Indoor and outdoor cooperative positioning method, device, equipment and medium

Abstract

The invention discloses an indoor and outdoor cooperative positioning method, which relates to the technical field of high-precision positioning and is used for solving the problem of inaccurate positioning of an existing buffer area, and the method comprises the following steps: obtaining positioning data of Bluetooth positioning and Beidou positioning; performing unified processing on the positioning data to enable different positioning data to follow the same data standard; performing feature extraction on the positioning data after the unified processing to obtain feature data; according to the feature data, training through a BP neural network to obtain a positioning fusion model; and positioning the buffer area according to the positioning fusion model. The invention further discloses an indoor and outdoor cooperative positioning device, electronic equipment and a computer storage medium. According to the invention, positioning fusion is carried out through the BP network, so that the buffer area is accurately positioned.

Description

Technical Field

[0001] The present invention relates to the field of high-precision positioning technology, and in particular to an indoor and outdoor collaborative positioning method, device, equipment and medium based on Beidou and Bluetooth technology. Background Art

[0002] The on-site working environment of power grids is complex and changeable, with many operators and overlapping work surfaces. On-site operators face high safety risks and usually need to rely on manual on-site control. Especially when the power grid is responding to major natural disasters such as floods, typhoons, and earthquakes, there are many on-site points, a wide range, a large scale, tight time, high work intensity, and high safety hazards.

[0003] There are the following difficulties in ensuring the safety of emergency workers at the scene: (1) There are many people at the scene of power disasters, the composition of the repair team is complex, it is difficult for personnel to obtain location information, and there are safety hazards in the coordination process; (2) During the on-site power emergency response, the on-site environment is complex, and it is difficult to obtain information on dangerous points and various resources; (3) The ways for emergency personnel to obtain response information and assistance at the scene are very limited, and the communication between the emergency command center and the site is not smooth, which affects the efficiency of emergency on-site repairs.

[0004] At present, on-site positioning of power operations is usually based on personnel wearing terminals, using single positioning methods such as Bluetooth, ultra-wideband UWB, Beidou navigation, and WIFI for positioning. These positioning methods have problems such as low positioning accuracy and large hardware deployment when positioning alone, and are not suitable for rapid deployment. Especially when the staff is located in a buffer zone where multiple positioning methods overlap, the positioning error is large. Considering the characteristics of high flexibility of on-site operations, wide positioning range, and high positioning accuracy requirements, there is an urgent need for a high-precision positioning method that integrates multiple positioning technologies for indoor and outdoor collaborative positioning. Summary of the invention

[0005] In order to overcome the shortcomings of the prior art, one of the objects of the present invention is to provide an indoor and outdoor collaborative positioning method, which obtains the positioning information of the buffer zone by fusing the positioning data of Bluetooth positioning and Beidou positioning.

[0006] One of the purposes of the present invention is achieved by the following technical solution:

[0007] An indoor and outdoor collaborative positioning method comprises the following steps:

[0008] Obtain positioning data of Bluetooth positioning and Beidou positioning;

[0009] Performing unified processing on the positioning data so that different positioning data follow the same data standard;

[0010] Extracting features from the unified positioning data to obtain feature data;

[0011] According to the feature data, a positioning fusion model is obtained by training a BP neural network;

[0012] The buffer zone is positioned according to the positioning fusion model.

[0013] Furthermore, the positioning data includes indoor and outdoor location coordinates and signal strength.

[0014] Furthermore, the positioning data is processed in a unified manner, including:

[0015] Perform outlier detection based on the 3α principle and remove abnormal data;

[0016] The filtered normal positioning data is normalized, and the positioning data is scaled or expanded to a preset range.

[0017] Further, feature extraction is performed on the unified positioning data, including:

[0018] Extracting Bluetooth signal strength value and Beidou signal strength value from the positioning data;

[0019] The Beidou positioning prior error is calculated using the least squares method.

[0020] Furthermore, the Beidou positioning a priori error includes satellite clock error, receiver clock error, and atmospheric delay; calculating the Beidou positioning a priori error by least square method to estimate the position of the receiver includes:

[0021] Model the error and calculate the error e for the data points i , satisfying e i =y i -f(x i ), where (x i ,y i ) is the data point, f(x) is the fitting function;

[0022] Adjust the model parameters to meet the following requirements: Among them, a is a parameter;

[0023] According to the adjusted model parameters, the error data is fitted through the fitting function.

