Non-contact industrial device positioning and tracking method and apparatus

Abstract

The application relates to the field of target positioning and tracking, and relates to a non-contact industrial equipment positioning and tracking method and device.The method comprises the following steps: acquiring a reference position, the reference position being used for installing a reference label; constructing a base station coordinate network based on the reference position; acquiring UWB original signals corresponding to mobile labels based on the base station coordinate network, the mobile labels comprising anti-metal labels installed on a crown block and UWB labels worn by maintenance workshop staff; performing feature extraction on the UWB original signals corresponding to the mobile labels to obtain feature vectors; judging whether the signals need to be compensated according to the feature vectors to obtain a compensated data set; and positioning and tracking the maintenance workshop staff and the crown block according to the compensated data set.The application solves the problem of insufficient precision caused by inaccurate benchmarks and signal interference in traditional positioning, guarantees the positioning reliability in an industrial scene, and realizes centimeter-level positioning in a complex industrial environment.

Description

Technical Field

[0001] This invention relates to the field of target object positioning and tracking, and more specifically, to a non-contact method and apparatus for positioning and tracking industrial equipment. Background Technology

[0002] In railway locomotive depot maintenance operations, protective barriers are traditionally used to isolate overhead cranes and workers at the regional level to mitigate operational safety risks. However, this method has significant limitations: firstly, its fixed isolation format is ill-suited to the complex and ever-changing work areas and dynamic maintenance needs within the depot, easily creating blind spots or limiting operational flexibility; secondly, this method only achieves safety protection at the "regional division" level, failing to accurately perceive the real-time positions of personnel and overhead cranes. If personnel accidentally enter dangerous areas or the overhead crane deviates from its trajectory, timely warnings are difficult to obtain, resulting in limited safety protection effectiveness. Therefore, a non-contact industrial equipment positioning and tracking method is urgently needed to overcome the shortcomings of traditional isolation methods. Summary of the Invention

[0003] The purpose of this invention is to provide a non-contact industrial equipment positioning and tracking method and apparatus to improve the above-mentioned problems.

[0004] To achieve the above objectives, the embodiments of this application provide the following technical solutions:

[0005] On the one hand, embodiments of this application provide a non-contact industrial equipment positioning and tracking method, the method comprising:

[0006] Obtain a reference location, which is used to install the reference label;

[0007] A base station coordinate network is constructed based on the reference location, and the base station coordinate network includes the corrected base station coordinates;

[0008] Based on the base station coordinate network, the raw UWB signal corresponding to the mobile tag is obtained. The mobile tag includes an anti-metal tag installed on the overhead crane and a UWB tag worn by maintenance workshop staff.

[0009] Feature extraction is performed on the raw UWB signal corresponding to the mobile tag to obtain a feature vector;

[0010] Based on the feature vector, determine whether to compensate the signal to obtain the compensated dataset;

[0011] Based on the compensated dataset, the maintenance workshop staff and overhead cranes are located and tracked.

[0012] Secondly, embodiments of this application provide a non-contact industrial equipment positioning and tracking device, the device comprising:

[0013] The first acquisition module is used to acquire a reference position, which is used to install a reference tag.

[0014] A first processing module is used to construct a base station coordinate network based on the reference location, wherein the base station coordinate network includes corrected base station coordinates;

[0015] The second acquisition module is used to acquire the original UWB signal corresponding to the mobile tag based on the base station coordinate network. The mobile tag includes an anti-metal tag installed on the overhead crane and a UWB tag worn by maintenance workshop staff.

[0016] The second processing module is used to extract features from the original UWB signal corresponding to the mobile tag to obtain a feature vector.

[0017] The third processing module is used to determine whether to compensate the signal based on the feature vector, and to obtain the compensated dataset.

[0018] The fourth processing module is used to locate and track maintenance workshop workers and overhead cranes based on the compensated dataset.

[0019] Thirdly, embodiments of this application provide a non-contact industrial equipment positioning and tracking device, the device including a memory and a processor. The memory is used to store a computer program; the processor is used to execute the computer program to implement the steps of the above-described non-contact industrial equipment positioning and tracking method.

[0020] Fourthly, embodiments of this application provide a readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described non-contact industrial equipment positioning and tracking method.

[0021] The beneficial effects of this invention are as follows:

[0022] This invention first obtains reference locations and installs reference tags, then constructs a base station coordinate network based on these locations to provide a precise coordinate reference for positioning. Next, it acquires the raw UWB signals from the overhead crane's anti-metal tag and the worker's UWB tag based on this network, extracts feature vectors through feature extraction, and uses these vectors to judge and compensate for the signals to optimize data quality. Ultimately, it achieves positioning and tracking of both. This invention solves the problems of inaccurate references and insufficient accuracy caused by signal interference in traditional positioning, ensuring positioning reliability in industrial scenarios and achieving centimeter-level positioning in complex industrial environments. Furthermore, it eliminates the need for contactless installation, adapting to the dynamic operation needs of maintenance workshops and improving the flexibility and applicability of equipment and personnel positioning.

