Three-dimensional GPS navigation system

Abstract

The present application discloses a three-dimensional GPS navigation system. A three-dimensional GPS navigation system includes: a map information collection device that obtains a GPS satellite map of a target area; an unmanned aerial vehicle (UAV) flight device that flies to a high altitude above the target area and connects to a base station signal or a GPS satellite signal to locate the UAV's position at high altitude; a high-altitude image acquisition module that obtains high-altitude image data of the target area and establishes a three-dimensional scene model based on the high-altitude image data; and a navigation information generation module that determines the position of the UAV in the three-dimensional scene model based on the UAV's position and generates a target route based on the three-dimensional scene model. The navigation solution provided by the present application, when used in the wild, can effectively reduce the impact of natural plant growth on navigation, and accurately find the satellite route through the constructed three-dimensional scene model.

Description

Technical Field

[0001] The present application relates to the field of neural network processing technology, and in particular to a three-dimensional GPS navigation system. Background Art

[0002] The application of satellite positioning technology faces two technical constraints when conducting navigation operations in complex mountainous terrain: Insufficient signal stability: The effectiveness of satellite positioning technology is highly dependent on continuous and stable signal reception. However, in mountainous terrain, features such as canyons and steep slopes can easily create signal obstructions, leading to satellite signal link interruptions or signal-to-noise ratios (SNRs) falling below the receiver capture threshold (typically ≤25dB-Hz). Lack of geographic information: Traditional two-dimensional contour maps lack three-dimensional terrain resolution. During periods of stagnant GNSS positioning data updates, dynamic path correction cannot be achieved by leveraging the spatial topological relationship between terrain elevation features and the trajectory, resulting in cumulative navigation errors.

[0003] The above technical defects lead to the following problems in actual operations: when there is no continuous positioning signal support, the two-dimensional map cannot generate safe path guidance that conforms to the undulating terrain; due to the multipath effect (Multipath Effect) and fluctuations in the number of visible satellites (Visible Satellites <4), the horizontal positioning error can reach 10-30 meters, which makes it difficult to meet the navigation accuracy requirements of narrow ridges or cliff terrain; the positioning blind spots caused by signal shielding force operators to rely on experience to judge the path, significantly increasing the risk of travel. Summary of the Invention

[0004] The content of this application is used to briefly introduce concepts that will be described in detail in the detailed description section below. The content of this application is not intended to identify key features or essential features of the technical solution for which protection is sought, nor is it intended to limit the scope of the technical solution for which protection is sought.

[0005] As a first aspect of the present application, in order to solve the technical problems mentioned in the above background technology section, some embodiments of the present application provide a three-dimensional GPS navigation system, including:

[0006] Map information collection device, obtaining GPS satellite map of the target area;

[0007] The drone flying device flies to a high altitude in the target area and connects with the signal from the base station or GPS satellite signal to locate the position of the drone at high altitude;

[0008] High-altitude image acquisition module, which acquires high-altitude image data of the target area and builds a three-dimensional scene model based on the high-altitude image data;

[0009] The navigation information generation module determines the position of the UAV in the three-dimensional scene model based on the position of the UAV and generates a target route according to the three-dimensional scene model.

[0010] The technical solution provided by this application utilizes a drone to acquire high-altitude imagery, taking into account the difficulty of navigation in mountainous areas. This solution also utilizes the fact that there are fewer mountains in the sky, making it easier to acquire signals. The drone's location is then determined based on the drone's location, thereby generating an optimal path within a three-dimensional scene model. When used in the wild, the navigation solution provided by this application can effectively reduce the impact of natural plant growth on navigation, allowing accurate satellite routing through the constructed three-dimensional scene model.

[0011] Furthermore, the GPS satellite map of the target area includes average altitude information of each area.

[0012] In the technical solution provided in the present application, the average altitude information of the target area is obtained, so that a basic three-dimensional model of the target area can be determined, and the rough outline of the target area can be constructed using the average altitude information.

[0013] When positioning in the wild, the signal is blocked by mountains and a large number of shrubs and trees will have a certain shielding effect on the signal. Therefore, even if you fly into the sky, it is difficult to accurately locate. To this end, this application provides the following technical solutions:

[0014] The drone flight device includes:

[0015] The drone body is used to fly high into the target area;

[0016] The base station information communication module is used to establish a signal connection with surrounding signal base stations and obtain the location information of surrounding base stations;

[0017] The distance calculation module determines the distance between the drone and surrounding base stations based on the communication loss with the surrounding base stations;

[0018] The positioning module determines the location of the drone based on the distance between the drone and surrounding base stations.

[0019] In the technical solution provided by this application, the characteristic of the drone being able to search for weak signals emitted by surrounding base stations at high altitudes is utilized to obtain the actual identity of the base station, and the location of the surrounding base stations is determined based on the identity of the base station. Then, based on the signal loss, the distance between the drone and the surrounding base stations is determined, so that the location of the drone body can be accurately determined. After determining the location of the drone body, the current location can be marked on the GPS satellite map to complete the direction guidance work.

[0020] When determining the current location based on the distance from the base station, sufficient positioning accuracy is required. To this end, this application provides the following technical solutions:

[0021] Furthermore, the base station information communication module obtains the location information of the base stations whose 3 direction vector differences exceed a preset threshold.

[0022] In this solution, the base stations whose three directional vector differences exceed the preset threshold are selected. The directions of these base stations are different from the direction of the drone, so the vector angle formed by the connection between the drone and each base station is larger, and the positioning calculation is more accurate.