[0024] Furthermore, the input vector of the BP neural network includes feature data of Beidou and Bluetooth positioning, and the output vector includes label values ​​of the Bluetooth positioning system and the Beidou positioning system; the number of neurons in the input layer of the BP neural network is 3, the number of neurons in the output layer is 3, and the number of neurons x in the hidden layer satisfies:

[0025] Among them, m is the number of neurons in the input layer, and n is the number of neurons in the output layer.

[0026] Furthermore, the BP neural network also includes a variable learning rate and additional momentum for model optimization, and the variable learning rate l satisfies:

[0027]

[0028] The weight of the additional momentum satisfies:

[0029] Δwij(t)=η×Δwij(t-1)+lo×Errj×Oi,

[0030] Among them, η is a constant between [0,1], l is the learning rate, Errj is the error of the next layer node, Oi is the output of the previous layer node, and t represents the number of iterations.

[0031] A second object of the present invention is to provide an indoor and outdoor collaborative positioning device.

[0032] The second object of the present invention is achieved by adopting the following technical solution:

[0033] An indoor and outdoor collaborative positioning device, comprising:

[0034] Acquisition module, used to obtain positioning data of Bluetooth positioning and Beidou positioning;

[0035] A data processing module is used to perform unified processing on the positioning data so that different positioning data follow the same data standard; and perform feature extraction on the positioning data after the unified processing to obtain feature data;

[0036] A model training module is used to train a BP neural network according to the feature data to obtain a positioning fusion model;

[0037] A positioning module is used to locate the buffer zone according to the positioning fusion model.

[0038] The third object of the present invention is to provide an electronic device for performing one of the objects of the invention, which includes a processor, a storage medium and a computer program, wherein the computer program is stored in the storage medium, and when the computer program is executed by the processor, the above-mentioned indoor and outdoor collaborative positioning method is implemented.

[0039] A fourth object of the present invention is to provide a computer-readable storage medium storing one of the objects of the invention, on which a computer program is stored, and when the computer program is executed by a processor, the above-mentioned indoor and outdoor collaborative positioning method is implemented.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] The present invention realizes high-precision positioning of the buffer zone by integrating Bluetooth positioning and Beidou positioning technology; the present invention realizes positioning fusion between Bluetooth and Beidou positioning systems and realizes smooth switching of different positioning through the output result of the BP neural network model, realizes seamless switching of indoor and outdoor environment positioning, solves the positioning blind spot problem in the indoor and outdoor transition area, and significantly improves the accuracy and reliability of positioning. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 This is a schematic diagram of Beidou carrier phase difference positioning technology in Example 1;

[0043] Figure 2 is a flow chart of the indoor and outdoor collaborative positioning method of embodiment 1;

[0044] Figure 3 is a schematic diagram of the indoor and outdoor unified coordinate system of the first embodiment;

[0045] Figure 4 is a structural block diagram of an indoor and outdoor collaborative positioning method and device according to Embodiment 2;

[0046] Figure 5 It is a structural block diagram of an electronic device of Embodiment 3. DETAILED DESCRIPTION

[0047] The present invention will be described in more detail below in conjunction with the accompanying drawings. It should be noted that the following description of the present invention with reference to the accompanying drawings is only illustrative and not restrictive. The various embodiments can be combined with each other to form other embodiments not shown in the following description.

[0048] Embodiment 1

[0049] Embodiment 1 provides an indoor and outdoor collaborative positioning method, which aims to obtain accurate buffer zone positioning by fusing positioning data of multiple positioning methods.

[0050] The positioning technology of this embodiment mainly involves Bluetooth positioning and Beidou positioning. Beidou high-precision positioning technology and Bluetooth AOA positioning technology have their own advantages and disadvantages. Beidou high-precision positioning technology has the common problems of satellite positioning technology, and there are situations where satellite positioning is difficult or impossible. The accuracy of Bluetooth AOA positioning depends on the quality of the Bluetooth signal. When the accuracy of positioning cannot be guaranteed by using a single positioning data, the method of this embodiment aims to perform data fusion on the effective positioning in multiple scenarios of the above two positioning data, avoid the scenario where one positioning technology fails and cannot be positioned, and output high-precision and reliable positioning coordinates.

