Tunnel personnel positioning method, device, equipment and storage medium

Abstract

This application relates to the field of positioning technology, and discloses a method, device, equipment and storage medium for positioning personnel in a tunnel. The method includes: obtaining first reference data, second reference data and third reference data; inputting the first reference data, second reference data and third reference data into a distance prediction model to predict the distance between a terminal to be located and a base station, so as to obtain a distance prediction value; and determining the terminal position of the terminal to be located according to the distance prediction value. The embodiments of this application can improve the accuracy of personnel positioning in a tunnel by combining multiple radio signals for personnel positioning in the tunnel.

Description

Technical Field

[0001] The present application relates to the technical field of positioning, and in particular, to a method, device, equipment and storage medium for positioning personnel in a tunnel. Background Art

[0002] Currently, the methods for positioning personnel in a tunnel are Ultra Wide Band (UWB), Radio Frequency Identification (RFID), or Wi-Fi positioning methods. That is, the terminal to be located and the base station mutually transmit and receive indoor radio signals, and calculate parameters such as the propagation time, signal strength, and multipath fading of the indoor radio signals to determine the distance or position between the terminal to be located and the base station, so as to achieve the positioning of the personnel carrying the terminal to be located. However, due to problems such as severe space occlusion and frequent signal interference in the tunnel, the above positioning methods are difficult to maintain the continuity and visibility of the positioning signals, and thus cannot normally locate the personnel in the tunnel. Summary of the Invention

[0003] The purpose of the present application is to provide a method, device, equipment and storage medium for positioning personnel in a tunnel, which can improve the positioning accuracy of personnel in a tunnel by combining multiple radio signals.

[0004] An embodiment of the present application provides a method for positioning personnel in a tunnel, including:

[0005] Obtaining first reference data, second reference data, and third reference data; the first reference data is communication quality data when the terminal to be located and the base station perform long-distance wireless communication, the second reference data is distance measurement data obtained by the terminal to be located and the base station through frequency hopping ranging, and the third reference data is communication quality data when the terminal to be located and the base station perform short-distance wireless communication;

[0006] Inputting the first reference data, the second reference data, and the third reference data into a distance prediction model to predict the distance between the terminal to be located and the base station, and obtaining a distance prediction value;

[0007] Determining the terminal position of the terminal to be located according to the distance prediction value.

[0008] In some embodiments, before obtaining the first reference data, second reference data, and third reference data, it further includes:

[0009] Obtaining a reference data packet forwarded by the base station from the terminal to be located;

[0010] Parse the reference data packet to obtain the first reference data, the second reference data, and the third reference data; the reference data packet is obtained by encapsulating the first reference data, the second reference data, and the third reference data after the to-be-located terminal performs long-distance wireless communication, frequency hopping ranging, and short-distance wireless communication with the base station in sequence.

[0011] In some embodiments, the method for generating the second reference data is as follows:

[0012] Determine reference distance data according to the distance measurement data set obtained by frequency hopping ranging;

[0013] Perform polynomial fitting on the reference distance data to obtain distance measurement fitting data;

[0014] Judge whether the distance measurement fitting data meets the preset ranging data conditions; the preset ranging data conditions are that the distance measurement fitting data is within a preset ranging data interval and the frequency error between the to-be-located terminal and the base station is within a preset frequency error interval;

[0015] If it meets the conditions, output the distance measurement fitting data as the second reference data.

[0016] In some embodiments, the step of inputting the first reference data, the second reference data, and the third reference data into a distance prediction model to predict the distance between the to-be-located terminal and the base station to obtain a distance prediction value includes:

[0017] Perform local feature extraction on the first reference data, the second reference data, and the third reference data to obtain a local data vector sequence containing a plurality of local data vectors;

[0018] Perform positional encoding on the local data vector sequence to obtain a positional embedding vector sequence;

[0019] Perform linear transformation processing on the positional embedding vector sequence, and perform attention weight calculation on the local data vectors after the linear transformation processing to obtain a global distance feature;

[0020] Predict the distance between the to-be-located terminal and the base station according to the global distance feature to obtain the distance prediction value.

[0021] In some embodiments, the step of determining the terminal position of the to-be-located terminal according to the distance prediction value includes:

[0022] Within a preset waiting duration, judge whether the distance prediction model outputs two distance prediction values obtained by predicting the distances between the same to-be-located terminal and two base stations;

[0023] If not, determine the relative direction between the terminal to be located and the base station according to the historical positioning data, and determine the terminal position of the terminal to be located in combination with the distance prediction value;

[0024] If so, calculate the terminal position of the terminal to be located according to the two distance prediction values and the distance value between the two base stations.

