Anti-metal interference industrial UWB positioning calibration method, device, equipment and medium

Abstract

The invention relates to an anti-metal interference industrial UWB positioning calibration method, device, equipment and medium, and the method comprises the steps: calculating an inertia prediction error through inertia prediction data and inertia measurement data of an ultra-wideband signal, inputting the positioning prediction data to a transmission state region model, and outputting a corresponding label region non-line-of-sight weight; then, performing weighted summation on the inertial prediction error and the non-line-of-sight weight of the label area, and performing calculation to obtain the non-line-of-sight weight; and finally, when the non-line-of-sight weight is greater than a preset non-line-of-sight threshold, performing non-line-of-sight compensation correction on the ultra-wideband signal through a dynamic attenuation compensation model to obtain an ultra-wideband positioning calibration signal, thereby effectively solving the problem of signal interference caused by a metal obstacle, compensating phase distortion caused by the metal obstacle, and recovering the integrity of the signal.

Description

Technical Field

[0001] This application relates to the field of industrial UWB positioning calibration technology, and in particular to an industrial UWB positioning calibration method, apparatus, equipment and medium that is resistant to metal interference. Background Technology

[0002] With the development of industrial Internet of Things (IIoT) technology, ultra-wideband (UWB) positioning technology has emerged. With its nanosecond-level narrow pulse characteristics, UWB technology has centimeter-level positioning accuracy, strong penetration capability, and anti-multipath interference potential, and has become a core supporting technology for applications such as equipment tracking, personnel navigation, and intelligent manufacturing in industrial scenarios.

[0003] However, current UWB positioning methods have significant limitations in metal-rich industrial environments: metal obstacles cause strong reflections of UWB signals, generating numerous coherent multipath components that make it difficult to distinguish between direct and reflected signals, and traditional threshold screening methods are prone to misjudging valid signals. Simultaneously, the electromagnetic properties of metal surfaces alter the signal phase, leading to phase distortion and directly affecting the measurement accuracy of Time of Arrival (TOA) / Time Difference of Arrival (TDOA), thus causing positioning errors. Furthermore, while adding anchor points can improve redundancy, it significantly increases deployment costs and the probability of signal collisions, and cannot fundamentally solve the systemic positioning errors caused by metal interference. Summary of the Invention

[0004] Therefore, it is necessary to provide an industrial UWB positioning calibration method, apparatus, equipment, and medium that can compensate for signal attenuation caused by metal reflection and phase distortion caused by metal obstacles, thereby addressing the aforementioned technical problems.

[0005] Firstly, this application provides an industrial UWB positioning calibration method resistant to metal interference, comprising: The positioning prediction data, inertial prediction data, and inertial measurement data of the ultra-wideband signal are acquired, and the difference between the inertial prediction data and the inertial measurement data is calculated as the inertial prediction error. The positioning prediction data is input into the transmission state region model, and the corresponding label region non-line-of-sight weights are output; the transmission state region model is used to characterize the ultra-wideband reflection path of the region corresponding to the positioning prediction data. The inertial prediction error is weighted and summed with the non-line-of-sight weight of the label region to obtain the non-line-of-sight weight; If the non-line-of-sight weight is greater than the preset non-line-of-sight threshold, the ultra-wideband signal is compensated and corrected for non-line-of-sight through a dynamic attenuation compensation model to obtain an ultra-wideband positioning calibration signal.

[0006] Furthermore, the process of constructing the transmission state region model includes: Step 1: Obtain obstacle distribution information, ultra-wideband base station positioning information, and the preset maximum number of reflections, and initialize the number of iterative reflections to zero; Step 2: Combining obstacle distribution information with the location of the ultra-wideband base station, set the minimum number of reflections required for the ultra-wideband signal to reach the ultra-wideband base station in the industrial positioning area to be equal to the label area category information and label area non-line-of-sight weight of the area corresponding to the number of iterative reflections, and increment the number of iterative reflections by one. Step 3: When the number of iterative reflections is less than or equal to the maximum number of reflections, step 2 is executed iteratively; when the number of iterative reflections is greater than the maximum number of reflections, the label area category information and the label area non-line-of-sight weight of the remaining area in the industrial positioning area are set as the blind zone label area category information and the blind zone label area non-line-of-sight weight, respectively, to obtain the transmission state area model.