[0051] The Beidou positioning module uses the positioning technology of the Beidou satellite navigation system. The Beidou satellite navigation system consists of three parts: the space segment, the ground segment and the user segment. It can provide high-precision, high-reliability positioning, navigation and timing services for all types of users around the world all day and all day, and has short message communication capabilities. It has initially acquired regional navigation, positioning and timing capabilities, with a positioning accuracy of 10 meters, a speed measurement accuracy of 0.2 meters per second, and a timing accuracy of 10 nanoseconds. With the help of a positioning system consisting of 5 geostationary orbit satellites, 27 medium earth satellites, and 3 inclined synchronous orbit satellites, its positioning accuracy can reach a horizontal accuracy of 100 meters (1σ), and 20 meters after the establishment of a calibration station (similar to the differential state).

[0052] The principle of Beidou satellite positioning is to establish a satellite network composed of multiple satellites in space. Through the rational design of satellite orbit distribution, users can observe at least three satellites at any location on the earth. Since the position of a satellite is fixed at a specific moment, users only need to measure the distance to them to calculate their own coordinates.

[0053] Due to various errors in the satellite positioning process, such as orbit error, clock error, ionospheric delay, tropospheric delay, receiver noise and multipath effects, the satellite positioning accuracy can only reach the meter level.

[0054] To achieve higher-precision positioning, differential positioning is used. On the basis of satellite positioning, a ground base station is set up. The ground base station obtains satellite positioning and compares it with the real position (since the base station is fixed, the position is absolute), calculates the positioning error, and sends the current error to the mobile terminal. The mobile terminal will use this error to correct the positioning obtained from the satellite. In this scenario, Beidou positioning technology is used at the center point of the communication network and the terminal node to locate the location information of the base station and the terminal, and the differential correction number is calculated based on the absolute coordinates of the base station, which is sent to each terminal node in real time to correct the positioning error, achieving centimeter-level positioning accuracy.

[0055] However, Beidou positioning also has inevitable limitations. The first is trajectory recalculation, which refers to the position point drift caused by signal obstruction in elevated areas or urban canyons; the second is the inability to locate, which refers to the low accuracy of inference in signal-free areas (indoors, canyons), resulting in large positioning errors; the third is height positioning errors, which refers to the combined gap distance must not be less than 9.6 meters when equipotential workers enter and exit strong electric fields. Workers are not allowed to enter the crossarms of three-phase wires at the same time. Among them, the direct cause of trajectory recalculation and inability to locate is the poor accuracy of Beidou positioning and the poor inference accuracy. Beidou positioning accuracy is determined by the observation environment and is difficult to improve; especially the calibration accuracy of trajectory recalculation and height positioning.

[0056] In order to overcome these limitations of satellite positioning, Bluetooth AOA positioning technology is used to locate the terminal when satellite positioning alone is difficult or impossible.

[0057] The technologies involved include the angle of arrival (AoA), which is the angle of the signal relative to the coordinates of the receiving antenna. The measurement of the angle of arrival depends on the difference in the phase of the signal sampled at a certain time T due to the different antenna positions during the transmission process. Bluetooth signals operate in the ISM (Industrial Scientific and Medical) 2.40GHz to 2.41GHz band with a bandwidth of 2MHz. Bluetooth is divided into three broadcast channels, 37, 38, and 39. In the Bluetooth 5.x specification, the extended broadcast channel of the Bluetooth LE part can be any channel from 0 to 39, which means that Bluetooth beacons can operate on any Bluetooth channel.

[0058] AOA (Ace Of Angels, AOA) Bluetooth positioning uses a Bluetooth positioning base station with a fixed position and direction. After obtaining the direction of arrival of the terminal signal, the position can be calculated by AOA and RSSI positioning algorithms. By measuring the direction of signal arrival and the signal strength (Received Signal Strength Indication, RSSI), the location of the device can be estimated together. Please refer to Figure 1 As shown in the figure, the ray from the base station intersects with the height plane where the terminal device is located, thereby determining its coordinates on the plane. The positioning range of a single base station is distributed in a cone shape, with the base station as the center. The angle error has a more significant impact on the plane coordinates at a distance from the base station, so the position error is also larger; while the positioning accuracy is higher at a distance close to the base station.