[0025] In some embodiments, the training method of the distance prediction model includes:

[0026] Obtain the current true distance value between the terminal to be located and the base station;

[0027] Input the first reference data, the second reference data, and the third reference data into the deep neural network model to be trained to obtain a training distance prediction value;

[0028] Determine the model loss information according to the training distance prediction value and the true distance value;

[0029] Judge whether the model loss information is within the loss threshold range;

[0030] If not, adjust the weight parameters of the neural network model to be trained; return to the step of inputting the first reference data, the second reference data, and the third reference data into the deep neural network model to be trained to obtain a training distance prediction value;

[0031] If it is, judge whether the training reset times reach the reset threshold;

[0032] If not, change the position of the terminal to be located; return to the step of obtaining the current true distance value between the terminal to be located and the base station;

[0033] If so, end the training to obtain the distance prediction model.

[0034] In some embodiments, the first reference data includes the long-distance wireless communication frequency error between the terminal to be located and the base station, the long-distance wireless communication signal strength value recorded by the base station, the long-distance wireless communication signal strength value recorded by the terminal to be located, and the long-distance wireless communication channel signal-to-noise ratio; the second reference data is the distance measurement data obtained by frequency hopping ranging when the terminal to be located and the base station perform long-distance wireless communication; the third reference data includes the short-distance wireless communication signal strength value recorded by the base station and the short-distance wireless communication signal strength value recorded by the terminal to be located.

[0035] An embodiment of the present application provides a tunnel personnel positioning device, including:

[0036] The first module is used to obtain the first reference data, the second reference data, and the third reference data; the first reference data is the communication quality data when the terminal to be located and the base station perform long-distance wireless communication, the second reference data is the distance measurement data obtained by the terminal to be located and the base station through frequency hopping ranging, and the third reference data is the communication quality data when the terminal to be located and the base station perform short-distance wireless communication;

[0037] The second module is used to input the first reference data, the second reference data, and the third reference data into a distance prediction model to predict the distance between the terminal to be located and the base station, and obtain a distance prediction value;

[0038] The third module is used to determine the terminal position of the terminal to be located according to the distance prediction value.

[0039] An embodiment of the present application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the above-mentioned method for locating personnel in a tunnel is implemented.

[0040] An embodiment of the present application provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, the above-mentioned method for locating personnel in a tunnel is implemented.

[0041] The beneficial effects of the present application: By using a pre-trained distance prediction model, based on the communication quality data when the terminal to be located and the base station perform long-distance wireless communication, the distance measurement data obtained by the terminal to be located and the base station through frequency hopping ranging, and the communication quality data when the terminal to be located and the base station perform short-distance wireless communication, the distance between the terminal to be located and the base station is predicted, so as to realize the positioning of personnel in the tunnel. Since the communication quality data obtained by combining two communication methods of long-distance wireless communication and short-distance wireless communication are mutually verified, and multiple distance measurement data obtained by using different channels are obtained through frequency modulation ranging, the pre-trained distance prediction model can predict the distance between the terminal to be located and the base station based on reference data in multiple dimensions, enhancing the accuracy and reliability of the distance prediction value and improving the accuracy of personnel positioning in the tunnel. Description of the Drawings

[0042] Figure 1 It is an implementation environment diagram of the method for locating personnel in a tunnel provided by an embodiment of the present application.

[0043] Figure 2 It is a flowchart of the method for locating personnel in a tunnel provided by an embodiment of the present application.

[0044] Figure 3 It is a flowchart of the method for generating the second reference data provided by an embodiment of the present application.

[0045] Figure 4 It is a flowchart of the specific method of step S202 provided by an embodiment of the present application.

[0046] Figure 5 It is a flowchart of the specific method of step S203 provided by an embodiment of the present application.

[0047] Figure 6 It is a flowchart of the training method of the distance prediction model provided by an embodiment of the present application.

[0048] Figure 7 It is a schematic structural diagram of a personnel positioning device in a tunnel provided by an embodiment of the present application.

[0049] Figure 8 It is a schematic hardware structure diagram of an electronic device provided by an embodiment of the present application.

[0050] Figure 9 It is a schematic model structure diagram of the distance prediction model provided by an embodiment of the present application. Specific embodiments

[0051] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0052] It should be noted that although functional module division is performed in the device schematic diagram and the logical order is shown in the flowchart, in some cases, the steps shown can be executed in a different module division in the device or a different order in the flowchart. Terms such as "first" and "second" in the description, claims and drawings are used to distinguish similar objects and are not used to describe a specific order or sequence.