[0007] Furthermore, the step of setting the minimum number of reflections required for the ultra-wideband signal to reach the ultra-wideband base station in the industrial positioning area to be equal to the number of iterative reflections, along with the label area category information and label area non-line-of-sight weight, includes: When the number of iteration reflections is zero, the category information of the label region corresponding to the region where the minimum number of reflections equals the number of iteration reflections is set as the category information of the line-of-sight region corresponding to the number of iteration reflections being zero; the non-line-of-sight weight of the label region corresponding to the region where the minimum number of reflections equals the number of iteration reflections is set as the non-line-of-sight weight of the line-of-sight region corresponding to the number of iteration reflections being zero. When the number of iterative reflections is greater than zero and less than or equal to the maximum number of reflections, the label region category information of the region corresponding to the minimum number of reflections equal to the number of iterative reflections is set as the category information of the non-line-of-sight region corresponding to the nth reflection when the number of iterative reflections is n; the label region non-line-of-sight weight of the region corresponding to the minimum number of reflections equal to the number of iterative reflections is set as the non-line-of-sight weight of the non-line-of-sight region corresponding to the nth reflection when the number of iterative reflections is n; the value of n is from 1 to the maximum number of reflections.

[0008] Furthermore, the step of performing non-line-of-sight compensation correction on the ultra-wideband signal using a deep learning-based dynamic attenuation compensation model to obtain an ultra-wideband positioning calibration signal includes: Based on the transmission state region model, obtain the label region category information corresponding to the location prediction data; Based on the label region category information corresponding to the positioning prediction data, and combined with the dynamic attenuation compensation model, the ultra-wideband signal is attenuated to obtain an ultra-wideband attenuation compensated signal. The waveform of the ultra-wideband attenuation compensation signal is corrected by an adaptive phase equalizer to obtain the ultra-wideband positioning calibration signal.

[0009] Furthermore, based on the label region category information corresponding to the positioning prediction data, and combined with the dynamic attenuation compensation model, signal attenuation compensation is performed on the ultra-wideband signal to obtain an ultra-wideband attenuation compensated signal, including: Based on the label region category information corresponding to the positioning prediction data, and combined with the transmission state region model, the number of reflections, path length, and metal obstacle attributes of the ultra-wideband reflection path are obtained. To obtain the amplitude and frequency of ultra-wideband signals; The number of reflections, the path length, the metal obstacle properties, the amplitude, and the frequency are input into the reflection path feature extraction module in the dynamic attenuation compensation model to extract the reflection path features; the reflection path feature extraction module is a neural network model built based on the Transformer model; The ultra-wideband signal and the reflection path features are input into the path feature compensation module in the dynamic attenuation compensation model to perform signal attenuation compensation on the positioning prediction data, thereby obtaining the ultra-wideband attenuation compensation signal; the path feature compensation module is a neural network model constructed based on the BP neural network model.

[0010] Furthermore, the step of performing waveform correction on the ultra-wideband attenuation compensation signal using an adaptive phase equalizer to obtain the ultra-wideband positioning calibration signal includes: By continuously updating the filter weight vector in the adaptive phase equalizer, the inverse channel response of the ultra-wideband attenuation compensation signal is generated. Based on the inverse response, the channel distortion of the ultra-wideband attenuation compensation signal is offset to obtain the ultra-wideband positioning calibration signal. The update formula for the filter weight vector in the adaptive phase equalizer is: ; ; in, express The filter weight vector at time t, express The filter weight vector at time t, Indicates the step size factor. express Error signal at time, express The ultra-wideband attenuation compensation signal vector at time t. This indicates the filter order of the adaptive phase equalizer. express Ultra-wideband attenuation compensation signal at any time, express Ultra-wideband attenuation compensation signal at any time, express Ultra-wideband attenuation compensation signal at any given time.

[0011] Furthermore, it also includes: Real-time metal distribution information is obtained by setting up anti-metal ultra-wideband anchor nodes; The obstacle distribution information is updated based on the real-time metal distribution information to obtain real-time obstacle distribution information; The transmission state region model is updated based on the real-time obstacle distribution information.