[0023] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing embodiments of the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings. Attached Figure Description

[0024] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a schematic diagram of the non-contact industrial equipment positioning and tracking method described in an embodiment of the present invention.

[0026] Figure 2 This is a schematic diagram of the non-contact industrial equipment positioning and tracking device described in an embodiment of the present invention.

[0027] Figure 3 This is a schematic diagram of the non-contact industrial equipment positioning and tracking device described in an embodiment of the present invention.

[0028] The diagram is labeled as follows: 800, Non-contact industrial equipment positioning and tracking device; 801, Processor; 802, Memory; 803, Multimedia component; 804, I / O interface; 805, Communication component; 901, First acquisition module; 902, First processing module; 903, Second acquisition module; 904, Second processing module; 905, Third processing module; 906, Fourth processing module. Detailed Implementation

[0029] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0030] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this invention, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0031] Example 1:

[0032] This embodiment provides a non-contact industrial equipment positioning and tracking method. It can be understood that in this embodiment, a scenario can be set up, such as an industrial scenario in a railway locomotive depot maintenance workshop where overhead cranes frequently lift parts and workers simultaneously carry out equipment maintenance operations.

[0033] See Figure 1 The figure shows that the method includes steps S1, S2, S3, S4, S5 and S6.

[0034] Step S1: Obtain a reference position, which is used to install a reference tag;

[0035] In this step, a 3D layout map containing elements such as load-bearing columns and overhead crane tracks is first drawn by scanning the workshop with a laser rangefinder. Areas with dense metal and vibration sources are marked. Then, signal propagation is simulated using simulation software to screen out areas with a line-of-sight ratio ≥95% and multipath signal strength ≤-90dBm. In a specific implementation, load-bearing steel columns with a height of 2.5-3m around the workshop and corner walls without equipment obstruction are selected to plan deployment points, ensuring that any three points are not collinear and the adjacent distance is ≥5m to meet the spatial positioning geometric constraints. Finally, the installation position of the reference tag is determined. In this way, areas with stable signal propagation, strong anti-interference ability and reliable structure can be screened out. At the same time, high-precision calibration ensures that the coordinates of the reference tag are accurate and stable in the long term, providing a reliable "anchor point" for base station calibration.

[0036] Step S2: Construct a base station coordinate network based on the reference location, wherein the base station coordinate network includes the corrected base station coordinates;

[0037] Step S2 further includes steps S21, S22, S23, S24, S25, and S26, which specifically include:

[0038] Step S21: Measure the reference position using the three-point intersection method to obtain the coordinates of the reference label center;

[0039] In this step, the three-dimensional coordinates of each reference position are measured using the cubic triangulation method, and the average value is taken to obtain the coordinates of the reference label center.

[0040] Step S22: Obtain the initial parameters of the UWB base station;

[0041] Step S23: Perform continuous ranging based on the reference tag center coordinates and the initial parameters of the UWB base station to obtain an effective ranging dataset;

[0042] Step S23 further includes steps S231, S232, S233, S234, and S235, which specifically include:

[0043] Step S231: Trigger continuous ranging based on the reference tag center coordinates and the initial parameters of the UWB base station and obtain the raw ranging data;

[0044] In this step, the raw ranging data includes ranging timestamp, received signal strength, signal-to-noise ratio, and signal waveform data.

[0045] Step S232: Process the original ranging data using a sliding window to obtain the first ranging dataset;

[0046] In this step, based on the measured distances sorted by timestamp in the original ranging data, a sliding window is constructed according to the rule of centering on the current k-th ranging value and including the two ranging values ​​before and after it. The median filtering algorithm is used to process the data in the window: first, the 5 measured distances are sorted, the maximum and minimum values ​​are removed, and the average of the remaining 3 values ​​is taken as the filtered data to obtain the first ranging dataset, which effectively filters the instantaneous ranging fluctuations caused by electromagnetic interference such as the start-up and shutdown of workshop equipment.

[0047] Step S233: Filter the first ranging dataset using the RSSI threshold to obtain the second ranging dataset;

[0048] In this step, the RSSI threshold is -85dBm. When RSSI < -85dBm, the signal is easily affected by metal reflections and electromagnetic interference in the workshop, resulting in a larger ranging error. Understandably, when RSSI ≥ -85dBm, the signal strength is considered to meet the standard, and the corresponding SNR value is further detected. If the SNR value ≥ 12dB, the corresponding data is retained.

