scale, such as by subtracting the mean and dividing by the standard deviation, to ensure that signals at different positions are comparable. The standardized electromagnetic field strength distribution matrix generated in this way can eliminate dimensional differences and be beneficial to improving the accuracy of comparison. The role of this processing is to enhance the robustness of the data, so that the system can more reliably identify signal patterns in complex environments, for example, in urban environments, signals may be disturbed by buildings. By standardizing, the true deviation rather than the noise influence can be highlighted, thereby improving the adaptability of trajectory planning.

[0150] S702, compare the standardized electromagnetic field strength distribution matrix with the final trajectory planning scheme data, calculate the deviation values of the two in the position coordinates, and generate a deviation distribution vector.

[0151] For example, for each coordinate point, the difference between the corresponding electromagnetic field strength value in the matrix and the preset value of the planning scheme is calculated, and is summarized into a deviation distribution vector. This vector represents the deviation degree of the actual signal distribution from the planning, which is beneficial to quantifying the inconsistency.

[0152] It should be noted that this comparison process can be verified from multiple aspects, such as checking signal fluctuations from a time sequence perspective or evaluating coverage uniformity from a spatial perspective. The deviation distribution vector generated in this way not only supports the judgment threshold, but also reveals potential reasons for environmental changes, for example, if the vector shows that the deviation in a specific area increases, it indicates that the planning scheme needs to be optimized to cope with sudden interference, thereby ensuring the real-time effectiveness of the trajectory.

[0153] S703, for the deviation distribution vector, judge whether it exceeds a preset deviation threshold, if it exceeds, adjust the speed and direction parameters in the trajectory planning basic parameter set according to the value of the deviation distribution vector, and generate an adjusted parameter set.

[0154] For example, reduce the speed parameter in the high deviation area to increase the signal acquisition time, or modify the direction parameter to avoid the weak electromagnetic field strength area, generate the adjusted parameter set. The effect of this adjustment is to dynamically respond to the actual signal feedback, which is beneficial to maintain the optimization state of the trajectory.

[0155] S704, according to the adjusted parameter set, apply the preset optimization calculation framework to iteratively update the trajectory planning data, and generate updated trajectory planning data.

[0156] For example, the framework evaluates the influence of parameters on the trajectory through multiple loops until the convergence condition is met to output the updated data. This iterative process includes initial parameter input, deviation minimization calculation and path re-planning steps in detail, which is beneficial to generate more accurate trajectory data.

[0157] S80, according to the updated trajectory planning data, generate a set of coordinated motion instructions for each joint, drive the crane to execute trajectory motion through the instruction set, and determine the state stability in the motion process.

[0158] Referring to Figure 9 as shown, specifically comprising the following steps:

[0159] S801, obtain the expected position and speed parameters of each joint from the updated trajectory planning data, and discretize the motion trajectory of each joint based on time sequence to generate an initial motion instruction set containing time stamp, position coordinate and velocity vector.

[0160] Specifically, when obtaining the expected position and speed parameters of each joint from the updated trajectory planning data, the actual operation of the crane in the warehouse environment can be considered, for example, the main beam needs to be moved horizontally to the specified shelf position, and the trolley needs to be adjusted vertically. By discretizing based on time sequence, the continuous trajectory can be decomposed into discrete points at multiple time stamps, thereby generating an initial motion instruction set, which is beneficial to improve the accuracy and executability of the instruction.

[0161] Specifically, this discretization process involves dividing the trajectory curve into equally spaced time periods, calculating the position coordinates and velocity vectors in each period, and ensuring that the instruction set covers the entire motion path, thereby avoiding calculation errors in continuous motion and bringing more stable control effect; if the trajectory planning data is updated to show that the main beam needs to be moved 5 meters in 10 seconds, it is discretized into one point per second, and instructions containing time stamps such as 0 seconds, 1 second, etc. are generated, which is beneficial to subsequent coordination.

[0162] It should be noted that this method can also adapt to sudden changes, such as when the data update reflects the presence of an obstacle, the parameters are adjusted in time to maintain safety.

[0163] S802, according to the initial motion instruction set, a constraint condition set containing speed limits and position boundaries is constructed for the kinematics of each joint of the crane, a coordinated motion instruction set is generated to ensure the synchronization and smoothness of each joint during motion.

