Short-time startup satellite positioning prediction method applied to intelligent safety helmet

Abstract

The embodiment of the invention provides a short-time startup satellite positioning prediction method and device applied to an intelligent safety helmet, equipment and a computer readable storage medium. The method comprises the following steps: acquiring positioning sequence data; mapping the positioning sequence data to a constructed environment feature matrix to obtain local environment features; inputting the positioning sequence data and the local environment features into a trained prediction model to obtain a positioning prediction value; and completing the prediction of the current position based on the positioning prediction value. In this way, the calculation cost is reduced, and meanwhile the positioning prediction accuracy is greatly improved.

Description

Technical Field

[0001] Embodiments of the present application relate to the field of data processing, and in particular to a short-time startup satellite positioning prediction method, device, equipment, and computer-readable storage medium applied to intelligent safety helmets. Background Art

[0002] With the development of satellite positioning technology and personal protective equipment, enterprises have begun to attempt to install satellite positioning modules on intelligent safety helmets to achieve intelligent management of the positioning and attendance of construction workers.

[0003] However, the power consumption of positioning modules is generally high. If they are turned on for a long time, the battery life of intelligent safety helmets will be sharply shortened. Therefore, they can only be periodically turned on for a short time to collect positioning data and then enter the sleep state. On this basis, due to the influence of multipath effects and satellite signal blockage in the construction environment, the positioning error of short-time startup in actual applications generally exceeds 20 meters, seriously restricting the positioning reliability.

[0004] Currently, methods for improving satellite positioning accuracy include: By installing inertial navigation hardware; Predicting the clock deviation of satellites using an LSTM neural network integrated with a self-attention mechanism; Satellite positioning assisted by long short-term memory networks; Among the above methods, improving the positioning accuracy by installing inertial navigation hardware greatly increases the hardware cost; Predicting the clock deviation of satellites using an LSTM neural network integrated with a self-attention mechanism requires obtaining satellite clock deviation data, which is costly and power-consuming, and is not suitable for mass installation and daily use on safety helmet terminals; Satellite positioning assisted by long short-term memory networks requires terminal devices to have a large amount of data storage space and the ability to perform model training, and is not applicable to safety helmet positioning terminals with limited storage and computing capabilities.

[0005] In summary, how to construct a satellite positioning method with high speed, small size, low computing power requirements, and high accuracy is an urgent problem to be solved at present. Summary of the Invention

[0006] According to the embodiments of the present application, a short-time startup satellite positioning prediction solution applied to intelligent safety helmets is provided. Through an environmental basic deviation correction mechanism, the systematic influence of environmental factors such as weather, ionospheric error, and on-site electromagnetic interference on the positioning of the same area is corrected; the matrix modeling of environmental characteristics is introduced to improve the accuracy of positioning prediction; data processing is performed through full connection in time steps and parameter quantization, greatly reducing the demand for computing resources (only less than 5KB of RAM), making the construction of the overall solution more lightweight.

[0007] In the first aspect of the present application, a short-term startup satellite positioning prediction method applied to an intelligent safety helmet is provided. The method includes: Obtain positioning sequence data; Map the positioning sequence data to the constructed environmental feature matrix to obtain local environmental features; Input the positioning sequence data and local environmental features into the trained prediction model to obtain a positioning prediction value; based on the positioning prediction value, complete the prediction of the current position.

[0008] Further, before inputting the positioning sequence data into the trained prediction model, it further includes: Clean the positioning sequence data to remove outliers in the positioning sequence data to obtain first positioning sequence data; Correct the longitude and latitude in the first positioning sequence data through the obtained environmental deviation value to obtain second positioning sequence data; Perform differential processing on the second positioning sequence data to obtain third positioning sequence data; Perform standardization processing on the third positioning sequence data to compress the numerical range of the third positioning sequence data to a preset range to obtain target positioning sequence data input to the prediction model.

[0009] Further, the step of correcting the longitude and latitude in the first positioning sequence data through the obtained environmental deviation value to obtain second positioning sequence data includes: Take the difference between the first positioning sequence data and the collected target data to obtain a plurality of differences; Calculate the average value of the plurality of differences to obtain an environmental deviation; Based on the environmental deviation, correct the first positioning sequence data to obtain second positioning sequence data.