[0059] Based on the above, the positioning system needs to model the actual scenario, deploy receiver routers in the observation area, and mark the corresponding location in the modeling map; the mobile terminal acts as a Bluetooth signal transmitter to send wireless signals, and the receiver deployed at a fixed location can determine that the terminal has entered the area after receiving the wireless signal.

[0060] After determining the area, the precise location of the mobile terminal can be calculated by measuring the signal arrival angle through multiple antennas and using the triangulation positioning method. As the name suggests, the triangulation positioning method determines the location of the mobile terminal through two sets of arrival angles. Two sets of antennas can measure one arrival angle, and two arrival angles can determine the terminal location.

[0061] Each Bluetooth base station can calculate the azimuth and pitch angle of the helmet through the received signal. Before information processing, the points with weak signal strength are discarded to avoid measurement errors caused by too weak signals. Through the spatial geometric relationship, it can be seen that the azimuth and pitch angles measured by the two groups of base stations can solve a spatial coordinate point of the helmet. Multiple groups of Bluetooth device data in the space are combined in pairs to obtain a group of coordinate points of the helmet. The points with large offsets are eliminated and the remaining coordinate points are averaged to improve the test accuracy.

[0062] According to the valid coordinates obtained by the above method, this embodiment uses BP neural network to intelligently fuse different location information sources in multiple scenarios. Multiple scenarios include Beidou positioning information and Bluetooth positioning information when the Beidou and Bluetooth information are good; Bluetooth positioning information when the Beidou signal can only observe less than 2 satellites; Beidou positioning information when the Bluetooth signal is weak.

[0063] The above BP neural network algorithm is a supervised learning algorithm. Its main idea is to learn the known actual positioning information and the detection positioning information sample group by using gradient search technology, and finally achieve the minimum mean square error between the actual output value of the detection positioning information and the expected output value. BP neural network is a feedforward neural network, which is the most widely used neural network at present. It generally includes three parts: input layer, several hidden layers and output layer.

[0064] In the learning process of BP neural network, the input signal is processed layer by layer from the input layer through the hidden layer unit and transmitted to the output layer of BP neural network. Each layer of BP neurons only affects the state of the neurons in the next layer. If the expected output is not obtained in the output layer of BP neural network, then the BP neural network will be back propagated to return the error of the output signal along the original connection path, and the mean square error will be minimized by continuously adjusting the weights of neurons in each layer.

[0065] Before network training, the data must first be normalized to ensure that the network layer output is not too small. Here, the initial value pi of the center vector is determined by the training sample, where pi = (pi1, pi2, pi3, pi4, pi5), i = 1, 2, ..., 10. If it belongs to the training sample set of the first type of data change, represented as {S1, S2, ..., Sm}, the initial value of each element in the center vector p1 is the average value of each element in these sample input vectors. Then the gradient descent method is used to adjust the center vector, width, coefficient and weight of the last layer respectively. During the training process, a larger learning rate is first selected. If the training error of the next step is greater than the training error of the previous step, the learning rate is reduced to make the network converge. Due to the addition of the prior information of the center vector of each category, the adjustment of the network weight is limited to a certain range to avoid falling into the local minimum point.

[0066] Beidou carrier phase differential technology (Real-Time Kinematic, RTK) positioning system is a high-precision satellite navigation positioning system. The system obtains floating point solutions in real time and accurately calculates the position of the mobile station through differential correction of the reference station (differential base station). Its principles mainly include the following aspects:

[0067] 1. Satellite positioning principle: The system uses the signals transmitted by Beidou satellites to determine the location of the mobile station. Beidou satellites are distributed in the earth's orbit. The mobile station receives signals from multiple satellites and calculates and processes the signals to determine its own position.

[0068] 2.RTK positioning principle: The reference station sends the carrier phase observation value and its own coordinate information to the mobile station, and the mobile station calculates the carrier phase correction value by itself. The mobile station can correct its position according to the differential correction value to achieve the purpose of high-precision positioning.