[0012] Secondly, this application also provides an industrial UWB positioning calibration device resistant to metal interference, comprising: The inertial prediction error calculation module is used to acquire positioning prediction data, inertial prediction data, and inertial measurement data of ultra-wideband signals, and calculate the difference between the inertial prediction data and the inertial measurement data as the inertial prediction error. The transmission state region identification module is used to input the positioning prediction data into the transmission state region model and output the corresponding label region non-line-of-sight weights; the transmission state region model is used to characterize the ultra-wideband reflection path of the region corresponding to the positioning prediction data. The non-line-of-sight weight calculation module is used to weight and sum the inertial prediction error and the non-line-of-sight weight of the label area to obtain the non-line-of-sight weight. The non-line-of-sight compensation and correction module is used to perform non-line-of-sight compensation and correction on the ultra-wideband signal through a dynamic attenuation compensation model if the non-line-of-sight weight is greater than a preset non-line-of-sight threshold, so as to obtain an ultra-wideband positioning calibration signal.

[0013] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method as described in the first aspect of this application.

[0014] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method as described in the first aspect of this application.

[0015] Beneficial Effects: This method calculates the inertial prediction error using inertial prediction data and inertial measurement data of ultra-wideband signals. The positioning prediction data is then input into a transmission state region model, outputting the corresponding label region non-line-of-sight (NLS) weights. Next, the NLS weights are weighted and summed to calculate the NLS weights. Finally, when the NLS weights exceed a preset NLS threshold, a dynamic attenuation compensation model is used to perform NLS compensation correction on the ultra-wideband signal, resulting in an ultra-wideband positioning calibration signal. This effectively solves the signal interference problem caused by metal obstacles, compensates for phase distortion caused by metal obstacles, and restores signal integrity. This method enables high-precision positioning of mobile robots and transportation equipment in environments with numerous metal obstacles, such as industrial warehouses and workshops, ensuring the stable operation of industrial automation systems. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart of an industrial UWB positioning calibration method against metal interference, as described in the embodiments of this application.

[0018] Figure 2 This is a flowchart illustrating the construction of the transmission state region model in this application embodiment.

[0019] Figure 3 This is a flowchart illustrating the generation of ultra-wideband positioning calibration signals in an embodiment of this application.

[0020] Figure 4 This is a schematic diagram of the structure of the industrial UWB positioning calibration device against metal interference in the embodiments of this application. Detailed Implementation

[0021] To make the above-mentioned objectives, features, and advantages of this application more apparent and understandable, the specific embodiments of this application are described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a thorough understanding of this application. However, this application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of this application. Therefore, this application is not limited to the specific embodiments disclosed below.

[0022] like Figure 1As shown, this embodiment provides an industrial UWB positioning calibration method to resist metal interference. This embodiment uses the application of this method to an ultra-wideband base station terminal as an example. It can be understood that this method can also be applied to ultra-wideband tag terminals, and can also be applied to systems including ultra-wideband tag terminals and ultra-wideband base station terminals, and is implemented through the interaction between the ultra-wideband base station terminal and the ultra-wideband tag terminal. In this embodiment, the method includes: S101: Acquire positioning prediction data, inertial prediction data, and inertial measurement data of the ultra-wideband signal, and calculate the difference between the inertial prediction data and the inertial measurement data as the inertial prediction error.

[0023] Optionally, the ultra-wideband base station terminal can estimate the positioning prediction data and inertial prediction data of the ultra-wideband signal of the ultra-wideband tag terminal based on the Kalman filter algorithm; the Kalman filter algorithm includes, but is not limited to, the traditional Kalman filter algorithm, the extended Kalman filter algorithm, and the adaptive unscented Kalman filter.

[0024] Optionally, the ultra-wideband tag terminal can integrate a miniature inertial measurement unit (INS), through which the ultra-wideband base station terminal can acquire inertial measurement data of the ultra-wideband signal. This inertial measurement data can help distinguish between line-of-sight (LOS) and non-line-of-sight (NLOS) states of the ultra-wideband signal, improving the accuracy of the ultra-wideband base station terminal's judgment of the signal transmission status.