[0164] For example, the inertia of the girder and the load response of the trolley can introduce speed limits such as a maximum of 2 meters per second and position boundaries such as not exceeding the end of the track, so that the coordinated motion instruction set is generated to ensure synchronization and prevent jitter caused by joint conflicts.

[0165] Wherein: the kinematics refers to the torque distribution of the joint when accelerating, the instruction is adjusted by the constraint set to match the speed of the girder and the trolley, for example, the trolley speed is not more than 50% of the girder, so as to realize smooth motion and prolong the service life of the equipment.

[0166] Specifically, in the bridge construction scene, if the instruction set is not constrained, the rapid movement of the girder may cause the trolley to be unstable, but after the boundary is added, the instruction set coordinates each joint to maintain overall balance, which is beneficial to reliability in complex environments.

[0167] S803, drive the control system of the crane through the coordinated motion instruction set, real-time collect the displacement and speed data of the girder and the trolley, and construct a position change model based on time series to record the real-time motion state of each joint.

[0168] Specifically, the displacement and speed of the girder and the trolley can be recorded every second during the operation of the crane, for example, the position change is captured using sensors to construct a position change model based on time series, which is a sequence of collected data organized in chronological order, recording displacement curves such as from the starting point to the end point, which is beneficial to capturing dynamic characteristics.

[0169] Wherein: the model construction process includes fitting the collected displacement data into a function form and analyzing the time point of the speed peak value to record the real-time motion state, which brings timely detection of abnormalities, such as sudden changes in speed indicating potential faults.

[0170] S804, extract the displacement and speed characteristics of the girder and the trolley from the position change model, combine the constraint condition set, judge the state stability during motion, and generate a trajectory planning basic parameter set suitable for complex scenes.

[0171] For example, extract the average speed of the girder and the maximum displacement deviation, if the deviation is less than the boundary threshold, it is determined to be stable, so as to generate a trajectory planning basic parameter set, which is beneficial to the application in complex scenes such as multi-obstacle environments.

[0172] Specifically, the extraction process involves calculating eigenvalues such as variance of velocity from the model sequence, combined with constraints such as position not exceeding 5 meters, determining stability, and generating a parameter set including optimized velocity upper limit, resulting in more robust planning.

[0173] Specifically, in factory automation, if the model shows that the trolley velocity characteristics exceed the limit and is determined to be unstable, the parameter set is adjusted to reduce the velocity to ensure state stability during motion.

[0174] Reference Figure 10 To achieve the above-mentioned bridge crane trajectory planning method, the present application also provides a bridge crane trajectory planning system for implementing the above-mentioned trajectory planning method, comprising:

[0175] The sensor array module detects the electromagnetic field intensity signal data of the main girder and the trolley during motion by an electromagnetic field intensity detection sensor array, and obtains the purified electromagnetic field intensity signal data set by denoising the electromagnetic field intensity signal data.

[0176] The feature extraction module extracts features and decomposes them into multiple component signal groups according to the purified electromagnetic field intensity signal data set, and determines the association between the component signal groups and the joint motion state.

[0177] The position mapping module maps the dynamic position coordinates of the main girder and the trolley according to the association between the component signal groups and the joint motion state, and obtains high-precision dynamic position data records.

[0178] The modeling analysis module constructs a position change model based on time series according to the high-precision dynamic position data records, analyzes the displacement and velocity characteristics of the main girder and the trolley at different time points, and determines the real-time motion state description of each joint.

[0179] The constraint generation module generates a multi-dimensional constraint condition set containing velocity limits and position boundaries according to the real-time motion state description, and obtains a trajectory planning basic parameter set suitable for complex scenarios.

[0180] The optimization adjustment module applies a preset optimization calculation framework to iteratively adjust the trajectory planning basic parameter set, and outputs the final trajectory planning scheme data.

[0181] The feedback adjustment module constructs a dynamic feedback mechanism according to the final trajectory planning scheme data, compares the real-time collected electromagnetic field intensity signal with the trajectory planning scheme data, and if the comparison result exceeds the preset deviation threshold, triggers the re-adjustment of the trajectory planning basic parameter set, and obtains the updated trajectory planning data.