[0049] Step S234: Send the data included in the second ranging dataset to the signal recognition model to obtain the recognition result, which includes the signal type and confidence level corresponding to each data point;

[0050] In this step, the training process of the signal recognition model is as follows: Between the calibrated reference label and the base station, three typical signal scenarios—LoS (unobstructed line-of-sight), NLoS (device / supporting column obstruction), and multipath (metal reflection)—are simulated, collecting over 100,000 samples. Each sample contains 12-dimensional features corresponding to the original ranging data (time-domain peak count, rise time, frequency-domain dominant frequency distribution, harmonic intensity, and TDoA, SNR, etc.) and manually labeled signal type labels (LoS / NLoS / multipath). The training set, validation set, and test set are divided in a 7:2:1 ratio. Next, a CNN-LSTM hybrid network structure is built, where the CNN layer extracts key local information from the signal features, and the LSTM layer captures the temporal correlation of the signal. A Dropout layer is introduced to prevent overfitting. During training, the cross-entropy loss function is used to measure the deviation between the model's predicted labels and the true labels. After each training round, the model's recognition accuracy is evaluated using a validation set. If the accuracy on the validation set does not improve for three consecutive rounds, an early stopping mechanism is triggered to avoid model overfitting. Finally, the model performance is verified using a test set, and the model parameters are adjusted using a confusion matrix until the model's recognition accuracy for the three types of signals is ≥99%. Training is then completed, and the model weights are saved to ensure that it can adapt to the signal type recognition requirements in the complex environment of the workshop. It can be understood that NLoS represents a scenario where the signal propagation path between the base station and the reference label is blocked by physical obstacles, and the signal cannot propagate directly through the "line-of-sight distance"; LoS represents a scenario where there are no obstructions between the base station and the reference label, and the signal propagates in a straight line.

[0051] Step S235: Filter the second ranging dataset according to the recognition results to obtain a valid ranging dataset.

[0052] In this step, if the identification result is LoS (unobstructed direct line of sight) and the confidence level is ≥0.95, it is marked as a valid distance measurement; if the identification result is NLoS or multipath or LoS confidence level <0.95, it is screened out.

[0053] Step S24: Calculate the theoretical distance based on the reference tag center coordinates and the base station initial coordinates;

[0054] Step S25: Calculate the distance error based on the theoretical distance and the effective distance measurement dataset;

[0055] Step S26: Use distance error to calibrate the base station coordinates to obtain the base station coordinate network.

[0056] In this step, the base station coordinates are used as the state variable and the distance error is used as the observation variable. The initial coordinates and state covariance matrix of the base station are first initialized by the EKF algorithm. Then, state prediction is performed, the observation matrix and Kalman gain are calculated, and finally the base station coordinates are corrected according to the distance error and iterated until the correction amount is ≤ ±1mm for 10 consecutive times, so as to realize the base station coordinate correction.

[0057] In this embodiment, multiple rounds of data screening are used to ensure the validity of ranging data, and error calibration is used to dynamically correct the base station coordinate deviation. This lays a precise spatial reference for subsequent mobile tag UWB signal reception, feature extraction, and positioning calculation, ensuring the centimeter-level accuracy of the entire positioning system from the source and overcoming the positioning reliability shortcomings caused by inaccurate base station references in traditional industrial scenarios.

[0058] Step S3: Based on the base station coordinate network, obtain the original UWB signal corresponding to the mobile tag. The mobile tag includes an anti-metal tag installed on the overhead crane and a UWB tag worn by maintenance workshop staff.

[0059] In this step, miniature UWB tags are worn by maintenance workshop workers, and anti-metal tags adapted to vibration environments are installed on the overhead crane.

[0060] Step S4: Extract features from the original UWB signal corresponding to the mobile tag to obtain a feature vector;

[0061] Step S4 further includes steps S41, S42, S43, and S44, which specifically include:

[0062] Step S41: Perform coherent demodulation on the raw UWB signal corresponding to the mobile tag to obtain the CIR curve;

[0063] In this step, the abstract raw signal is transformed into a visual curve that can intuitively reflect the propagation path, providing a concrete analysis carrier for subsequent feature extraction. Coherent demodulation of the raw UWB signal is a well-known technical solution in the art, so it will not be described in detail here.