[0025] S102: Input the positioning prediction data into the transmission state region model and output the corresponding label region non-line-of-sight weights; the transmission state region model is used to characterize the ultra-wideband reflection path of the region corresponding to the positioning prediction data.

[0026] Optional, such as Figure 2 As shown, this embodiment provides the process for constructing a transmission state region model, including: Step 1: Obtain obstacle distribution information, ultra-wideband base station positioning information, and the preset maximum number of reflections, and initialize the number of iterative reflections to zero; Step 2: Combining obstacle distribution information with the location of the ultra-wideband base station, set the minimum number of reflections required for the ultra-wideband signal to reach the ultra-wideband base station in the industrial positioning area to be equal to the label area category information and label area non-line-of-sight weight of the area corresponding to the number of iterative reflections, and increment the number of iterative reflections by one. Step 3: When the number of iterative reflections is less than or equal to the maximum number of reflections, step 2 is executed iteratively; when the number of iterative reflections is greater than the maximum number of reflections, the label area category information and the label area non-line-of-sight weight of the remaining area in the industrial positioning area are set as the blind zone label area category information and the blind zone label area non-line-of-sight weight, respectively, to obtain the transmission state area model.

[0027] Furthermore, the step of setting the minimum number of reflections required for the ultra-wideband signal to reach the ultra-wideband base station in the industrial positioning area to be equal to the number of iterative reflections, along with the label area category information and label area non-line-of-sight weight, includes: When the number of iteration reflections is zero, the category information of the label region corresponding to the region where the minimum number of reflections equals the number of iteration reflections is set as the category information of the line-of-sight region corresponding to the number of iteration reflections being zero; the non-line-of-sight weight of the label region corresponding to the region where the minimum number of reflections equals the number of iteration reflections is set as the non-line-of-sight weight of the line-of-sight region corresponding to the number of iteration reflections being zero. When the number of iterative reflections is greater than zero and less than or equal to the maximum number of reflections, the label region category information of the region corresponding to the minimum number of reflections equal to the number of iterative reflections is set as the category information of the non-line-of-sight region corresponding to the nth reflection when the number of iterative reflections is n; the label region non-line-of-sight weight of the region corresponding to the minimum number of reflections equal to the number of iterative reflections is set as the non-line-of-sight weight of the non-line-of-sight region corresponding to the nth reflection when the number of iterative reflections is n; the value of n is from 1 to the maximum number of reflections.

[0028] For example, let's take a maximum reflection count of two as an example: The label region category information of the corresponding region includes the category information of the line-of-sight region when the iteration reflection number parameter is set to zero, and the label region non-line-of-sight weight of the corresponding region includes the non-line-of-sight weight of the line-of-sight region when the iteration reflection number parameter is set to zero. The label area category information of the corresponding area also includes the category information of the non-line-of-sight area corresponding to the first reflection when the iteration reflection number parameter is set to one; the label area non-line-of-sight weight of the corresponding area also includes the non-line-of-sight weight of the non-line-of-sight area corresponding to the first reflection when the iteration reflection number parameter is set to one. The label region category information of the corresponding region also includes the category information of the secondary reflection non-line-of-sight region when the iteration reflection number parameter is set to two, and the label region non-line-of-sight weight of the corresponding region also includes the non-line-of-sight weight of the secondary reflection non-line-of-sight region when the iteration reflection number parameter is set to two.

[0029] In the aforementioned anti-metal interference industrial UWB positioning calibration method, by acquiring the preset maximum number of reflections, obstacle distribution information, and ultra-wideband base station positioning information, and combining iterative reflection number parameters to dynamically set tag area category information and non-line-of-sight weights, efficient and accurate modeling of signal propagation paths in industrial positioning areas can be achieved, improving the reliability of the positioning system. By iteratively setting tag area category information and non-line-of-sight weights, and setting the remaining area as blind zone tag area category information and blind zone tag area non-line-of-sight weights after the iterative reflection number parameters reach the maximum number of reflections, complete coverage of signal propagation paths in industrial positioning areas can be achieved, further improving the stability of the positioning system.

[0030] S103: The inertial prediction error and the non-line-of-sight weight of the label area are weighted and summed to obtain the non-line-of-sight weight.

[0031] Optionally, non-line-of-sight weights can be used to characterize the impact of metallic obstacles on ultra-wideband signals.