[0010] Further, the prediction model can be constructed in the following manner: Generate a training sample set, where the training samples include positioning sequence data with annotation information and local environmental features; the annotation information is the positioning prediction value; Use the samples in the training sample set to train the prediction model, use the sample file as the input, and the positioning prediction value as the output. When the unification rate of the output positioning prediction value and the annotated positioning prediction value meets the preset threshold, the training of the prediction model is completed.

[0011] Further, the environmental feature matrix includes: Construct an environmental feature matrix with the collection point as the center according to the construction project drawings; Among them, the value of each grid in the environmental feature matrix is the value after normalizing the building height at that position: ; Among them, is the building height.

[0012] Furthermore, the loss function of the prediction model includes: Calculate the mean absolute error of longitude and latitude respectively; Sum the mean absolute error of longitude and the mean absolute error of latitude to obtain the loss function of the prediction model.

[0013] Furthermore, the mean absolute error of longitude and / or latitude can be calculated in the following manner: ; Among them, is the predicted value of longitude and latitude; is the true value of longitude and latitude.

[0014] In the second aspect of the present application, a short-time startup satellite positioning prediction device applied to an intelligent safety helmet is provided. The device includes: An acquisition module for acquiring positioning sequence data; A processing module for mapping the positioning sequence data to the constructed environmental feature matrix to obtain local environmental features; A positioning module for inputting the positioning sequence data and local environmental features into the trained prediction model to obtain a positioning prediction value; and completing the prediction of the current position based on the positioning prediction value.

[0015] In the third aspect of the present application, an electronic device is provided. The electronic device includes: a memory and a processor, and a computer program is stored on the memory, and when the processor executes the program, the method as described above is implemented.

[0016] In the fourth aspect of the present application, a computer-readable storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, the method according to the first aspect of the present application is implemented.

[0017] The short-time startup satellite positioning prediction method applied to an intelligent safety helmet provided by the embodiments of the present application obtains positioning sequence data; maps the positioning sequence data to the constructed environmental feature matrix to obtain local environmental features; inputs the positioning sequence data and local environmental features into the trained prediction model to obtain a positioning prediction value; and completes the prediction of the current position based on the positioning prediction value, which greatly improves the positioning prediction accuracy while reducing the calculation cost.

[0018] It should be understood that the content described in the Summary of the Invention section is not intended to limit the key or important features of the embodiments of the present application, nor to limit the scope of the present application. Other features of the present application will become easily understandable through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In conjunction with the accompanying drawings and with reference to the following detailed description, the above and other features, advantages, and aspects of the embodiments of the present application will become more apparent. In the drawings, the same or similar reference numerals denote the same or similar elements, where: Figure 1 FIG. is a flowchart of a short-time startup satellite positioning prediction method applied to an intelligent safety helmet according to an embodiment of the present application; Figure 2 FIG. is a schematic diagram of environmental matrix modeling according to an embodiment of the present application; Figure 3 FIG. is a schematic diagram of environmental matching according to an embodiment of the present application; Figure 4 FIG. is a schematic diagram of a neural network model structure according to an embodiment of the present application; Figure 5 FIG. is a block diagram of a short-time startup satellite positioning prediction device applied to an intelligent safety helmet according to an embodiment of the present application; Figure 6 FIG. is a schematic diagram of the structure of a terminal device or a server suitable for implementing the embodiments of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0020] To make the objectives, technical solutions, and advantages of the embodiments of the present disclosure clearer, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present disclosure. Apparently, the described embodiments are some, but not all, of the embodiments of the present disclosure. All other embodiments obtained by those of ordinary skill in the art without making creative efforts based on the embodiments of the present disclosure fall within the scope of protection of the present disclosure.

[0021] In addition, the term "and / or" in this document is merely a description of an association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this document generally represents an "or" relationship between the associated objects before and after.

[0022] Figure 1 FIG. shows a flowchart of a short-time startup satellite positioning prediction method applied to an intelligent safety helmet according to an embodiment of the present disclosure. The method includes: S110, obtaining positioning sequence data.

[0023] In some embodiments, positioning sequence data can be obtained through a positioning module pre - set inside a safety helmet. For example, after starting the positioning module, when initialization is completed and the longitude and latitude are successfully calculated, continuously reading data at a frequency of 1 Hz for 300 seconds is regarded as the completion of a set of data collection, and positioning sequence data is obtained. That is, when the positioning module is started, it actively searches for and captures satellite signals, demodulates and decodes the captured satellite signals, extracts key parameters such as time and orbital parameters, performs calculations to obtain NMEA standard positioning data, and outputs an original positioning data sequence with a time interval of 1 second.