[0182] The motion control module generates a coordinated motion instruction set of each joint according to the updated trajectory planning data, drives the crane to perform trajectory motion through the instruction set, and determines the state stability in the motion process.

[0183] Obviously, the above embodiments are only examples for clearly illustrating the present application, and are not intended to limit the embodiments. Based on the above description, other different forms of changes or variations can be made by those skilled in the art. Here, all the embodiments are not required to be exhausted, and the obvious changes or variations derived therefrom are still within the protection scope of the present application.

Claims

1. A method of trajectory planning for a bridge crane, characterized by: The method comprises the following steps: Collecting electromagnetic field intensity signal data of the main girder and the trolley during movement through an electromagnetic field intensity detection sensor array, performing denoising processing on the electromagnetic field intensity signal data to obtain a purified electromagnetic field intensity signal data set; According to the purified electromagnetic field intensity signal data set, extracting features and decomposing into multiple component signal groups, determining the correlation between the component signal groups and the joint motion state; According to the correlation between the component signal groups and the joint motion state, mapping the dynamic position coordinates of the main girder and the trolley to obtain high-precision dynamic position data records; According to the high-precision dynamic position data records, constructing a position change model based on time series, analyzing the displacement and speed characteristics of the main girder and the trolley at different time points, and determining the real-time motion state description of each joint; According to the real-time motion state description, generating a multi-dimensional constraint condition set containing speed limits and position boundaries to obtain a trajectory planning basic parameter set suitable for complex scenarios; According to the trajectory planning basic parameter set, applying a preset optimization calculation framework for iterative adjustment to output a final trajectory planning scheme data; According to the final trajectory planning scheme data, a dynamic feedback mechanism is constructed to compare the real-time collected electromagnetic field intensity signal with the trajectory planning scheme data. If the comparison result exceeds the preset deviation threshold, the trajectory planning basic parameter set is re-adjusted to obtain updated trajectory planning data. The method comprises the following steps: obtaining signal intensity data from the real-time collected electromagnetic field intensity signal, constructing an electromagnetic field intensity distribution data set containing time stamps and position coordinates, performing standardization processing on the electromagnetic field intensity distribution data set to generate a standardized electromagnetic field intensity distribution matrix; comparing the standardized electromagnetic field intensity distribution matrix with the final trajectory planning scheme data, calculating the deviation value of the two in the position coordinates, and generating a deviation distribution vector; judging whether the deviation distribution vector exceeds the preset deviation threshold. If it exceeds, adjust the speed and direction parameters in the trajectory planning basic parameter set according to the value of the deviation distribution vector to generate an adjusted parameter set; according to the adjusted parameter set, applying a preset optimization calculation framework to iteratively update the trajectory planning data to generate updated trajectory planning data; According to the updated trajectory planning data, a coordinated motion instruction set for each joint is generated to drive the crane to execute trajectory motion and determine the state stability during movement.

2. The bridge crane trajectory planning method of claim 1, wherein: Through an electromagnetic field intensity detection sensor array, electromagnetic field intensity signal data during movement of the main girder and the trolley is collected, and denoising processing is performed on the electromagnetic field intensity signal data to obtain a purified electromagnetic field intensity signal data set, which comprises: A plurality of electromagnetic field intensity detection sensors are arranged along the track to form a sensor array covering the movement path of the main girder and the trolley. When the main girder and the trolley move, electromagnetic field intensity signal data is collected in real time to obtain an initial electromagnetic field intensity signal set; For the initial electromagnetic field intensity signal set, a preset filtering mechanism is used to process the signal to remove noise interference and generate a purified electromagnetic field intensity signal data set; The purified electromagnetic field intensity signal data set is segmented, and continuous signal data is divided into multiple time segment signal subsets, so that subsequent fine data calibration is performed on each time segment signal subset; For the divided signal subsets, data smoothing is performed one by one to further eliminate small fluctuations in the signal, and an optimized electromagnetic field intensity signal data set is obtained for subsequent monitoring of the motion state of the girder and the trolley.