[0032] S104: If the non-line-of-sight weight is greater than the preset non-line-of-sight threshold, the ultra-wideband signal is compensated and corrected for non-line-of-sight through a dynamic attenuation compensation model to obtain an ultra-wideband positioning calibration signal.

[0033] In the aforementioned anti-metal interference industrial UWB positioning calibration method, multi-source data fusion and dynamic error correction technology can achieve deep collaborative optimization of ultra-wideband positioning data and inertial measurement data, and track phase shift and path distortion in the signal transmission process in real time. Through transmission state region model and non-line-of-sight intelligent recognition technology, it can efficiently and accurately identify the ultra-wideband signal reflection path in the industrial environment. By converting the signal transmission state in the positioning scene into quantifiable regional attributes, it can capture the non-line-of-sight state characteristics of the signal in real time. By performing non-line-of-sight compensation correction on the ultra-wideband signal when the non-line-of-sight weight is greater than a preset threshold, it can achieve accurate optimization and real-time adjustment of the positioning results, improve the positioning accuracy of the positioning system in complex environments, and enhance the reliability of the positioning system.

[0034] like Figure 3 As shown, the process of generating an ultra-wideband positioning calibration signal may include: S301: Based on the transmission state region model, obtain the label region category information corresponding to the positioning prediction data.

[0035] S302: Based on the label region category information corresponding to the positioning prediction data, and in conjunction with the dynamic attenuation compensation model, the ultra-wideband signal is attenuated to obtain an ultra-wideband attenuation compensated signal.

[0036] Specifically, ultra-wideband base stations can use tag area category information to simulate the reflection path of metal surfaces through ray tracing algorithms, and combine this with a deep learning-based dynamic attenuation compensation model to compensate for the signal attenuation of ultra-wideband signals caused by metal reflection, thus obtaining an ultra-wideband attenuation-compensated signal.

[0037] Optionally, when the industrial scenario changes, the ultra-wideband base station terminal can perform transfer learning on the pre-trained dynamic attenuation compensation model. By fine-tuning the model parameters using a small amount of new scenario data, the dynamic attenuation compensation model can quickly adapt to the new scenario, avoiding the cost of retraining and ensuring the stability of attenuation compensation in dynamic environments. Changes in the industrial scenario may include, but are not limited to, the addition of new metal equipment and alterations to the workshop layout.

[0038] Furthermore, based on the label region category information corresponding to the positioning prediction data, and combined with the dynamic attenuation compensation model, signal attenuation compensation is performed on the ultra-wideband signal to obtain an ultra-wideband attenuation compensated signal, including: Based on the label region category information corresponding to the positioning prediction data, and combined with the transmission state region model, the number of reflections, path length, and metal obstacle attributes of the ultra-wideband reflection path are obtained.

[0039] Optionally, the attributes of the metal obstacle include, but are not limited to, the metal obstacle material parameters, the metal obstacle thickness, and the metal obstacle surface roughness.

[0040] Obtain the amplitude and frequency of the ultra-wideband signal.

[0041] The number of reflections, the path length, the properties of the metal obstacle, the amplitude, and the frequency are input into the reflection path feature extraction module in the dynamic attenuation compensation model mounted on the ultra-wideband base station terminal to extract the reflection path features.

[0042] Optionally, the reflection path feature extraction module is a neural network model built based on the Transformer model.

[0043] The ultra-wideband signal and the reflection path features are input into the path feature compensation module in the dynamic attenuation compensation model carried by the ultra-wideband base station terminal to perform signal attenuation compensation on the positioning prediction data, thereby obtaining the ultra-wideband attenuation compensation signal.

[0044] Optionally, the path feature compensation module is a neural network model built based on the BP neural network model.

[0045] In the aforementioned industrial UWB positioning calibration method against metal interference, by combining tag area category information and transmission state area model to obtain the ultra-wideband reflection path and performing signal attenuation compensation, the signal propagation path in complex environments can be accurately identified and the ultra-wideband signal can be efficiently compensated, thereby improving the adaptability and accuracy of the positioning system.

[0046] S303: The waveform of the ultra-wideband attenuation compensation signal is corrected by an adaptive phase equalizer to obtain the ultra-wideband positioning calibration signal.