[0024] S120, map the positioning sequence data to the constructed environmental feature matrix to obtain local environmental features.

[0025] In some embodiments, the environment around each collection point can be modeled to output an environmental matrix, so that the model can capture the influence of environmental features on positioning errors. In the present disclosure, taking the collection point as the center, a square area with a side length of 200 meters is modeled (the maximum error after startup does not exceed 100 meters). When modeling, the grid is divided with a granularity of 10 meters, and the 200 - meter * 200 - meter square area is divided into 20 * 20 grids. The value of each grid represents the value of the building height at that position after normalization: ; where is the building height.

[0026] The environmental matrix modeled around a certain collection point is as Figure 2 shown. The blank grid indicates that the building height at that place is 0. Since the center point, that is, the data collection point, has known actual longitude and latitude, the longitude and latitude ranges of each grid can be deduced. That is, the specific position of any given longitude and latitude in the grid can be deduced.

[0027] Furthermore, when the environment is complex, that is, the construction project is not a standard square and the scale is usually greater than 200 meters, actual modeling can be carried out according to the construction project drawings. Similarly, the grid is divided with a granularity of 10 meters, and the building height of each grid is normalized. At this time, the longitude and latitude of the four vertices of the lower - left - hand grid also need to be measured so that the longitude and latitude ranges of each grid can be deduced subsequently, and thus the specific position of any given longitude and latitude in the grid can be deduced.

[0028] It should be noted that the above - mentioned side - length range and the granularity of grid division are only for illustrative purposes, and the specific values can be adjusted according to the actual application scenario.

[0029] In some embodiments, in order to obtain more accurate positioning prediction data and solve problems such as the large absolute value of the longitude and latitude data in the positioning sequence data and the small effective change (for example, the longitude and latitude are 121.456789°E, 31.234567°N, and the change is approximately on the order of 0.00001), the positioning sequence data obtained in step S110 can be cleaned first.

[0030] Specifically, the positioning sequence data is preliminarily cleaned to remove abnormal values ​​in the positioning sequence data to obtain the first positioning sequence data: For example, a total of n valid longitude and latitude values ​​of the previous n seconds (such as the previous 20 seconds) are continuously read at a frequency of 1 Hz. Data cleaning is performed while collecting, and data with a distance threshold that is too large between the previous X seconds and the nth second, and data with completely consistent longitude and latitude in the previous X seconds are discarded, and the preset segment data is taken from the nth second onwards to ensure that the data size is n and the data is available, and the first positioning sequence data is obtained; the X, distance threshold and / or preset segment can be preset according to the actual application scenario.

[0031] Furthermore, a reference device can be deployed at the construction project site. The reference device positioning is always on, and the environmental deviation (the difference between the precise longitude and latitude and the positioning test longitude and latitude) is reported to the server once a minute. After the hard hat end completes the positioning data collection, it obtains the latest environmental deviation value of the current construction project from the server end. The longitude and latitude in the first positioning sequence data are corrected by the acquired environmental deviation value to obtain the second positioning sequence data; compared with the prior art, the present disclosure corrects the systematic impact of environmental factors such as weather, ionospheric errors, and on-site electromagnetic interference on the positioning of the same area: The environmental deviation value can be determined by the reference value measured by the RTK device. That is, the environmental deviation is obtained by subtracting and averaging the reference value measured by the reference device and the RTK device; the reference device is a positioning module of a helmet, which is fixed and works continuously, and the real value of its deployment position can be obtained by RTK measurement: ; ; in, and The latitude and longitude information collected by the PTK equipment respectively; and The first latitude and longitude information respectively; and The second latitude and longitude information respectively; The first positioning sequence data is corrected based on the environmental deviation to obtain the second positioning sequence data: ; ; wherein, and are the latitude and longitude collected by the experimental equipment (the positioning module in the safety helmet), respectively.

[0032] Further, perform differential processing on the second positioning sequence data to obtain third positioning sequence data. That is, the coordinates collected in the first n seconds (for example, 20 seconds) of each group of data can be added to the real coordinates, and the longitude and latitude of a total of n + 1 pieces of data are subtracted in sequence to be converted into the longitude and latitude change amounts between adjacent points , , which is used to eliminate the influence of absolute coordinates and enhance the sensitivity of the model.