[0069] 3. Real-time dynamic principle: The system can receive satellite signals in real time, process the signals quickly, and calculate the position of the mobile station in real time. At the same time, because the system adopts dynamic positioning, it can track the movement trajectory of the target in real time and has certain dynamic monitoring capabilities.

[0070] According to the above principles, please refer to Figure 2 As shown, an indoor and outdoor collaborative positioning method includes the following steps:

[0071] S1. Obtain positioning data of Bluetooth positioning and Beidou positioning;

[0072] The above positioning data includes indoor and outdoor location coordinates and signal strength.

[0073] S2. Performing unified processing on the positioning data so that different positioning data follow the same data standard;

[0074] In this embodiment, the Beidou coordinate system is used as the reference system. Since the location information obtained by the Bluetooth positioning technology using the solution platform is relative location information, it is impossible to directly obtain the absolute location information. Therefore, it is necessary to convert the Bluetooth coordinate system into the Beidou coordinate system to construct a unified indoor and outdoor coordinate system. Please refer to Figure 3 As shown in the figure, the indoor coordinate system is the original Bluetooth positioning coordinate system, and WGS64 is the Beidou positioning coordinate system. Suppose the coordinate point to be converted in the Beidou positioning coordinate system is (x, y), and the corresponding coordinate point to be converted in the Bluetooth positioning coordinate system is (x1, y1), then there is a corresponding relationship between the two coordinate systems: where M is the scale factor, w and h are the coordinate systems of the map, and x0 and y0 are the coordinates of the reference point.

[0075]

[0076] S2 performs unified processing on the positioning data, including:

[0077] Perform outlier detection based on the 3α principle and remove abnormal data;

[0078] The filtered normal positioning data is normalized, and the positioning data is scaled or expanded to a preset range.

[0079] The above-mentioned outlier detection by the 3α principle is a commonly used statistical method for identifying outliers in a data set. This method is based on the characteristics of normal distribution, that is, in the case of normal distribution, most of the data (about 99.7%) will fall within the range of ±3 times the standard deviation of the mean. If the data points are outside this range, they may be considered outliers.

[0080] S3, extracting features from the unified positioning data to obtain feature data;

[0081] S3 performs feature extraction on the unified positioning data, including:

[0082] Extracting Bluetooth signal strength value and Beidou signal strength value from the positioning data;

[0083] The Beidou positioning prior error is calculated using the least squares method.

[0084] The Bluetooth signal strength value (RSSI) mentioned above can be directly extracted. It is expressed as decibel milliwatts (dBm) and is used to determine the link quality and whether the broadcast transmission strength needs to be increased. The Beidou signal strength value can also be extracted.

[0085] The Beidou positioning prior error includes satellite clock error, receiver clock error, and atmospheric delay. In the single point positioning (SPP) algorithm, these errors are modeled and an attempt is made to estimate the receiver's position through the least squares optimization method.

[0086] The basic idea of ​​the least squares method is:

[0087] (1) Sum of squared errors: First, define an error function, which is usually the square of the difference between the observed value and the model predicted value. For a set of data points (x1, y1), (x2, y2), …, (xn, yn), the function f(x) can be used to fit this set of data.

[0088] (2) Minimize the error by adjusting the parameters in the model so that the sum of squared errors of all data points is minimized.

[0089] (3) Solving parameters: Through the above optimization, a set of parameters a can be obtained so that the model f(x) best fits the data.

[0090] Specifically, the Beidou positioning prior error is calculated by the least squares method to estimate the position of the receiver, including:

[0091] Model the error and calculate the error e for the data points i , satisfying e i =y i -f(x i ), where (x i ,y i ) is the data point, f(x) is the fitting function;

[0092] Adjust the model parameters to meet the following requirements: Among them, a is a parameter;

[0093] According to the adjusted model parameters, the error data is fitted through the fitting function.

[0094] S4, training the BP neural network according to the feature data to obtain a positioning fusion model;

[0095] S4 also needs to divide the data into training set and test set before training, for example, the training set accounts for 70% of the total data set, and the test set accounts for 30% of the total data set. The training set has been optimized through data preprocessing, and the test set is used to continuously adjust the weights and biases of the model through the back propagation algorithm and gradient descent optimization algorithm to enable it to better fit the data.