[0047] Optionally, the ultra-wideband base station terminal can use an adaptive phase equalizer to perform waveform correction on the ultra-wideband attenuation compensation signal of the ultra-wideband signal in a non-line-of-sight state, compensate for the phase distortion caused by metal obstacles, restore the integrity of the signal, and obtain ultra-wideband positioning calibration signal information.

[0048] In the aforementioned industrial UWB positioning calibration method against metal interference, by combining tag area category information with a signal attenuation compensation model to compensate for the signal attenuation of the ultra-wideband signal, it is possible to accurately compensate for the propagation loss of the ultra-wideband signal in complex environments. This avoids the decrease in positioning accuracy caused by signal attenuation in traditional positioning systems, further improving the accuracy and reliability of the positioning system. By using an adaptive phase equalizer to correct the waveform of the ultra-wideband attenuation compensation signal, it is possible to accurately adjust and optimize the waveform of the ultra-wideband signal, thereby improving the stability of the positioning system.

[0049] Furthermore, the step of performing waveform correction on the ultra-wideband attenuation compensation signal using an adaptive phase equalizer to obtain the ultra-wideband positioning calibration signal includes: By continuously updating the filter weight vector in the adaptive phase equalizer, the inverse channel response of the ultra-wideband attenuation compensation signal is generated. Based on the inverse response, the channel distortion of the ultra-wideband attenuation compensation signal is offset to obtain the ultra-wideband positioning calibration signal. Optionally, the update formula for the filter weight vector in the adaptive phase equalizer is: ; ; in, express The filter weight vector at time t, express The filter weight vector at time t, Indicates the step size factor. express Error signal at time, express The ultra-wideband attenuation compensation signal vector at time t. This indicates the filter order of the adaptive phase equalizer. express Ultra-wideband attenuation compensation signal at any time, express Ultra-wideband attenuation compensation signal at any time, express Ultra-wideband attenuation compensation signal at any given time.

[0050] Furthermore, such as Figure 2 As shown, the process of generating ultra-wideband positioning calibration signals may further include: S304: Obtain real-time metal distribution information through the configured anti-metal ultra-wideband anchor node.

[0051] Specifically, the ultra-wideband tag terminal can be equipped with anti-metal ultra-wideband anchor nodes. The anti-metal ultra-wideband anchor nodes can be electromagnetically shielded and encapsulated, with built-in metal detection sensors (such as eddy current probes) to sense the surrounding metal distribution in real time and transmit the metal distribution information to the processing equipment for signal compensation and calibration.

[0052] S305: Update the obstacle distribution information based on the real-time metal distribution information to obtain real-time obstacle distribution information.

[0053] S306: Update the transmission state region model based on the real-time obstacle distribution information.

[0054] In the aforementioned anti-metal interference industrial UWB positioning calibration method, real-time metal distribution information is obtained by setting anti-metal ultra-wideband anchor nodes, and obstacle distribution information and transmission state area model are updated based on this. This enables dynamic perception and modeling of metal obstacles in the industrial environment and real-time updating of the transmission state area model.

[0055] In the process of generating ultra-wideband positioning calibration signals, a signal attenuation compensation model is established by simulating the reflection path of metal surfaces through a ray tracing algorithm, which can compensate for signal attenuation caused by metal reflection. By using an inertial measurement unit, inertial data is used to help distinguish between line-of-sight and non-line-of-sight states, which can improve the accuracy of signal transmission status. By using an adaptive phase equalizer to correct the waveform of the ultra-wideband attenuation compensation signal, the phase distortion caused by metal obstacles can be compensated, and the integrity of the signal can be restored.

[0056] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0057] Based on the same inventive concept, this application also provides an industrial UWB positioning calibration device for implementing the aforementioned anti-metal interference industrial UWB positioning calibration method. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of the anti-metal interference industrial UWB positioning calibration device provided below can be found in the limitations of the anti-metal interference industrial UWB positioning calibration method described above, and will not be repeated here.