[0053] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application and are not intended to limit the present application.

[0054] The execution subject of the personnel positioning method in a tunnel provided by an embodiment of the present application is applied to end-side devices and can also be applied to cloud-side devices, and this embodiment does not make any limitations in this regard.

[0055] The method provided by an embodiment of the present application can be applied to Figure 1The illustrated implementation environment includes a terminal to be located 110, multiple base stations 120, and an execution entity 130. The base stations 120 are arranged in the tunnel and are spaced at intervals along the length of the tunnel. When construction workers carry the terminal to be located 110 into the tunnel, the terminal to be located 110 and the base stations 120 can perform long-distance wireless communication or short-distance wireless communication, and the execution entity 130 can communicate with the base stations 120 through a communication network. Among them, the communication network uses standard communication technologies and / or protocols, usually the Internet, but it can also be any network, including but not limited to any combination of Bluetooth, local area network, metropolitan area network, wide area network, mobile, private network, or virtual private network. In some embodiments, customized or dedicated data communication technologies can be used to replace or supplement the above data communication technologies.

[0056] Refer to Figure 2 , Figure 2 is a flowchart of a method for locating personnel in a tunnel provided by an embodiment of the present application. In some embodiments, Figure 2 the method in includes but is not limited to steps S201 to S203.

[0057] Step S201, obtain first reference data, second reference data, and third reference data.

[0058] Among them, the first reference data is communication quality data when the terminal to be located and the base station perform long-distance wireless communication, the second reference data is distance measurement data obtained by the terminal to be located and the base station through frequency hopping ranging, and the third reference data is communication quality data when the terminal to be located and the base station perform short-distance wireless communication.

[0059] In specific implementation, construction workers carry the terminal to be located into the tunnel. The terminal to be located performs wireless communication interaction with the two nearest base stations and obtains the first reference data, the second reference data, and the third reference data. After the interaction between the terminal to be located and the base station is completed, the execution entity obtains the first reference data, the second reference data, and the third reference data from the terminal to be located or the base station.

[0060] Step S202, input the first reference data, the second reference data, and the third reference data into a distance prediction model to predict the distance between the terminal to be located and the base station, and obtain a distance prediction value.

[0061] Among them, the distance prediction model is obtained by training a depth neural network model to be trained based on the first reference data, the second reference data, the third reference data, and the true distance value between the terminal to be located and the base station.

[0062] In specific implementation, the first reference data, the second reference data, and the third reference data are input into the distance prediction model. The distance prediction model extracts local features from among the first reference data, the second reference data, and the third reference data, and understands the long-distance dependence relationship between the extracted local features based on the multi-head attention mechanism and the feed-forward neural network. Combining the local features and the corresponding long-distance dependence relationship, the distance between the terminal to be located and the base station is predicted to obtain a distance prediction value.

[0063] Step S203: Determine the terminal position of the terminal to be located according to the distance prediction value.

[0064] In specific implementation, in the one-dimensional environment of a tunnel, after a terminal to be located communicates with at least two base stations and the distance prediction model predicts and outputs two distance prediction values, the terminal position of the terminal to be located can be determined, or after the distance prediction model predicts and outputs one distance prediction value, the terminal position of the terminal to be located is determined according to the historical positioning record.

[0065] In a specific embodiment, before step S201, it further includes: obtaining a reference data packet forwarded by the base station from the terminal to be located; parsing the reference data packet to obtain the first reference data, the second reference data, and the third reference data. The reference data packet is obtained by encapsulating the first reference data, the second reference data, and the third reference data after the terminal to be located performs long-distance wireless communication, frequency hopping ranging, and short-distance wireless communication with the base station in sequence.

[0066] In specific implementation, after the interaction between the terminal to be located and the base station is completed, the terminal to be located encapsulates the first reference data, the second reference data, and the third reference data obtained during the interaction in the reference data packet and sends it to the base station. The execution entity obtains the reference data packet forwarded by the base station from the terminal to be located, and parses the reference data packet to obtain the first reference data, the second reference data, and the third reference data.