[0096] The input vector of the BP neural network includes the feature data of Beidou and Bluetooth positioning, and the output vector includes the label values ​​of the Bluetooth positioning system and the Beidou positioning system; the number of neurons in the input layer of the BP neural network is 3, the number of neurons in the output layer is 3, and the number of neurons x in the hidden layer satisfies:

[0097] Among them, m is the number of neurons in the input layer, and n is the number of neurons in the output layer.

[0098] The neural network uses cross entropy as the cost function, and the output layer neurons use the softmax function as the activation function. The output vector is a 1×3 row vector, which is used to represent the probability of the neural network selecting different positioning systems. The number of neurons in the hidden layer is 4, and the activation function is the sigmoid function.

[0099] It should be noted that, in the whole positioning process, there are three different positioning states. When in the indoor area or the outdoor area, since there is only a single positioning system, both the indoor area and the outdoor area can obtain stable positioning results. However, in the indoor and outdoor buffer areas, there are different positioning systems, which may cause the occurrence of positioning jumps. At this time, it is necessary to fuse the positioning data of the two positioning systems to obtain more accurate positioning coordinates. Considering the excellent performance of the BP neural network in positioning, therefore, the BP neural network is used in the indoor and outdoor buffer areas, and the three positioning states of the indoor area, the indoor and outdoor buffer areas and the outdoor area are judged by pattern recognition. By training the BP neural network, the correct selection probability of the three positioning states in the positioning process is improved. However, the classic BP neural network has many iterations and the network structure is difficult to select. These problems will cause the BP neural network to have a large error. Therefore, the present embodiment is also improved on the basis of the original BP neural network, and the method of variable learning rate and additional momentum is introduced.

[0100] The BP neural network also includes a variable learning rate and additional momentum for model optimization, and the variable learning rate l satisfies:

[0101]

[0102] The weight of the additional momentum satisfies:

[0103] Δwij(t)=η×Δwij(t-1)+l×Errj×Oi,

[0104] Among them, η is a constant between [0,1], l is the learning rate, Errj is the error of the next layer node, Oi is the output of the previous layer node, and t represents the number of iterations.

[0105] S5. Positioning the buffer zone according to the positioning fusion model.

[0106] S5 smoothly switches between Bluetooth and Beidou positioning systems based on the output results of the BP neural network model. It uses Bluetooth and Beidou positioning systems for positioning in indoor and outdoor areas respectively, and integrates the positioning data of the two positioning systems in the indoor and outdoor buffer areas to obtain more accurate positioning results.

[0107] The positioning fusion strategy based on weighted fusion means that after the Beidou and Bluetooth positioning systems output positioning results for the same location, the positioning results of the two are weighted to obtain the final positioning result. Due to the influence of the environment, Beidou positioning cannot immediately obtain high-precision positioning results in the indoor and outdoor buffer areas. It is necessary to judge whether the positioning results are reliable based on the number of observed satellites. The positioning fusion strategy introduces parameters such as the number of Beidou satellite observations and the signal quality of Bluetooth to determine the allocation factor. The specific allocation factor values ​​are shown in Table 1 below:

[0108] Table 1

[0109] Number of Beidou observation satellites (x) <![CDATA[n1 (Beidou)]]> <![CDATA[n2 (Bluetooth)]]> x≥4 1 0 1≤x≤3 (0.1,0.5) (0.5,0.9) x＝0 0 1

[0110] The position calculation formula based on weighted fusion is as follows:

[0111]

[0112] Among them, n1 and n2 represent the information allocation factor values ​​of the Beidou and Bluetooth positioning systems, respectively, PBD and PLY represent the longitude positioning results received by the Beidou and Bluetooth positioning systems, respectively, QBD and QLY represent the longitude positioning results received by the Beidou and Bluetooth positioning systems, respectively. X represents the positioning result after fusion, and P and Q represent the longitude and latitude results after fusion, respectively.

[0113] Embodiment 2

[0114] Embodiment 2 discloses a device corresponding to the indoor and outdoor collaborative positioning method of the above embodiment, which is a virtual device structure of the above embodiment. Please refer to Figure 4 As shown, including:

[0115] An acquisition module 210 is used to acquire positioning data of Bluetooth positioning and Beidou positioning;

[0116] The data processing module 220 is used to perform unified processing on the positioning data so that different positioning data follow the same data standard; perform feature extraction on the positioning data after the unified processing to obtain feature data;

[0117] A model training module 230 is used to train a BP neural network according to the feature data to obtain a positioning fusion model;

[0118] The positioning module 240 is used to locate the buffer zone according to the positioning fusion model.