[0058] like Figure 4 As shown, this embodiment provides an industrial UWB positioning calibration device 400 with resistance to metal interference, comprising: The inertial prediction error calculation module 401 is used to acquire positioning prediction data, inertial prediction data and inertial measurement data of ultra-wideband signals, and calculate the difference between the inertial prediction data and the inertial measurement data as the inertial prediction error. The transmission state region identification module 402 is used to input the positioning prediction data into the transmission state region model and output the corresponding label region non-line-of-sight weights; the transmission state region model is used to characterize the ultra-wideband reflection path of the region corresponding to the positioning prediction data. The non-line-of-sight weight calculation module 403 is used to weight and sum the inertial prediction error and the non-line-of-sight weight of the label area to obtain the non-line-of-sight weight. The non-line-of-sight compensation and correction module 404 is used to perform non-line-of-sight compensation and correction on the ultra-wideband signal through a dynamic attenuation compensation model if the non-line-of-sight weight is greater than a preset non-line-of-sight threshold, so as to obtain an ultra-wideband positioning calibration signal.

[0059] This embodiment also provides a computer device, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the industrial UWB positioning calibration method against metal interference as described above.

[0060] Optionally, the computer device further includes a signal processing device, which can be used to receive data from the tag and the anti-metal ultra-wideband anchor node, perform operations such as ray tracing algorithm, signal attenuation compensation, and adaptive phase equalization, and realize the calibration of the ultra-wideband signal and the accurate calculation of the target position.

[0061] This embodiment also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the industrial UWB positioning calibration method against metal interference as described above.

[0062] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0063] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. An anti-metal interference industrial UWB positioning calibration method, characterized in that, The method comprises the following steps: obtaining positioning prediction data, inertial prediction data and inertial measurement data of the ultra-wideband signal, and calculating the difference between the inertial prediction data and the inertial measurement data as an inertial prediction error; inputting the positioning prediction data into a transmission state area model to output corresponding label area non-line-of-sight weights; the transmission state area model is used to represent the ultra-wideband reflection path of the positioning prediction data corresponding area; weighting and summing the inertial prediction error and the label area non-line-of-sight weights to obtain a non-line-of-sight weight; if the non-line-of-sight weight is greater than a preset non-line-of-sight threshold, performing non-line-of-sight compensation correction on the ultra-wideband signal through a dynamic attenuation compensation model to obtain an ultra-wideband positioning calibration signal.

2. The method of claim 1, wherein, The construction process of the transmission state area model comprises: Step 1: obtaining obstacle distribution information, ultra-wideband base station positioning information and a preset maximum reflection number, and initializing an iterative reflection number to zero; the ultra-wideband base station positioning information comprises the position of the ultra-wideband base station; Step 2: combining the obstacle distribution information and the position of the ultra-wideband base station, setting the minimum reflection number of the ultra-wideband signal reaching the ultra-wideband base station in the industrial positioning area according to the iterative reflection number, and performing a plus operation on the iterative reflection number; Step 3: when the iterative reflection number is less than or equal to the maximum reflection number, iteratively executing Step 2; when the iterative reflection number is greater than the maximum reflection number, setting the label area category information and the label area non-line-of-sight weight of the remaining area in the industrial positioning area to blind area label area category information and blind area label area non-line-of-sight weight respectively to obtain the transmission state area model.

3. The method of claim 2, wherein, The setting of the minimum reflection number of the ultra-wideband signal reaching the ultra-wideband base station in the industrial positioning area according to the iterative reflection number to the label area category information and the label area non-line-of-sight weight of the area corresponding to the iterative reflection number comprises: when the iterative reflection number is zero, setting the minimum reflection number equal to the label area category information of the area corresponding to the iterative reflection number to the category information of the line-of-sight area corresponding to the iterative reflection number being zero; setting the minimum reflection number equal to the label area non-line-of-sight weight of the area corresponding to the iterative reflection number to the non-line-of-sight weight of the line-of-sight area corresponding to the iterative reflection number being zero; when the iterative reflection number is greater than zero and less than or equal to the maximum reflection number, setting the minimum reflection number equal to the label area category information of the area corresponding to the iterative reflection number to the category information of the n-time reflection non-line-of-sight area corresponding to the iterative reflection number being n; setting the minimum reflection number equal to the label area non-line-of-sight weight of the area corresponding to the iterative reflection number to the non-line-of-sight weight of the n-time reflection non-line-of-sight area corresponding to the iterative reflection number being n; the value of n is in the order from 1 to the maximum reflection number.