[0067] The specific process of the interaction between the terminal to be located and the base station to generate the first reference data, the second reference data, and the third reference data includes a first interaction stage, a second interaction stage, and a third interaction stage. In the first interaction stage, the terminal to be located and the base station perform long-distance wireless communication. The terminal to be located sends a long-distance communication request message to the base station. The base station receives the long-distance communication request message and verifies the identity information of the terminal to be located in the message. If the verification result shows that the terminal to be located is a legitimate terminal, the base station records the communication quality data when receiving the long-distance communication request message and writes it into the long-distance communication feedback message, and sends the long-distance communication feedback message to the terminal to be located. The terminal to be located records the communication quality data when receiving the long-distance communication consent message and the communication quality data when the base station records receiving the long-distance communication request message, and obtains the first reference data. In the second interaction stage, the terminal to be located and the base station perform long-distance wireless communication and switch to the frequency hopping ranging mode. The entire communication bandwidth is divided into multiple channels according to the frequency value. The terminal to be located and the base station sequentially complete message sending and message receiving through each channel, and determine the distance measurement data obtained by ranging using this channel according to the message sending and receiving duration. After traversing all channels, the distance measurement data that can represent the distance between the terminal to be located and the base station is selected to obtain the second reference data. In the third interaction stage, the terminal to be located and the base station perform short-distance wireless communication. The terminal to be located sends a short-distance communication request message to the base station. The base station receives the short-distance communication request message and verifies the identity information of the terminal to be located in the message. If the verification result shows that the terminal to be located is a legitimate terminal, the base station records the communication quality data when receiving the short-distance communication request message and writes it into the short-distance communication feedback message, and sends the short-distance communication feedback message to the terminal to be located. The terminal to be located records the communication quality data when receiving the short-distance communication consent message and the communication quality data when the base station records receiving the short-distance communication request message, and obtains the third reference data.

[0068] Figure 3 It is a flowchart of the method for generating the second reference data provided by the embodiments of the present application. Refer to Figure 3 In some embodiments, the method includes but is not limited to steps S301 to S304.

[0069] Step S301: Determine the reference distance data according to the distance measurement data set obtained by frequency hopping ranging.

[0070] Step S302: Perform polynomial fitting on the reference distance data to obtain the distance measurement fitting data.

[0071] Step S303: Determine whether the distance measurement fitting data meets the preset ranging data conditions.

[0072] Among them, the preset ranging data condition is that the distance measurement fitting data is within the preset ranging data interval and the frequency error between the terminal to be located and the base station is within the preset frequency error interval.

[0073] If it meets the conditions, execute step S304; if it does not meet the conditions, discard the distance measurement fitting data.

[0074] Step S304, output the distance measurement fitting data as the second reference data.

[0075] In a specific implementation, when the second interaction stage is completed, multiple distance measurement data for message transmission and reception based on multiple channels are calculated to obtain a distance measurement data set. Select the distance measurement data as the median from the distance measurement data set, or calculate the average value of each distance measurement data in the distance measurement data set to obtain the reference distance data. Then, use the expert experience library to perform polynomial fitting on the reference distance data to obtain the distance measurement fitting data, and determine whether the distance measurement fitting data meets the preset ranging data conditions. When it meets the preset ranging data conditions, use the distance measurement fitting data as the second reference data.

[0076] In a specific embodiment, the first reference data includes the long-distance wireless communication frequency error between the terminal to be located and the base station, the long-distance wireless communication signal strength value recorded by the base station, the long-distance wireless communication signal strength value recorded by the terminal to be located, and the long-distance wireless communication channel signal-to-noise ratio. The second reference data is the distance measurement data obtained by frequency hopping ranging when the terminal to be located and the base station perform long-distance wireless communication. The third reference data includes the short-distance wireless communication signal strength value recorded by the base station and the short-distance wireless communication signal strength value recorded by the terminal to be located.

[0077] Preferably, LoRa communication is used when the terminal to be located and the base station perform long-distance wireless communication, which has good signal penetration and can better maintain the continuity and visibility of the positioning signal. BLE communication is used when the terminal to be located and the base station perform short-distance wireless communication.

[0078] Figure 4 It is a flowchart of the specific method of step S202 provided by the embodiment of the present application. Refer to Figure 4 , in some embodiments, the method includes but is not limited to steps S401 to S404.

[0079] Step S401, perform local feature extraction on the first reference data, the second reference data, and the third reference data to obtain a local data vector sequence including multiple local data vectors.

[0080] Step S402, perform position encoding on the local data vector sequence to obtain a position embedding vector sequence.

[0081] Step S403: Perform a linear transformation on the position embedding vector sequence, and perform an attention weight operation on the local data vectors after the linear transformation to obtain global distance features.

[0082] Step S404: Predict the distance between the terminal to be located and the base station based on the global distance features to obtain a distance prediction value.

[0083] In some embodiments, the first reference data, the second reference data, and the third reference data are preprocessed first, and then local feature extraction is performed on the preprocessed first reference data, second reference data, and third reference data. For example, normalize the number of successful frequency modulation ranging times, convert the format of the time stamp, and standardize other data, etc.