3. The bridge crane trajectory planning method of claim 1, wherein: According to the purified electromagnetic field intensity signal data set, features are extracted and decomposed into multiple component signal groups, and the correlation between the component signal groups and the joint motion state is determined, including: From the purified electromagnetic field intensity signal data set, time sequence features and spatial distribution features are extracted, and for the nonlinear changes in the signal caused by the multi-degree-of-freedom structure, the purified electromagnetic field intensity signal data set is split into multiple independent component signal groups, and the signal components related to different motion dimensions are preliminarily separated; For the component signal groups, the pre-established motion state correlation database is combined, the fluctuation characteristics of each component signal group are compared with the joint motion patterns recorded in the database, the correspondence between each component signal group and the specific joint motion state is determined, and a mapping table of component signals and motion states is formed; According to the mapping table of component signals and motion states, the change law of each component signal group in the time dimension is extracted, and for the nonlinear characteristics presented in the change law, a preset signal processing tool is used for smoothing processing to generate a smoothed component signal feature set, which is used for further verification of the correlation. Using the smoothed component signal feature set, combined with the mapping table of joint motion states, by comparing the signal changes in the feature set with the dynamic changes of the motion state one by one, the correlation between the component signal groups and the joint motion states is refined to ensure the accuracy and reliability of the correlation, and the feature decomposition and state correlation of the purified electromagnetic field intensity signal data set are completed.

4. The bridge crane trajectory planning method of claim 1, wherein: According to the correlation between the component signal groups and the joint motion state, the dynamic position coordinates of the girder and the trolley are mapped to obtain high-precision dynamic position data records, including: According to the correlation between the component signal groups and the joint motion state, combined with the preset electromagnetic field intensity change and position coordinate mapping relationship database, a nonlinear regression analysis method is used to map the component signal groups to the dynamic position coordinates of the girder and the trolley to generate preliminary position data records; For the preliminary position data records, Kalman filtering is applied to smooth the dynamic position coordinates to eliminate random errors in the coordinate data, and high-precision dynamic position data records are obtained; According to the high-precision dynamic position data records, a position change model based on time series is constructed to calculate the displacement and speed of the girder and the trolley at different time points to generate dynamic position data records.

5. The bridge crane trajectory planning method of claim 1, wherein: According to the high-precision dynamic position data records, a position change model based on time series is constructed to analyze the displacement and speed characteristics of the girder and the trolley at different time points to determine the real-time motion state description of each joint, including: From the high-precision dynamic position data record, the position coordinate information of the main beam and the trolley at multiple time points is obtained, and is arranged into a time sequence-based position data set. The position data is arranged in time sequence to form a continuous time position sequence; For the continuous time position sequence, the displacement between adjacent time points of the main beam and the trolley is calculated, and the speed characteristics of each time point are further determined by the ratio of displacement to time interval, and a characteristic curve reflecting the displacement and speed changes with time is constructed; From the characteristic curve, the displacement change trend and the speed change trend of the main beam and the trolley at different time points are extracted, and the motion state of each joint at the corresponding time point is judged whether it meets the expectation in combination with the preset joint motion range parameters, and a preliminary motion state description is formed; For the preliminary motion state description, the real-time motion state of each joint is refined and labeled in combination with the time sequence position data set, and the specific motion posture of the main beam and the trolley at each time point is determined, and finally the time sequence-based position change model construction and real-time motion state description of each joint are completed.

6. The bridge crane trajectory planning method of claim 1, wherein: According to the real-time motion state description, a multi-dimensional constraint condition set containing speed limit and position boundary is generated, and a trajectory planning basic parameter set suitable for complex scenes is obtained, including: From the real-time motion state description, the current speed and position data of the crane are obtained, and a constraint range containing the upper and lower limits of the speed is constructed for these data, and the position boundary range is determined, forming an initial multi-dimensional constraint condition set; The initial multi-dimensional constraint condition set is adjusted for scene adaptation, and the speed constraint and position boundary are refined in combination with the obstacle distribution and work area restriction in the complex scene, forming an adapted constraint condition set; Key parameters are extracted from the adapted constraint condition set to construct a trajectory planning basic parameter set suitable for complex scenes, ensuring that the parameter set can cover the multi-dimensional constraint requirements of speed limit and position boundary; For the trajectory planning basic parameter set, data integration is performed, and it is compared with the real-time motion state description. If it is found that the parameter set deviates from the actual state by more than a preset threshold, the parameter set is adjusted locally to ensure that the finally generated parameter set is suitable for the trajectory planning requirements of complex scenes.