[0064] Step S42: Extract features from the CIR curve to obtain first feature information, wherein the first feature information includes a numerical description of the CIR curve;

[0065] In the LosS scenario, the CIR curve exhibits a single main peak with weak side lobes, meaning the direct wave has the strongest energy, forming a sharp main peak, while the side lobes are only a small amount of environmental reflection, with weak energy and few in number. In the NLoS / multipath scenario, the CIR curve exhibits multiple peaks, meaning that after the direct wave is blocked, the reflected wave becomes the main component, and 2-3 side lobes of similar intensity appear after the main peak, or even the main peak shifts. Therefore, in this step, the main peak amplitude, main peak delay, number of side lobes, and total side lobe energy are extracted to numerically describe the CIR curve.

[0066] Step S43: Calculate the CIR curve to obtain second feature information, the second feature information including the quantization features of the CIR curve;

[0067] In this step, the second feature information includes signal strength, angle of arrival, main peak width, and multipath component ratio. It should be noted that the extraction process of multipath component ratio is to calculate the ratio of the total energy of the side lobes to the total energy of the main peak plus the side lobes in the CIR curve.

[0068] Step S44: Construct a feature vector based on the first feature information and the second feature information.

[0069] In this step, the first and second feature information are normalized and then concatenated into a single feature vector. The first feature information is crucial in directly reflecting the state of the signal propagation path, while the second feature information supplements the quantitative attributes of signal propagation. Only by combining the two can a comprehensive evaluation of both qualitative morphology and quantitative parameters be achieved. If only the first feature information is retained, although the signal path type can be determined, the interference intensity cannot be quantified, resulting in a lack of accurate basis for subsequent compensation judgments. If only the second feature information is retained, although quantitative data can be obtained, the root cause of poor signal quality cannot be traced, making targeted compensation difficult. The complementary choice of the two provides irreplaceable feature support for the accurate processing of UWB signals in complex industrial environments.

[0070] Step S5: Determine whether to compensate the signal based on the feature vector to obtain the compensated dataset;

[0071] Step S5 further includes steps S51, S52, and S53, which specifically include:

[0072] Step S51: Send the feature vector to the AI ​​model for evaluation to obtain the evaluation result, which includes the signal type and multipath interference level;

[0073] Step S52: Mark the confidence level corresponding to the signal according to the evaluation result;

[0074] Step S53: Determine whether to compensate the signal based on the confidence level to obtain the compensated dataset.

[0075] In this embodiment, AI evaluation and confidence level judgment are used to achieve accurate and intelligent decision-making for signal compensation, providing a high-quality dataset for subsequent positioning and tracking. This not only solves the problem of resource waste or insufficient compensation caused by traditional "one-size-fits-all" compensation, but also reduces the positioning error of the compensated data. At the same time, it reduces invalid compensation operations, reduces the power consumption of mobile tags, and balances positioning accuracy and device battery life.

[0076] Step S53 further includes steps S531 and S532, which specifically include:

[0077] Step S531: When the confidence level is determined to be high and the multipath interference level is less than or equal to level 2, no signal compensation is required.

[0078] Step S532: When the confidence level is determined to be low or the multipath interference level is greater than or equal to level 3, the signal is compensated.

[0079] In this step, signal compensation specifically involves: obtaining the interference type; when the interference type is metallic obstruction, compensation is performed using a first compensation calculation formula; when the interference type is personnel obstruction, compensation is performed using a second compensation calculation formula. The first compensation calculation formula includes:

[0080] C1 = 0.8 × L × (1 - R)

[0081] In the above formula, C1 represents the compensation value; L represents the multipath interference level; and R represents the signal strength.

[0082] The second compensation calculation formula includes:

[0083] C² = 0.5 × L × (1 - W)

[0084] In the above formula, C2 represents the compensation value; L represents the multipath interference level; and W represents the proportion of multipath components.

[0085] Step S6: Locate and track the maintenance workshop staff and overhead cranes based on the compensated dataset.

[0086] Step S6 further includes steps S61, S62, S63, S64, and S65, which specifically include:

[0087] Step S61: Obtain first information and second information. The first information includes IMU data corresponding to the staff and the crane, and the second information includes image information collected by the camera installed on the top of the crane.

[0088] Step S62: Standardize the first information, the second information, and the compensated dataset to obtain a standardized dataset;

[0089] In this step, the IMU data is statically initialized for 3 seconds to remove zero bias and denoised using Kalman filtering to obtain clean acceleration and angular velocity. The visual data is processed using the ORB-SLAM3 algorithm to extract feature points, and frames with ≥20 matching points are selected to calculate the relative pose increment. Finally, using the 200Hz IMU data timestamp as a reference, the 10Hz UWB data and 30Hz visual data are synchronized to the same time axis through interpolation, resulting in a standardized dataset that includes aligned UWB ranging, denoised IMU data, effective visual increment, and corresponding timestamps.