[0084] Refer to Figure 9 In the embodiments of the present application, the distance prediction model includes a convolutional network layer, a position encoding network layer, an encoder network layer, and a decoder network layer.

[0085] Performing local feature extraction on the first reference data, the second reference data, and the third reference data is executed by invoking the convolutional network layer of the distance prediction model. In specific implementation, the first reference data, the second reference data, and the third reference data are input into the convolutional network layer, and layer-by-layer convolution operations are performed in the convolutional network layer to extract the local features among the first reference data, the second reference data, and the third reference data. The last hidden layer of the convolutional network layer is connected to an activation module, and the activation module uses an activation function to perform activation processing on the result of the last convolution operation of the convolutional network layer, so as to obtain local data vectors. After aggregating each local data vector, a local data vector sequence is obtained.

[0086] Performing position encoding on the local data vector sequence is executed by invoking the position encoding network layer of the distance prediction model. In specific implementation, the local data vector sequence is input into the position encoding network layer to embed corresponding position information into each local data vector in the local data vector sequence to represent the time sequence relationship of the corresponding local features. After the position encoding network layer performs position encoding on the local data vector sequence, the local data vector sequence is mapped into a vector sequence composed of multiple local data vectors after embedding position information, that is, a position embedding vector sequence.

[0087] Perform a linear transformation on the sequence of position embedding vectors, and perform attention weight operations on the local data vectors after the linear transformation, which is executed by calling the encoder network layer of the distance prediction model. In a specific implementation, after obtaining the sequence of position embedding vectors, perform a linear transformation on the local data vectors after each embedding position information in the sequence of position embedding vectors, and perform attention weight operations on each local data vector after the linear transformation to obtain global distance features. More specifically, input the sequence of position embedding vectors into the encoder network layer. The encoder network layer connects the multi-head mechanism and the feed-forward layer through the residual network result. The multi-head mechanism performs multiple linear transformation operations on the input vectors to obtain different linear values, and then performs operations on the attention weights. The attention weight operation formula is as follows:

[0088] ,

[0089] ,

[0090] where Q, K, and V are the input local data vector matrices, , and are trainable weight matrices, Attention is the attention weight operation, Concat is the concatenation process, is the i-th hyperparameter head, and i is an integer greater than or equal to 1 and less than or equal to h. Through the above formula, the principle of performing a linear transformation on the local data vectors after each embedding position information in the sequence of position embedding vectors and performing attention weight operations on each local data vector after the linear transformation is: map Q, K, and V through the corresponding weight matrices and then perform the Attention operation. After repeating h times, concatenate the calculation results to obtain the global distance features reflecting the long-distance dependence relationship between local features.

[0091] Predicting the distance between the terminal to be located and the base station based on the global distance features is executed by calling the decoder network layer of the distance prediction model. After obtaining the global distance features, understand the global distance information of the first reference data, the second reference data, and the third reference data layer by layer through the global distance features to predict the distance between the terminal to be located and the base station. In a specific implementation, input the global distance features into the decoder network layer. The decoder network layer connects the multi-head mechanism and the feed-forward layer through the residual network result. The multi-head mechanism performs multiple linear transformation operations on the input vectors to obtain different linear values, and then performs operations on the attention weights to capture the representations of different features obtained by different linear transformations, thereby capturing the global information in the first reference data, the second reference data, and the third reference data. Combine the global information in the first reference data, the second reference data, and the third reference data to predict the distance between the terminal to be located and the base station and obtain the distance prediction value.

[0092] In some embodiments, the terminal to be located may also collect geomagnetic data. The execution entity obtains the geomagnetic data collected by the terminal to be located and inputs it together with the first reference data, the second reference data, and the third reference data into the distance prediction model. When predicting the distance between the terminal to be located and the base station, the distance prediction model determines the moving direction of the terminal to be located through the geomagnetic data to obtain a more accurate distance prediction value.

[0093] In some embodiments, the base stations may also communicate with each other for distance detection to obtain base station interval data. The execution entity obtains the base station interval data, can determine the position information and position change information of the base stations, and inputs the base station interval data together with the first reference data, the second reference data, and the third reference data into the distance prediction model. When predicting the distance between the terminal to be located and the base station, more reference information is provided to obtain a more accurate distance prediction value.

[0094] Figure 5 It is a flowchart of the specific method of step S203 provided by the embodiments of the present application. Refer to Figure 5 In some embodiments, the method includes but is not limited to steps S501 to S503.

[0095] Step S501: Within a preset waiting duration, determine whether the distance prediction model outputs two distance prediction values obtained by predicting the distances between the same terminal to be located and two base stations.