4. The method of claim 2, wherein, The non-line-of-sight compensation correction on the ultra-wideband signal through the dynamic attenuation compensation model based on deep learning to obtain the ultra-wideband positioning calibration signal comprises: Based on the transmission state area model, the label area category information corresponding to the positioning prediction data is obtained; Based on the label area category information corresponding to the positioning prediction data, and combined with the dynamic attenuation compensation model, the signal attenuation compensation is performed on the ultra-wideband signal to obtain an ultra-wideband attenuation compensation signal; The adaptive phase equalizer is used to correct the waveform of the ultra-wideband attenuation compensation signal to obtain the ultra-wideband positioning calibration signal.

5. The method of claim 4, wherein, The signal attenuation compensation is performed on the ultra-wideband signal based on the label area category information corresponding to the positioning prediction data and combined with the dynamic attenuation compensation model to obtain an ultra-wideband attenuation compensation signal. Based on the label area category information corresponding to the positioning prediction data, the transmission state area model is combined to obtain the reflection number, path length and metal obstacle attribute of the ultra-wideband reflection path; The amplitude and frequency of the ultra-wideband signal are obtained; The reflection number, path length, metal obstacle attribute, amplitude and frequency are input into the reflection path feature extraction module in the dynamic attenuation compensation model to extract the reflection path feature; the reflection path feature extraction module is a neural network model constructed based on the Transformer model; The ultra-wideband signal and the reflection path feature are input into the path feature compensation module in the dynamic attenuation compensation model to perform signal attenuation compensation on the positioning prediction data to obtain the ultra-wideband attenuation compensation signal; the path feature compensation module is a neural network model constructed based on the BP neural network model.

6. The method of claim 4, wherein, The adaptive phase equalizer is used to correct the waveform of the ultra-wideband attenuation compensation signal to obtain the ultra-wideband positioning calibration signal. The inverse response of the channel of the ultra-wideband attenuation compensation signal is generated by continuously updating the filter weight vector in the adaptive phase equalizer, the channel distortion of the ultra-wideband attenuation compensation signal is offset based on the inverse response, and the ultra-wideband positioning calibration signal is obtained; The update formula of the filter weight vector in the adaptive phase equalizer is: ； ； wherein denotes the filter weight vector at time instant denotes the filter weight vector at time instant denotes a step size factor denotes the error signal at time instant denotes the ultra-wideband attenuation compensation signal vector at time instant denotes the filter order of the adaptive phase equalizer denotes the ultra-wideband attenuation compensation signal at time instant denotes the ultra-wideband attenuation compensation signal at time instant denotes the ultra-wideband attenuation compensation signal at time instant 7. The method of claim 2, wherein, Further comprising: Real-time metal distribution information is obtained through the set anti-metal ultra-wideband anchor node; The obstacle distribution information is updated based on the real-time metal distribution information to obtain real-time obstacle distribution information; The transmission state area model is updated based on the real-time obstacle distribution information.

8. An anti-metal interference industrial UWB positioning calibration device, characterized in that, Comprising: An inertial prediction error calculation module is configured to obtain positioning prediction data, inertial prediction data and inertial measurement data of an ultra-wideband signal, and calculate the difference between the inertial prediction data and the inertial measurement data as an inertial prediction error; A transmission state area identification module is configured to input the positioning prediction data into a transmission state area model to output corresponding label area non-line-of-sight weights; the transmission state area model is used to represent the ultra-wideband reflection path of the area corresponding to the positioning prediction data; A non-line-of-sight weight calculation module is configured to weight and sum the inertial prediction error and the label area non-line-of-sight weight to obtain a non-line-of-sight weight; A non-line-of-sight compensation correction module is configured to, if the non-line-of-sight weight is greater than a preset non-line-of-sight threshold, perform non-line-of-sight compensation correction on the ultra-wideband signal by a dynamic attenuation compensation model to obtain an ultra-wideband positioning calibration signal. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-8 when the computer program is executed by the processor. The computer program is executed by the processor to implement the method in any one of claims 1 to 7.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the method in any one of claims 1 to 7.

Images