[0090] Step S63: Construct a factor graph structure based on the standardized dataset, wherein the factor graph structure includes constraint factors;

[0091] In this step, the first frame of valid UWB data is extracted from the standardized dataset to calculate the initial pose, which is used as the starting node of the factor graph. A set of denoised IMU data is read from the standardized dataset, and motion factors are constructed based on the physical motion model to constrain the relationship between adjacent nodes, thus obtaining the motion factors of associated adjacent nodes. A set of aligned UWB data is read from the standardized dataset, and distance-constrained UWB observation factors are constructed by combining confidence weights. The coordinates of the current node and the base station are associated, and the UWB observation factors of the bound node and the base station are output. A set of visual incremental data is read from the standardized dataset, and relative pose-constrained visual observation factors are constructed to constrain the relationship between adjacent nodes, thus outputting the visual observation factors of associated adjacent nodes. The above nodes and various factors are integrated to obtain a factor graph structure including motion factors, UWB observation factors, and visual observation factors. This application constructs a factor graph structure with constrained factors based on the standardized dataset, transforming multi-source data into pose constraint relationships, and realizing deep coupling of UWB, IMU, and visual data.

[0092] Step S64: Solve the global error residual minimization problem using the Gauss-Newton method based on the factor graph structure to obtain the optimized pose sequence;

[0093] In this step, an initial guess is made for each pose node in the factor graph based on the IMU motion model to obtain the initial pose of the node. Based on the initial pose of the node, the motion factor residual, UWB observation factor residual, and visual observation factor residual are calculated respectively. The residuals are linearized at the initial pose, the Jacobian matrix is ​​constructed, and the pose correction is solved. The initial pose is updated based on the pose correction, and the residual calculation and incremental solution process is repeated until the residual change is less than a preset threshold, and the optimized pose sequence is output. The problem of minimizing the global error residual of the factor graph is solved by using the Gauss-Newton method. Iterative optimization can reduce the influence of the superposition of errors from multiple sources and improve the accuracy of pose calculation.

[0094] Step S65: Obtain the real-time six-degree-of-freedom poses of the workers and the overhead crane based on the optimized pose sequence.

[0095] In this step, the pose data of the workers' labels are retained in the Z-axis range of 1.5-2.0m, and the pose data of the crane labels are retained in the Z-axis range of 3-8m.

[0096] In this embodiment, not only is centimeter-level positioning accuracy achieved, but also, through multi-source data complementarity, positioning continuity can still be maintained through IMU and visual data when UWB signals are blocked, greatly reducing the positioning interruption rate.

[0097] Following step S65, steps S66, S67, S68, and S69 are further included, specifically comprising:

[0098] Step S66: Send the real-time six-degree-of-freedom poses of the staff and the overhead crane to the digital twin platform for dynamic model binding to obtain a synchronized image;

[0099] In this step, the digital twin platform is a 1:1 replica of the maintenance workshop in three dimensions, including static elements such as crane tracks, load-bearing columns, maintenance platforms, and safety passages. The platform coordinate system is fully aligned with the base station coordinate network. Personnel poses are bound to the three-dimensional personnel model in the digital twin, and crane poses are bound to the crane hook model, achieving a visualization effect of actual movement-virtual synchronization.

[0100] Step S67: Calculate the real-time distance between the worker and the overhead crane based on the worker's center of gravity posture and the crane hook posture in the synchronized image;

[0101] In this step, the digital twin platform calculates the shortest distance between the center of the overhead crane hook and the center of gravity of the worker in real time based on the three-dimensional distance formula between the two points, thus obtaining the real-time distance between the worker and the overhead crane.

[0102] Step S68: Determine the risk level based on the real-time distance;

[0103] In this step, when the real-time distance between the staff and the crane is greater than the first safety threshold, it is a safe level; when the real-time distance between the staff and the crane is less than or equal to the first safety threshold but greater than the second safety threshold, it is a warning level; when the real-time distance between the staff and the crane is less than or equal to the second safety threshold, it is an emergency level. This application does not limit the first safety threshold and the second safety threshold. The first safety threshold and the second safety threshold are different depending on the density of the staff.

[0104] Step S69: Send control instructions and warning instructions according to the risk level. The control instructions are used to reduce the speed of the overhead crane, and the warning instructions are used to warn the staff.

[0105] In this step, when the risk level is at the warning level, a control command is sent to the overhead crane to reduce its speed, and the worker's work badge vibrates at a low frequency and a yellow light indicates a warning; when the risk level is at the emergency level, a stop command is issued, and the work badge vibrates at a high frequency, a red light is emitted, and a buzzer sounds to prompt the worker to evacuate immediately.