[0119] Preferably, the positioning data includes indoor and outdoor location coordinates and signal strength.

[0120] Preferably, the positioning data is processed in a unified manner, including:

[0121] Perform outlier detection based on the 3α principle and remove abnormal data;

[0122] The filtered normal positioning data is normalized, and the positioning data is scaled or expanded to a preset range.

[0123] Preferably, feature extraction is performed on the unified positioning data, including:

[0124] Extracting Bluetooth signal strength value and Beidou signal strength value from the positioning data;

[0125] The Beidou positioning prior error is calculated using the least squares method.

[0126] Preferably, the Beidou positioning a priori error includes satellite clock error, receiver clock error, and atmospheric delay; calculating the Beidou positioning a priori error by least square method to estimate the position of the receiver includes:

[0127] Model the error and calculate the error e for the data points i , satisfying e i =y i -f(x i ), where (x i ,y i ) is the data point, f(x) is the fitting function;

[0128] Adjust the model parameters to meet the following requirements: Among them, a is a parameter;

[0129] According to the adjusted model parameters, the error data is fitted through the fitting function.

[0130] Preferably, the input vector of the BP neural network includes feature data of Beidou and Bluetooth positioning, and the output vector includes label values ​​of the Bluetooth positioning system and the Beidou positioning system; the number of neurons in the input layer of the BP neural network is 3, the number of neurons in the output layer is 3, and the number of neurons x in the hidden layer satisfies:

[0131] Among them, m is the number of neurons in the input layer, and n is the number of neurons in the output layer.

[0132] Preferably, the BP neural network further includes a variable learning rate and additional momentum for model optimization, and the variable learning rate l satisfies:

[0133]

[0134] The weight of the additional momentum satisfies:

[0135] Δwij(t)=η×Δwij(t-1)+lo×Errj×Oi,

[0136] Among them, η is a constant between [0,1], l is the learning rate, Errj is the error of the next layer node, Oi is the output of the previous layer node, and t represents the number of iterations.

[0137] Embodiment 3

[0138] Figure 5 A schematic diagram of the structure of an electronic device provided in Embodiment 3 of the present invention is shown in FIG. Figure 5 As shown, the electronic device includes a processor 310, a memory 320, an input device 330 and an output device 340; the number of processors 310 in the computer device can be one or more. Figure 5 A processor 310 is taken as an example; the processor 310, the memory 320, the input device 330 and the output device 340 in the electronic device can be connected via a bus or other means. Figure 5 The example of connecting through bus is taken in the following.

[0139] The memory 320, as a computer-readable storage medium, can be used to store software programs, computer executable programs and modules, such as program instructions / modules corresponding to the indoor and outdoor collaborative positioning method in the embodiment of the present invention. The processor 310 executes various functional applications and data processing of the electronic device by running the software programs, instructions and modules stored in the memory 320, that is, the indoor and outdoor collaborative positioning method of the above-mentioned embodiment 1 is realized.

[0140] The memory 320 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system and at least one application required for a function; the data storage area may store data created according to the use of the terminal, etc. In addition, the memory 320 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some instances, the memory 320 may further include a memory remotely arranged relative to the processor 310, and these remote memories may be connected to the electronic device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0141] The input device 330 may be used to receive input user identity information, location data, etc. The output device 340 may include a display device such as a display screen.

[0142] Embodiment 4

[0143] Embodiment 4 of the present invention further provides a storage medium containing computer executable instructions, and the storage medium can be used for a computer to execute an indoor and outdoor collaborative positioning method, the method comprising:

[0144] Obtain positioning data of Bluetooth positioning and Beidou positioning;

[0145] Performing unified processing on the positioning data so that different positioning data follow the same data standard;

[0146] Extracting features from the unified positioning data to obtain feature data;

[0147] According to the feature data, a positioning fusion model is obtained by training a BP neural network;

[0148] The buffer zone is positioned according to the positioning fusion model.