[0033] Further, perform normalization processing on the third positioning sequence data to compress the numerical range of the third positioning sequence data to a preset range to obtain the target positioning sequence data input to the prediction model. For example, perform Min - Max normalization on the third positioning sequence data to compress the numerical range to the interval [-1, 1] to accelerate the convergence of the model: ; ; ; ; ; ; In some embodiments, as Figure 3 shown, map the longitude and latitude of each time step in the target positioning sequence data to the corresponding grid of the environment matrix, and read the values of the grids adjacent to the upper, lower, left, and right of this grid as the 4 environmental feature vectors (North_i, South_i, West_i, East_i) of this time step.

[0034] S130, input the positioning sequence data and local environmental features into the trained prediction model to obtain a positioning prediction value; based on the positioning prediction value, complete the prediction of the current position.

[0035] In some embodiments, the prediction model can be trained in the following manner: Generate a training sample set, wherein the training samples include positioning sequence data with annotation information and local environmental features; the annotation information is the positioning prediction value; Use the samples in the training sample set to train the prediction model, with the sample file as the input and the positioning prediction value as the output. When the unification rate of the output positioning prediction value and the labeled positioning prediction value meets the preset threshold, the training of the prediction model is completed.

[0036] Further, when generating the training sample set, in order to improve the accuracy of the neural network model positioning and prevent the influence of the inherent errors of a single device on the data set. In the present disclosure, 7 positioning modules of the same model are used for data collection, among which 5 are used for data collection, and the other 2 modules are always on and used as environmental reference devices for measuring environmental deviations. The device is placed statically on a wooden frame 1.5 m high for positioning data collection to simulate the height of the positioning device when construction workers actually wear safety helmets. When collecting data, after starting the positioning module, place it on the wooden frame. After the module initialization is completed and the longitude and latitude are successfully calculated, continuously read the data for 300 seconds at a frequency of 1 Hz as a set of data collection completed. After the data collection at a point is completed, use the RTK device to measure the longitude and latitude of the device placement position. The RTK error is less than 0.1 m, which is two orders of magnitude lower than the error of the positioning module (at the 10 m level) and can be ignored. Therefore, the RTK measurement value can be directly used as the true value of the device position. In order to cover various environments during construction, the present disclosure selects data collection points between 120° east longitude and 122° east longitude, and 31° north latitude and 32° north latitude. The number of data collection points is not less than 10, covering various environments such as open areas, single-side close buildings, both-side close buildings, surrounded buildings, and the edge of dense woods. In order to collect as much data as possible for model training, 5 rounds of data collection are carried out at each of the 10 collection points. Each round collects 10 times, and each time 5 devices each collect a set of positioning data for 300 seconds after startup. Finally, a total of 2500 sets of data are collected for neural network model training. The always-on module reads the measured longitude and latitude after 1 hour of startup during each round of collection at each collection point, and uses RTK to collect its accurate longitude and latitude.

[0037] In some embodiments, the prediction model of the present disclosure is as Figure 4 shown, which is a multi-layer neural network structure. The input layer contains 19 time steps between the 1st and 20th seconds, with 6 nodes in each time step, accumulating 114 nodes. The first hidden layer is a feature cross layer, divided into 19 time steps, with 8 nodes in each time step. The 8 nodes inside each time step are fully connected to the 6 nodes in the corresponding time step of the input layer, and there is no connection between different time steps. The second hidden layer is a fully connected layer, with a total of 16 nodes, which are fully connected to the 8×19 = 152 nodes of the feature cross layer. The output layer is a fully connected layer, with a total of 2 nodes, representing the true position longitude and latitude 。The loss function of the model uses the mean absolute error (MAE), and calculates the MAE for longitude and latitude respectively: ; Among them, is the predicted value of longitude and latitude; is the true value of longitude and latitude; The sum of the longitude MAE and the latitude MAE is used as the loss function of the model.

[0038] The algorithm of the present disclosure adopts a method of fully connected with time steps and parameter quantization, which only requires less than 5KB of RAM, is lighter and has lower computing resource requirements, and is suitable for embedded hardware.