[0106] In this embodiment, proactive safety protection for maintenance workshop personnel and overhead cranes is achieved through digital twin visualization and hierarchical linkage control. This not only solves the shortcomings of traditional fixed guardrails and lack of early warning, but also assists in analyzing high-risk areas and time periods based on historical data from synchronized images, optimizing the efficiency of subsequent workshop operations. At the same time, the hierarchical early warning mechanism reduces unnecessary overhead crane downtime, improves workshop operation efficiency, and balances safety protection and production efficiency.

[0107] Example 2:

[0108] like Figure 2 As shown, this embodiment provides a non-contact industrial equipment positioning and tracking device. The device includes a first acquisition module 901, a first processing module 902, a second acquisition module 903, a second processing module 904, a third processing module 905, and a fourth processing module 906, specifically including:

[0109] The first acquisition module 901 is used to acquire a reference position, the reference position being used to install a reference tag;

[0110] The first processing module 902 is used to construct a base station coordinate network based on the reference location, wherein the base station coordinate network includes corrected base station coordinates.

[0111] The second acquisition module 903 is used to acquire the original UWB signal corresponding to the mobile tag based on the base station coordinate network. The mobile tag includes an anti-metal tag installed on the overhead crane and a UWB tag worn by maintenance workshop staff.

[0112] The second processing module 904 is used to extract features from the original UWB signal corresponding to the mobile tag to obtain a feature vector.

[0113] The third processing module 905 is used to determine whether to compensate the signal based on the feature vector, and to obtain the compensated dataset.

[0114] The fourth processing module 906 is used to locate and track maintenance workshop workers and overhead cranes based on the compensated dataset.

[0115] In one specific embodiment of this disclosure, the first processing module further includes a first processing unit, a first acquisition unit, a second processing unit, a third processing unit, a fourth processing unit, and a fifth processing unit, specifically including:

[0116] The first processing unit is used to measure the reference position using the three-point intersection method to obtain the coordinates of the reference label center.

[0117] The first acquisition unit is used to acquire the initial parameters of the UWB base station;

[0118] The second processing unit is used to perform continuous ranging based on the reference tag center coordinates and the initial parameters of the UWB base station to obtain an effective ranging dataset.

[0119] The third processing unit is used to calculate the theoretical distance based on the reference tag center coordinates and the base station initial coordinates;

[0120] The fourth processing unit is used to calculate the distance error based on the theoretical distance and the effective ranging dataset;

[0121] The fifth processing unit is used to calibrate the base station coordinates using distance errors to obtain the base station coordinate network.

[0122] In one specific embodiment of this disclosure, the second processing unit further includes a second acquisition unit, a sixth processing unit, a seventh processing unit, an eighth processing unit, and a ninth processing unit, specifically including:

[0123] The second acquisition unit is used to trigger continuous ranging based on the reference tag center coordinates and the initial parameters of the UWB base station and acquire raw ranging data.

[0124] The sixth processing unit is used to process the original ranging data using a sliding window to obtain the first ranging dataset;

[0125] The seventh processing unit is used to filter the first ranging dataset using the RSSI threshold to obtain the second ranging dataset.

[0126] The eighth processing unit is used to send the data included in the second ranging dataset to the signal recognition model to obtain the recognition result, which includes the signal type and confidence level corresponding to each data point;

[0127] The ninth processing unit is used to filter the second ranging dataset based on the recognition result to obtain a valid ranging dataset.

[0128] In one specific embodiment of this disclosure, the second processing module further includes a tenth processing unit, an eleventh processing unit, a twelfth processing unit, and a thirteenth processing unit, specifically including:

[0129] The tenth processing unit is used to coherently demodulate the raw UWB signal corresponding to the mobile tag to obtain the CIR curve;

[0130] The eleventh processing unit is used to extract features from the CIR curve to obtain first feature information, wherein the first feature information includes a numerical description of the CIR curve.

[0131] The twelfth processing unit is used to calculate the CIR curve to obtain second feature information, the second feature information including the quantization features of the CIR curve;

[0132] The thirteenth processing unit is used to construct a feature vector based on the first feature information and the second feature information.

[0133] It should be noted that the specific manner in which each module performs its operation in the apparatus described in the above embodiments has been described in detail in the embodiments of the method, and will not be elaborated here.

[0134] Example 3:

[0135] Corresponding to the above method embodiments, this embodiment also provides a non-contact industrial equipment positioning and tracking device. The non-contact industrial equipment positioning and tracking device described below and the non-contact industrial equipment positioning and tracking method described above can be referred to each other.