[0096] If not, execute step S502; if so, execute step S503.

[0097] Step S502: Determine the relative direction between the terminal to be located and the base station according to the historical positioning data, and combine the distance prediction value to determine the terminal position of the terminal to be located.

[0098] Step S503: Calculate the terminal position of the terminal to be located according to the two distance prediction values and the distance value between the two base stations.

[0099] When the distance prediction model outputs two distance prediction values obtained by predicting the distances between the same terminal to be located and two base stations, the calculation formula for the terminal position of the terminal to be located is:

[0100] ,

[0101] ,

[0102] ,

[0103] where is the terminal position of the terminal to be located, and are all weight coefficients, is the predicted distance value between the terminal to be located and one of its base stations, is the predicted distance value between the terminal to be located and another base station, is the distance value between the two base stations.

[0104] Figure 6 is the flowchart of the training method of the distance prediction model provided by the embodiments of the present application. Refer to Figure 6 In some embodiments, the method includes but is not limited to steps S601 to S608.

[0105] Step S601, obtain the current true distance value between the terminal to be located and the base station.

[0106] Step S602, input the first reference data, the second reference data, and the third reference data into the neural network model to be trained, and obtain the training distance prediction value.

[0107] Step S603, determine the model loss information according to the training distance prediction value and the true distance value.

[0108] Step S604, determine whether the model loss information is within the loss threshold interval.

[0109] If not, execute step S605; if so, execute step S606.

[0110] Step S605, adjust the weight parameters of the neural network model to be trained. Return to step S602.

[0111] Step S606, determine whether the training reset times reach the reset threshold.

[0112] If not, execute step S607; if so, execute step S608.

[0113] Step S607, change the position of the terminal to be located. Return to step S601.

[0114] Step S608, end the training and obtain the distance prediction model.

[0115] In a specific implementation, a loss threshold interval and a reset threshold are preset as the training end conditions. When the model loss information is within the loss threshold interval and the training reset count reaches the reset threshold, the training is ended, and the obtained deep neural network model to be trained in the last iteration is the distance prediction model. When the training reset count has not reached the reset threshold, according to the degree of deviation of the model loss information from the loss threshold interval, the weight parameters of the neural network model to be trained are adjusted using the backpropagation method, so that the model loss information gradually approaches the loss threshold interval during the iteration process and finally falls within the loss threshold interval. Then, the first reference data, the second reference data, and the third reference data are input into the deep neural network model to be trained again until the model loss information is within the loss threshold interval. The position of the terminal to be located is changed, and the current true distance value between the terminal to be located and the base station is obtained again, and the neural network model to be trained is iteratively trained. The above steps are repeated until the training reset count reaches the reset threshold, and then the training is ended to obtain the distance prediction model.

[0116] Please refer to Figure 7 , an embodiment of the present application further provides a personnel positioning device in a tunnel, which can implement the above-mentioned personnel positioning method in the tunnel. The device includes:

[0117] A first module 701, configured to obtain first reference data, second reference data, and third reference data; the first reference data is communication quality data when the terminal to be located and the base station perform long-distance wireless communication, the second reference data is distance measurement data obtained by the terminal to be located and the base station through frequency hopping ranging, and the third reference data is communication quality data when the terminal to be located and the base station perform short-distance wireless communication;

[0118] A second module 702, configured to input the first reference data, the second reference data, and the third reference data into the distance prediction model to predict the distance between the terminal to be located and the base station, and obtain a distance prediction value;

[0119] A third module 703, configured to determine the terminal position of the terminal to be located according to the distance prediction value.

[0120] The specific implementation manner of this personnel positioning device in the tunnel is basically the same as the specific embodiments of the above-mentioned personnel positioning method in the tunnel, and will not be elaborated here.

[0121] Figure 8 is a block diagram of an electronic device shown according to an exemplary embodiment.

[0122] Next, refer to Figure 8 to describe the electronic device 800 according to this embodiment of the present disclosure. Figure 8 The shown electronic device 800 is only an example and should not impose any limitation on the functions and usage scope of the embodiments of the present disclosure.

[0123] As Figure 8 shown, the electronic device 800 is presented in the form of a general-purpose computing device. The components of the electronic device 800 may include, but are not limited to: at least one processing unit 810, at least one storage unit 820, a bus 830 connecting different system components (including the storage unit 820 and the processing unit 810), a display unit 840, etc.

[0124] Among them, the storage unit stores program code, and the program code can be executed by the processing unit 810, so that the processing unit 810 executes the steps according to various exemplary embodiments of the present disclosure described in the above-mentioned tunnel personnel positioning method part of this specification.