[0149] Of course, the storage medium containing computer executable instructions provided by an embodiment of the present invention is not limited to the method operations described above, and can also execute related operations in the indoor and outdoor collaborative positioning method provided by any embodiment of the present invention.

[0150] Through the above description of the implementation methods, the technicians in the relevant field can clearly understand that the present invention can be implemented by means of software and necessary general hardware, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions for an electronic device (which can be a mobile phone, a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0151] It is worth noting that in the above-mentioned embodiment of the device based on the indoor and outdoor collaborative positioning method, the various units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific name of each functional unit is only for the convenience of distinguishing each other, and is not used to limit the scope of protection of the present invention.

[0152] For those skilled in the art, various other corresponding changes and deformations can be made according to the technical solutions and concepts described above, and all of these changes and deformations should fall within the protection scope of the claims of the present invention.

Claims

1. An indoor and outdoor collaborative positioning method, characterized in that: The following steps are involved: Obtain positioning data of Bluetooth positioning and Beidou positioning; Performing unified processing on the positioning data so that different positioning data follow the same data standard; Performing feature extraction on the unified positioning data to obtain feature data; According to the feature data, a positioning fusion model is obtained by training a BP neural network; The buffer zone is positioned according to the positioning fusion model.

2. The indoor and outdoor collaborative positioning method according to claim 1, characterized in that: The positioning data includes indoor and outdoor location coordinates and signal strength.

3. The indoor and outdoor collaborative positioning method according to claim 1, characterized in that: The positioning data is processed in a unified manner, including: Perform outlier detection based on the 3α principle and remove abnormal data; The filtered normal positioning data is normalized, and the positioning data is scaled or expanded to a preset range.

4. The indoor and outdoor collaborative positioning method according to claim 1, characterized in that: Performing feature extraction on the unified positioning data includes: Extracting Bluetooth signal strength value and Beidou signal strength value from the positioning data; The Beidou positioning prior error is calculated using the least squares method.

5. The indoor and outdoor collaborative positioning method according to claim 4, characterized in that: The Beidou positioning priori error includes satellite clock error, receiver clock error, and atmospheric delay; the Beidou positioning priori error is calculated by the least square method to estimate the position of the receiver, including: Model the error and calculate the error e for the data points i , satisfying e i =y i -f(x i ), where (x i ,y i ) is the data point, f(x) is the fitting function; Adjust the model parameters to meet the following requirements: Among them, a is a parameter; According to the adjusted model parameters, the error data is fitted through the fitting function.

6. The indoor and outdoor collaborative positioning method according to claim 1, characterized in that: The input vector of the BP neural network includes the feature data of Beidou and Bluetooth positioning, and the output vector includes the label values ​​of the Bluetooth positioning system and the Beidou positioning system; the number of neurons in the input layer of the BP neural network is 3, the number of neurons in the output layer is 3, and the number of neurons x in the hidden layer satisfies: Among them, m is the number of neurons in the input layer, and n is the number of neurons in the output layer.

7. The indoor and outdoor collaborative positioning method according to claim 6, characterized in that: The BP neural network also includes a variable learning rate and additional momentum for model optimization, and the variable learning rate l satisfies: The weight of the additional momentum satisfies: Δwij(t)=η×Δwij(t-1)+l×Errj×Oi, Among them, η is a constant between [0,1], l is the learning rate, Errj is the error of the next layer node, Oi is the output of the previous layer node, and t represents the number of iterations.

8. An indoor and outdoor collaborative positioning device, characterized in that: It includes: Acquisition module, used to obtain positioning data of Bluetooth positioning and Beidou positioning; A data processing module is used to perform unified processing on the positioning data so that different positioning data follow the same data standard; and perform feature extraction on the positioning data after the unified processing to obtain feature data; A model training module is used to train a BP neural network according to the feature data to obtain a positioning fusion model; A positioning module is used to locate the buffer zone according to the positioning fusion model.

9. An electronic device comprising a processor, a storage medium and a computer program, wherein the computer program is stored in the storage medium, wherein: When the computer program is executed by a processor, the indoor and outdoor collaborative positioning method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the indoor and outdoor collaborative positioning method according to any one of claims 1 to 7 is implemented.

Images