[0039] Furthermore, after the model training and verification are completed, it can be deployed to run on the Quectel Air820UG communication module, which integrates the Beidou / GPS / GLONASS multi-mode satellite positioning function. First, the method provided by pytorch is used to dynamically quantize the model, and then it is exported as an ONNX file, which contains quantization information. Finally, the ONNX2Lua toolchain is used to convert the model into LUA language and burn it into the embedded hardware, thus completing the model download. In addition, the environmental matrix after the construction project modeling must also be stored in the terminal. After the model download is completed, the predict method is implemented in LUA language on the safety helmet side to perform positioning prediction. This method receives the positioning sequence transmitted by the positioning module, performs differential and normalization preprocessing, then performs environmental matching and feature splicing, uses the spliced vector as the model input, and then performs lightweight forward propagation calculation through the downloaded model parameters to obtain the model prediction value. Finally, inverse normalization and inverse difference are performed to obtain the predicted value of longitude and latitude, thus realizing the model operation. That is, the above-generated positioning data sequence and environmental matrix are input into the predict prediction algorithm implemented on the safety helmet side, and the predicted longitude and latitude are output.

[0040] Furthermore, the ROM and RAM occupied by the prediction model can be evaluated as follows. In the feature cross layer, there are 6×8=48 weight parameters and 8 bias parameters for each time step, totaling 56. For 19 time steps, there are 56×19=1064 parameters. The fully connected layer has 152×16+16=2448 parameters. The output layer has 16×2+2=34 parameters. There are 3564 model parameters, which are stored in 8-bit fixed-point after quantization, occupying about 3.5KB of ROM. Taking a large construction site of 1km*1km as an example, the environment matrix has 10,000 grid parameters, which are stored in 8-bit fixed-point after quantization, occupying about 9.8KB of ROM. A total of 13.3KB of ROM is required. When running the algorithm in memory, first read the 3.5KB model parameters into memory, then the input layer 6×19=114 variables, the feature cross layer 8×19=152 variables, the fully connected layer 16 variables, the output layer 2 variables, a total of 282 temporary variables, quantized and stored in 32-bit fixed point, occupying about 1.1KB of RAM, a total of 4.6KB of RAM. Both ROM and RAM occupancy are within the hardware parameter range of Hezhou Air820UG. It can be seen that the prediction model constructed by the present disclosure can run smoothly on the helmet terminal.

[0041] According to the embodiments of the present disclosure, the following technical effects are achieved: Through the neural network, the law between the positioning change trend of short-term power-on in historical data and the real position is captured, and finally the positioning prediction is realized, which improves the positioning accuracy of the satellite positioning module when it is powered on for a short time; The environmental feature training model is used to fully capture the regularity between the positioning changes and the real coordinates of short-term startups under different environmental features, greatly improving the model prediction effect; The prediction model is lightweight, has fast calculation speed, small size, low computing power requirements, and is suitable for embedded hardware.

[0042] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily required by the present application.

[0043] The above is an introduction to the method embodiment. The following is a further explanation of the scheme described in this application through an apparatus embodiment.

[0044] Figure 5FIG. 500 is a block diagram of a short-term startup satellite positioning prediction device applied to an intelligent safety helmet according to an embodiment of the present application, as Figure 5 shown including: An acquisition module 510, configured to acquire positioning sequence data; A processing module 520, configured to map the positioning sequence data to a pre-constructed environmental feature matrix to obtain local environmental features; A positioning module 530, configured to input the positioning sequence data and local environmental features into a pre-trained prediction model to obtain a positioning prediction value; and complete the prediction of the current position based on the positioning prediction value.

[0045] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the described modules can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0046] Figure 6 FIG. shows a schematic structural diagram of a terminal device or a server suitable for implementing the embodiments of the present application.

[0047] As Figure 6 shown, the terminal device or the server includes a central processing unit (CPU) 601, which can execute various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage section 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the terminal device or the server are also stored. The CPU 601, the ROM 602, and the RAM 603 are connected to each other through a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0048] The following components are connected to the I / O interface 605: an input section 606 including a keyboard, a mouse, etc.; an output section 607 including, for example, a cathode ray tube (CRT), a liquid crystal display (LCD), etc. and a speaker, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN card, a modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the I / O interface 605 as required. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as required, so that a computer program read from it can be installed into the storage section 608 as required.

[0049] In particular, according to an embodiment of the present application, the above method flow steps can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a machine-readable medium, and the computer program includes program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 609, and / or installed from the removable medium 611. When the computer program is executed by the central processing unit (CPU) 601, the above functions defined in the system of the present application are executed.