[0125] The storage unit 820 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 8201 and / or a cache storage unit 8202, and may further include a read-only storage unit (ROM) 8203.

[0126] The storage unit 820 may further include a program / utilities 8204 having a set (at least one) of program modules 8205. Such program modules 8205 include, but are not limited to: an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include the implementation of a network environment.

[0127] The bus 830 may represent one or more of several types of bus structures, including a storage unit bus or a storage unit controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus using any of the multiple bus structures.

[0128] The electronic device 800 may also communicate with one or more external devices 800' (such as a keyboard, a pointing device, a Bluetooth device, etc.), and may also communicate with one or more devices that enable a user to interact with the electronic device 800, and / or communicate with any device that enables the electronic device 800 to communicate with one or more other computing devices (such as a router, a modem, etc.). Such communication may be performed through an input / output (I / O) interface 850. And, the electronic device 800 may also communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through a network adapter 860. The network adapter 860 may communicate with other modules of the electronic device 800 through the bus 830. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in combination with the electronic device 800, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0129] The embodiment of the present application also provides a computer-readable storage medium storing a computer program, which when executed by a processor, implements the above-mentioned method for locating personnel in a tunnel.

[0130] The method, device, equipment and storage medium for locating personnel in a tunnel provided by the embodiment of the present application utilize a pre-trained distance prediction model to predict the distance between a terminal to be located and a base station based on the communication quality data when the terminal to be located and the base station perform long-distance wireless communication, the distance measurement data obtained by the terminal to be located and the base station through frequency hopping ranging, and the communication quality data when the terminal to be located and the base station perform short-distance wireless communication, so as to locate the personnel in the tunnel. Since the communication quality data obtained by combining the two communication methods of long-distance wireless communication and short-distance wireless communication are mutually verified, and frequency modulation ranging is used to obtain multiple distance measurement data obtained by using different channels, the pre-trained distance prediction model can predict the distance between the terminal to be located and the base station based on reference data in multiple dimensions, enhancing the accuracy and reliability of the distance prediction value and improving the accuracy of personnel location in the tunnel.

[0131] Through the description of the above embodiments, those skilled in the art can easily understand that the example embodiments described herein can be implemented by software or by a combination of software and necessary hardware. Therefore, the technical solutions according to the embodiments of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions for causing a computing device (which can be a personal computer, a server, or a network device, etc.) to execute the above-mentioned method according to the embodiments of the present disclosure.

[0132] The program product can adopt any combination of one or more readable media. The readable media can be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0133] A computer-readable storage medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a readable program code is carried. Such a propagated data signal may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing. The readable storage medium may also be any readable medium other than the computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the readable storage medium may be transmitted by any appropriate medium, including but not limited to wireless, wired, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0134] Those skilled in the art can understand that the above-mentioned modules can be distributed in the device according to the description of the embodiments, or can be correspondingly changed and distributed in one or more devices that are only different from this embodiment. The modules of the above embodiments can be combined into one module, or can be further split into multiple sub-modules.

[0135] The exemplary embodiments of the present disclosure have been specifically shown and described above. It should be understood that the present disclosure is not limited to the detailed structures, settings, or implementation methods described herein; on the contrary, the present disclosure is intended to cover various modifications and equivalent settings included within the spirit and scope of the appended claims.

Claims

1. A method for positioning personnel in a tunnel, characterized in that Including: Obtain first reference data, second reference data, and third reference data; The first reference data is communication quality data when the to-be-located terminal and the base station perform long-distance wireless communication, the second reference data is distance measurement data obtained by the to-be-located terminal and the base station through frequency hopping ranging, and the third reference data is communication quality data when the to-be-located terminal and the base station perform short-distance wireless communication; Input the first reference data, the second reference data, and the third reference data into a distance prediction model to predict the distance between the to-be-located terminal and the base station, and obtain a distance prediction value; Determine the terminal position of the to-be-located terminal according to the distance prediction value; The step of inputting the first reference data, the second reference data, and the third reference data into a distance prediction model to predict the distance between the to-be-located terminal and the base station, and obtaining a distance prediction value includes: Perform local feature extraction on the first reference data, the second reference data, and the third reference data to obtain a local data vector sequence containing a plurality of local data vectors; Perform positional encoding on the local data vector sequence to obtain a positional embedding vector sequence; Perform linear transformation processing on the positional embedding vector sequence, perform attention weight calculation on the local data vectors after the linear transformation processing, and obtain a global distance feature; Predict the distance between the to-be-located terminal and the base station according to the global distance feature, and obtain the distance prediction value.