[0136] Figure 3 This is a block diagram illustrating a non-contact industrial equipment positioning and tracking device 800 according to an exemplary embodiment. Figure 3 As shown, the non-contact industrial equipment positioning and tracking device 800 may include: a processor 801 and a memory 802. The non-contact industrial equipment positioning and tracking device 800 may also include one or more of a multimedia component 803, an I / O interface 804, and a communication component 805.

[0137] The processor 801 controls the overall operation of the non-contact industrial equipment positioning and tracking device 800 to complete all or part of the steps in the aforementioned non-contact industrial equipment positioning and tracking method. The memory 802 stores various types of data to support the operation of the non-contact industrial equipment positioning and tracking device 800. This data may include, for example, instructions for any application or method operating on the non-contact industrial equipment positioning and tracking device 800, as well as application-related data such as contact data, sent and received messages, images, audio, video, etc. The memory 802 can be implemented using any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The multimedia component 803 may include a screen and an audio component. The screen may be, for example, a touchscreen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signals may be further stored in the memory 802 or transmitted via the communication component 805. The audio component also includes at least one speaker for outputting audio signals. I / O interface 804 provides an interface between processor 801 and other interface modules, such as keyboards, mice, and buttons. These buttons can be virtual or physical. Communication component 805 is used for wired or wireless communication between the non-contact industrial equipment positioning and tracking device 800 and other devices. Wireless communication includes Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, or 4G, or a combination thereof. Therefore, the corresponding communication component 805 may include a Wi-Fi module, a Bluetooth module, or an NFC module.

[0138] In an exemplary embodiment, the non-contact industrial equipment positioning and tracking device 800 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the non-contact industrial equipment positioning and tracking method described above.

[0139] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, which, when executed by a processor, implement the steps of the aforementioned non-contact industrial equipment positioning and tracking method. For example, the computer-readable storage medium may be the aforementioned memory 802 including program instructions, which may be executed by the processor 801 of the non-contact industrial equipment positioning and tracking device 800 to complete the aforementioned non-contact industrial equipment positioning and tracking method.

[0140] Example 4:

[0141] Corresponding to the above method embodiments, this embodiment also provides a readable storage medium. The readable storage medium described below can be referred to in conjunction with the non-contact industrial equipment positioning and tracking method described above.

[0142] A readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the non-contact industrial equipment positioning and tracking method described in the above method embodiments.

[0143] Specifically, the readable storage medium can be a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, or any other readable storage medium capable of storing program code.

[0144] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0145] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A non-contact industrial equipment positioning and tracking method, characterized in that, include: Obtain a reference location, which is used to install the reference label; A base station coordinate network is constructed based on the reference location, and the base station coordinate network includes the corrected base station coordinates; Based on the base station coordinate network, the raw UWB signal corresponding to the mobile tag is obtained. The mobile tag includes an anti-metal tag installed on the overhead crane and a UWB tag worn by maintenance workshop staff. Feature extraction is performed on the raw UWB signal corresponding to the mobile tag to obtain a feature vector; Based on the feature vector, determine whether to compensate the signal to obtain the compensated dataset; Based on the compensated dataset, the maintenance workshop staff and overhead cranes are located and tracked. The process of constructing a base station coordinate network based on the reference location includes: The reference position was measured using the three-point intersection method to obtain the coordinates of the reference label center. Obtain the initial parameters of the UWB base station; Continuous ranging is performed based on the reference tag center coordinates and the initial parameters of the UWB base station to obtain an effective ranging dataset. The theoretical distance is calculated based on the reference tag center coordinates and the base station initial coordinates; Calculate the distance error based on the theoretical distance and the effective ranging dataset; The base station coordinates are calibrated using distance error to obtain the base station coordinate network; Feature extraction is performed on the raw UWB signal corresponding to the mobile tag to obtain a feature vector, including: The raw UWB signal corresponding to the mobile tag is coherently demodulated to obtain the CIR curve; Feature extraction is performed on the CIR curve to obtain first feature information, which includes a numerical description of the CIR curve. The CIR curve is calculated to obtain second feature information, which includes the quantization features of the CIR curve. A feature vector is constructed based on the first feature information and the second feature information.

2. The non-contact industrial equipment positioning and tracking method according to claim 1, characterized in that, Continuous ranging is performed based on the reference tag center coordinates and the initial parameters of the UWB base station, including: Continuous ranging is triggered based on the reference tag center coordinates and the initial parameters of the UWB base station, and raw ranging data is obtained. The original ranging data is processed using a sliding window to obtain the first ranging dataset; The first ranging dataset is filtered using an RSSI threshold to obtain the second ranging dataset; The data included in the second ranging dataset are sent to the signal recognition model to obtain the recognition result, which includes the signal type and confidence level corresponding to each data point. The second ranging dataset is filtered based on the recognition results to obtain a valid ranging dataset.