[0050] It should be noted that the computer-readable medium shown in the present application can be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. The computer-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 of the computer-readable storage medium can include, but are not limited to: an electrical connection with one or more wires, a portable computer 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. In the present application, the computer-readable storage medium can be any tangible medium that contains or stores a program, and the program can be used by or in combination with an instruction execution system, apparatus, or device. In the present application, the computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, which carries the computer-readable program code. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium can also be any computer-readable medium other than the computer-readable storage medium, and the computer-readable medium can send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted by any suitable medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0051] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagram may represent a module, a segment of a program, or a portion of code that contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than that marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, as well as combinations of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system that performs the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0052] The units or modules involved in the embodiments described in the present application can be implemented in software or in hardware. The described units or modules can also be provided in a processor. Among them, the names of these units or modules do not, in some cases, constitute a limitation on the units or modules themselves.

[0053] As another aspect, the present application also provides a computer-readable storage medium, which can be included in the electronic device described in the above embodiments; or can exist separately without being assembled into the electronic device. The above computer-readable storage medium stores one or more programs, and when the foregoing programs are executed by one or more processors, the methods described in the present application are implemented.

[0054] The above description is only a preferred embodiment of the present application and an explanation of the technical principles applied. Those skilled in the art should understand that the scope of the application involved in the present application is not limited to the technical solutions formed by the specific combination of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the foregoing inventive concept. For example, the technical solutions formed by mutually replacing the above features with the technical features (but not limited to) having similar functions described in the present application.

Claims

1. A short-time startup satellite positioning prediction method applied to an intelligent safety helmet, characterized in that, Including: Obtain positioning sequence data; Map the positioning sequence data to the constructed environmental feature matrix to obtain local environmental features; Input the positioning sequence data and the local environmental features into the trained prediction model to obtain a positioning prediction value; based on the positioning prediction value, complete the prediction of the current position.

2. The method according to claim 1, wherein Before inputting the positioning sequence data into the trained prediction model, it further includes: Clean the positioning sequence data to remove outliers in the positioning sequence data to obtain first positioning sequence data; Correct the longitude and latitude in the first positioning sequence data through the obtained environmental deviation value to obtain second positioning sequence data; Perform differential processing on the second positioning sequence data to obtain third positioning sequence data; Perform normalization processing on the third positioning sequence data to compress the numerical range of the third positioning sequence data to a preset range to obtain the target positioning sequence data input to the prediction model.

3. The method according to claim 2, wherein The step of correcting the longitude and latitude in the first positioning sequence data through the obtained environmental deviation value to obtain second positioning sequence data includes: Subtract the first positioning sequence data from the collected target data to obtain multiple differences; Calculate the average value of the multiple differences to obtain an environmental deviation; Based on the environmental deviation, correct the first positioning sequence data to obtain second positioning sequence data.

4. The method according to claim 3, wherein The prediction model can be trained in the following manner: Generate a training sample set, where the training samples include positioning sequence data with annotation information and local environmental features; the annotation information is the positioning prediction value; Use the samples in the training sample set to train the prediction model, use the sample file as the input, and the positioning prediction value as the output. When the unification rate of the output positioning prediction value and the annotated positioning prediction value meets the preset threshold, complete the training of the prediction model.

5. The method according to claim 4, wherein The environmental feature matrix includes: Construct an environmental feature matrix centered on the collection point according to the construction project drawings; Among them, the value of each grid in the environmental feature matrix is the value after normalizing the building height at that position: ； Among them, is the building height.

6. The method according to claim 5, characterized in that, The loss function of the prediction model includes: Calculate the mean absolute error of longitude and latitude respectively; Sum the longitude mean absolute error and the latitude mean absolute error to obtain the loss function of the prediction model.

7. The method according to claim 6, characterized in that, The mean absolute error of longitude and / or latitude can be calculated in the following manner: ； Among them, is the predicted value of longitude and latitude; is the true value of longitude and latitude.

8. A short-time startup satellite positioning prediction device applied to an intelligent safety helmet, characterized in that, Including: An acquisition module for acquiring positioning sequence data; A processing module for mapping the positioning sequence data to the constructed environmental feature matrix to obtain local environmental features; A positioning module for inputting the positioning sequence data and the local environmental features into the trained prediction model to obtain a positioning prediction value; based on the positioning prediction value, complete the prediction of the current position.

9. An electronic device, comprising a memory and a processor, wherein a computer program is stored on the memory, characterized in that, When the processor executes the computer program, it implements the method according to any one of claims 1 to 7.

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

Images