2. The method for positioning personnel in a tunnel according to claim 1, wherein Before obtaining the first reference data, the second reference data, and the third reference data, it further includes: Obtain a reference data packet forwarded by the base station from the to-be-located terminal; Parse the reference data packet to obtain the first reference data, the second reference data, and the third reference data; the reference data packet is obtained by encapsulating the first reference data, the second reference data, and the third reference data after the to-be-located terminal successively performs long-distance wireless communication, frequency hopping ranging, and short-distance wireless communication with the base station.

3. The method for positioning personnel in a tunnel according to claim 1, characterized in that The generation method of the second reference data is: Determine reference distance data according to a distance measurement data set obtained by frequency hopping ranging; Perform polynomial fitting on the reference distance data to obtain distance measurement fitting data; Judge whether the distance measurement fitting data meets a preset ranging data condition; The preset ranging data condition is that the distance measurement fitting data is within a preset ranging data interval and the frequency error between the to-be-located terminal and the base station is within a preset frequency error interval; If it meets the condition, output the distance measurement fitting data as the second reference data.

4. The method for positioning personnel in a tunnel according to claim 1, wherein The step of determining the terminal position of the to-be-located terminal according to the distance prediction value includes: Within a preset waiting duration, judge whether the distance prediction model outputs two distance prediction values obtained by predicting the distances between the same to-be-located terminal and two base stations; If not, determine the relative direction between the to-be-located terminal and the base station according to historical positioning data, and combine the distance prediction value to determine the terminal position of the to-be-located terminal; If so, calculate the terminal position of the to-be-located terminal according to the two distance prediction values and the distance value between the two base stations.

5. The method for positioning personnel in a tunnel according to claim 1, wherein The training method of the distance prediction model includes: Obtain the current true distance value between the to-be-located terminal and the base station; Input the first reference data, the second reference data, and the third reference data into the to-be-trained deep neural network model to obtain a training distance prediction value; Determine model loss information according to the training distance prediction value and the true distance value; Judge whether the model loss information is within the loss threshold range; If not, adjust the weight parameters of the to-be-trained neural network model; return to the step of inputting the first reference data, the second reference data, and the third reference data into the to-be-trained deep neural network model to obtain a training distance prediction value; If so, judge whether the training reset times reach the reset threshold; If not, change the position of the to-be-located terminal; return to the step of obtaining the current true distance value between the to-be-located terminal and the base station; If so, end the training to obtain the distance prediction model.

6. The method for positioning personnel in a tunnel according to any one of claims 1 to 5, characterized in that, The first reference data includes the long-distance wireless communication frequency error between the to-be-located terminal and the base station, the long-distance wireless communication signal strength value recorded by the base station, the long-distance wireless communication signal strength value recorded by the to-be-located terminal, and the signal-to-noise ratio of the long-distance wireless communication channel; the second reference data is the distance measurement data obtained by frequency hopping ranging when the to-be-located terminal and the base station perform long-distance wireless communication; the third reference data includes the short-distance wireless communication signal strength value recorded by the base station and the short-distance wireless communication signal strength value recorded by the to-be-located terminal.

7. A personnel positioning device in a tunnel, characterized in that, Includes: The first module is used to obtain the first reference data, the second reference data, and the third reference data; The first reference data is the communication quality data when the to-be-located terminal and the base station perform long-distance wireless communication, the second reference data is the distance measurement data obtained by frequency hopping ranging between the to-be-located terminal and the base station, and the third reference data is the communication quality data when the to-be-located terminal and the base station perform short-distance wireless communication; The second module is used to input the first reference data, the second reference data, and the third reference data into the distance prediction model to predict the distance between the to-be-located terminal and the base station to obtain a distance prediction value; The third module is used to determine the terminal position of the to-be-located terminal according to the distance prediction value; The step of inputting the first reference data, the second reference data, and the third reference data into the distance prediction model to predict the distance between the to-be-located terminal and the base station to obtain a distance prediction value includes: Perform local feature extraction on the first reference data, the second reference data, and the third reference data to obtain a local data vector sequence containing multiple local data vectors; Perform position encoding on the local data vector sequence to obtain a position embedding vector sequence; Perform a linear transformation process on the sequence of position embedding vectors, and perform an attention weight operation on the local data vectors after the linear transformation process to obtain global distance features; According to the global distance features, predict the distance between the terminal to be located and the base station to obtain the distance prediction value.

8. An electronic device, characterized in that, The electronic device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the method for positioning personnel in a tunnel according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, the method for positioning personnel in a tunnel according to any one of claims 1 to 6 is implemented.

Images