3. The non-contact industrial equipment positioning and tracking method according to claim 1, characterized in that, Determining whether to compensate the signal based on the feature vector includes: The feature vector is sent to an AI model for evaluation to obtain an evaluation result, which includes the signal type and multipath interference level. The confidence level of the signal is marked according to the evaluation results; Based on the confidence level, determine whether to compensate the signal to obtain the compensated dataset.

4. The non-contact industrial equipment positioning and tracking method according to claim 1, characterized in that, Based on the compensated dataset, the maintenance workshop staff and overhead cranes are located and tracked, including: Acquire first information and second information, wherein the first information includes IMU data corresponding to the staff and the overhead crane, and the second information includes image information collected by a camera installed on the top of the overhead crane; The first information, the second information, and the compensated dataset are standardized to obtain a standardized dataset. A factor graph structure is constructed based on the standardized dataset, and the factor graph structure includes constraint factors. Based on the aforementioned factor graph structure, the global error residual minimization problem is solved using the Gauss-Newton method to obtain the optimized pose sequence. The real-time six-degree-of-freedom poses of the staff and the overhead crane are obtained based on the optimized pose sequence.

5. The non-contact industrial equipment positioning and tracking method according to claim 4, characterized in that, After obtaining the real-time six-DOF poses of the worker and the overhead crane based on the optimized pose sequence, the following steps are also included: The real-time six-DOF poses of the staff and the overhead crane are sent to the digital twin platform for dynamic model binding to obtain a synchronized view; The real-time distance between the worker and the overhead crane is calculated based on the worker's center of gravity posture and the crane hook posture in the synchronized image. The risk level is determined based on the real-time distance. Control commands and warning commands are sent according to the risk level. The control commands are used to reduce the speed of the overhead crane, and the warning commands are used to warn the staff.

6. A non-contact industrial equipment positioning and tracking device, characterized in that, include: The first acquisition module is used to acquire a reference position, which is used to install a reference tag. A first processing module is used to construct a base station coordinate network based on the reference location, wherein the base station coordinate network includes corrected base station coordinates; The second acquisition module is used to acquire the original UWB signal corresponding to the mobile tag based on the base station coordinate network. The mobile tag includes an anti-metal tag installed on the overhead crane and a UWB tag worn by maintenance workshop staff. The second processing module is used to extract features from the original UWB signal corresponding to the mobile tag to obtain a feature vector. The third processing module is used to determine whether to compensate the signal based on the feature vector, and to obtain the compensated dataset. The fourth processing module is used to locate and track the maintenance workshop staff and overhead cranes based on the compensated dataset. The first processing module includes: The first processing unit is used to measure the reference position using the three-point intersection method to obtain the coordinates of the reference label center. The first acquisition unit is used to acquire the initial parameters of the UWB base station; The second processing unit is used to perform continuous ranging based on the reference tag center coordinates and the initial parameters of the UWB base station to obtain an effective ranging dataset. The third processing unit is used to calculate the theoretical distance based on the reference tag center coordinates and the base station initial coordinates; The fourth processing unit is used to calculate the distance error based on the theoretical distance and the effective ranging dataset; The fifth processing unit is used to calibrate the base station coordinates using distance errors to obtain the base station coordinate network; The second processing module includes: The tenth processing unit is used to coherently demodulate the raw UWB signal corresponding to the mobile tag to obtain the CIR curve; The eleventh processing unit is used to extract features from the CIR curve to obtain first feature information, wherein the first feature information includes a numerical description of the CIR curve. The twelfth processing unit is used to calculate the CIR curve to obtain second feature information, the second feature information including the quantization features of the CIR curve; The thirteenth processing unit is used to construct a feature vector based on the first feature information and the second feature information.

7. The non-contact industrial equipment positioning and tracking device according to claim 6, characterized in that, The second processing unit includes: The second acquisition unit is used to trigger continuous ranging based on the reference tag center coordinates and the initial parameters of the UWB base station and acquire raw ranging data. The sixth processing unit is used to process the original ranging data using a sliding window to obtain the first ranging dataset; The seventh processing unit is used to filter the first ranging dataset using the RSSI threshold to obtain the second ranging dataset. The eighth processing unit is used to send the data included in the second ranging dataset to the signal recognition model to obtain the recognition result, which includes the signal type and confidence level corresponding to each data point; The ninth processing unit is used to filter the second ranging dataset based on the recognition result to obtain a valid ranging dataset.

Images