7. The bridge crane trajectory planning method of claim 1, wherein: According to the trajectory planning basic parameter set, a preset optimization calculation framework is applied for iterative adjustment, and final trajectory planning scheme data is output, including: From the preset trajectory planning basic parameter set, initial parameters are obtained, including the motion range and speed constraint of the crane joint, a preliminary geometric path of the target trajectory is constructed based on the initial parameters, the position and angle of the path node are adjusted through iterative calculation, and it is judged whether the adjusted path meets the preset convergence condition. If it meets, the preliminary trajectory planning scheme data is generated; Based on the preliminary trajectory planning scheme data, the coordinates and speed information of the path node are extracted, a path smoothness evaluation function is constructed, and the path node is smoothed by least squares method to generate smoothed trajectory planning scheme data; For the smoothed trajectory planning scheme data, the time series assignment of each path node is calculated, the coordinated motion time series of each joint is generated combined with the dynamics constraints of the crane, and the continuity requirement of the time series is determined; According to the coordinated motion time series, a motion instruction set for each joint is generated, which includes joint angle and speed control parameters for driving the crane to execute trajectory motion, and the final trajectory planning scheme data is output.

8. The bridge crane trajectory planning method of claim 1, wherein: According to the updated trajectory planning data, the coordinated motion instruction set for each joint is generated, and the crane is driven to execute trajectory motion through the instruction set to determine the stability of the motion process, including: From the updated trajectory planning data, the expected position and speed parameters of each joint are obtained, the motion trajectory of each joint is discretized based on the time series, and an initial motion instruction set containing time stamp, position coordinates and velocity vector is generated; According to the initial motion instruction set, a constraint condition set containing speed limit and position boundary is constructed for the dynamics characteristics of each joint of the crane, and a coordinated motion instruction set is generated to ensure the synchronization and smoothness of each joint during the motion process; Through the coordinated motion instruction set, the control system of the crane is driven, the displacement and speed data of the main beam and the trolley are collected in real time, a position change model based on time series is constructed, and the real-time motion state of each joint is recorded; From the position change model, the displacement and speed characteristics of the main beam and the trolley are extracted, the stability of the motion process is judged combined with the constraint condition set, and a trajectory planning basic parameter set suitable for complex scenes is generated.

9. A bridge crane trajectory planning system for implementing the method of any of the preceding claims 1 to 8, characterized in that: It includes: A sensor array module acquires electromagnetic field intensity signal data during the motion of the main beam and the trolley through an electromagnetic field intensity detection sensor array, and obtains a purified electromagnetic field intensity signal data set by denoising the electromagnetic field intensity signal data; A feature extraction module extracts features and decomposes them into multiple component signal groups based on the purified electromagnetic field intensity signal data set, and determines the association between the component signal groups and the joint motion state; A position mapping module maps the dynamic position coordinates of the main beam and the trolley based on the association between the component signal groups and the joint motion state, and obtains high-precision dynamic position data records; A modeling analysis module constructs a position change model based on time series based on the high-precision dynamic position data records, analyzes the displacement and speed characteristics of the main beam and the trolley at different time points, and determines the real-time motion state description of each joint; A constraint generation module generates a multi-dimensional constraint condition set containing speed limit and position boundary based on the real-time motion state description, and obtains a trajectory planning basic parameter set suitable for complex scenes; An optimization adjustment module applies a preset optimization calculation framework to iteratively adjust the trajectory planning basic parameter set based on the trajectory planning basic parameter set, and outputs the final trajectory planning scheme data; A feedback adjustment module constructs a dynamic feedback mechanism based on the final trajectory planning scheme data, compares the real-time collected electromagnetic field intensity signal with the trajectory planning scheme data, and if the comparison result exceeds the preset deviation threshold, triggers the re-adjustment of the trajectory planning basic parameter set to obtain updated trajectory planning data. The motion control module generates a coordinated motion instruction set of each joint according to the updated trajectory planning data, drives the crane to perform trajectory motion through the instruction set, and determines the state stability in the motion process.

Images