[0023] Furthermore, the calculation formula for the distance between the drone body and the surrounding base stations is as follows:

[0024] ;

[0025] Among them, L0 is the reference distance, n is the environmental attenuation factor, d0 is the path loss, X σ is the shadow fading margin, d represents the actual distance from the UAV to the base station, d0 represents the reference distance, and L represents the path loss.

[0026] The technical solution provided in this application predicts the degree of signal interference received by a drone based on the environmental attenuation factor. In practice, because drones are located at high altitudes, the propagation medium between them and the signal base stations is mostly air, so the attenuation is stable, allowing the actual distance to surrounding base stations to be accurately calculated.

[0027] The environment in each region is different. When encountering complex outdoor situations, the signal attenuation may be inconsistent. To address this problem, this application provides the following technical solutions:

[0028] Furthermore, a long-short memory model is pre-trained, wherein the input data of the long-short memory model is the effective mountain length between the signal base station and the target line, and the output data is the shadow fading margin and the environmental attenuation factor.

[0029] In the technical solution provided in this application, a long-short memory model is used to find the relationship between the shadow fading margin and the environmental attenuation factor and the influence of the mountain on the signal, so that the shadow fading margin and the environmental attenuation factor can be accurately calculated, and then the subsequent distance calculation between the drone body and the surrounding base stations can be more accurate.

[0030] Furthermore, the long short-term memory model includes: an input layer for inputting effective mountain length;

[0031] LSTM units selectively retain or forget information through a gating mechanism;

[0032] The output layer generates shadow fading margin and environmental attenuation factor based on the information output by the LSTM unit.

[0033] Furthermore, the LSTM unit includes:

[0034] Input gate, used to control the inflow of new information:

[0035] I t =σ(W xi x t +W hi h t-1 +b i );

[0036] Among them, I t表示 The output of the input gate, t is the current time step, W xi Represents the weight matrix input to the input gate, x t represents the input vector, W hi represents the weight matrix from the hidden state to the input gate, b i represents the input gate bias term; σ represents the Sigmoid function, h t-1 Indicates the hidden state at the previous moment

[0037] The forget gate is used to determine the retention ratio of old information;

[0038] f t =σ(W xf x t +W hf h t-1 +b f );

[0039] W xf Represents the weight matrix input to the forget gate, W hf The weight matrix from the hidden state to the forget gate, b f represents the forget gate bias term, f t Represents the output of the forget gate;

[0040] Candidate cell states:

[0041] C t =tanh(W xc x t +W hc h t-1 +b c );W xc Represents the weight matrix input to the candidate state, W hc Represents the weight matrix from hidden state to candidate state, b c represents the candidate state bias term, C t Indicates candidate status;

[0042] The cell state update unit is used to integrate the input gate, forget gate, and candidate cell states to generate output information;

[0043] D t =f t ⊙D t-1 +I t ⊙C t , D t表示 Output information of the LSTM unit, ⊙ represents the element-wise product.

[0044] In the technical solution provided in this application, by quantifying the reference distance into a time label, the information development law between the shadow fading margin and the environmental factor under different distance information is found, and then the shadow fading margin and the environmental factor are accurately calculated when the effective mountain length is known. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The drawings constituting a part of this application are used to provide a further understanding of this application and make other features, purposes and advantages of this application more apparent. The drawings and descriptions of the exemplary embodiments of this application are used to explain this application and do not constitute an improper limitation on this application.

[0046] In addition, throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and that the elements and components are not necessarily drawn to scale.

[0047] In the attached figure:

[0048] Figure 1 It is a structural diagram of the three-dimensional space GPS navigation system. DETAILED DESCRIPTION

[0049] The following describes embodiments of the present application in more detail with reference to the accompanying drawings. Although certain embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be construed as being limited to the embodiments described herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the present application. It should be understood that the drawings and embodiments of the present application are for illustrative purposes only and are not intended to limit the scope of protection of the present application.

[0050] It should also be noted that, for ease of description, only the parts related to the invention are shown in the drawings. In the absence of conflict, the embodiments and features in the embodiments of this application can be combined with each other.

[0051] The present application will be described in detail below with reference to the accompanying drawings and in combination with embodiments.

[0052] Reference Figure 1The three-dimensional GPS navigation system includes: a map information collection device, a UAV flight device, a high-altitude image acquisition module, and a navigation information generation module. The map information collection device obtains a GPS satellite map of the target area; the UAV flight device flies to the high altitude of the target area and connects with the signal of the signal base station or the GPS satellite signal to locate the position of the UAV at high altitude;

[0053] The high-altitude image acquisition module obtains high-altitude image data of the target area and establishes a three-dimensional scene model based on the high-altitude image data; the navigation information generation module determines the position of the drone in the three-dimensional scene model based on the position of the drone and generates the target route according to the three-dimensional scene model.

[0054] The GPS satellite map of the target area includes the average elevation of each area. Specifically, a map information collection device obtained a 1:10,000 scale GPS satellite map of the target area from the National Basic Geographic Information Center. The data had an accuracy better than 5 meters and a grid resolution of 10 x 10 meters. ArcGIS Pro software was used to preprocess the raw DEM data, including terrain smoothing (using a 3 x 3 median filter kernel), hillshade generation (at a solar altitude of 45°), and terrain feature extraction (ridgelines, cliff boundaries, and slope classification).

[0055] The target area is the specific area required for navigation. For example, if navigation is required in a mountainous area, map information of the mountainous area is pre-entered. This allows the elevation of each location in the mountainous area and the corresponding 3D scene information to be marked on the digital map.

[0056] When navigating in mountainous areas, a map alone is not enough; you need to know your actual location. However, satellite and base station signals are weak in mountainous areas, making precise positioning impossible. Therefore, this solution utilizes a high-altitude drone. After reaching a preset altitude, the drone communicates with satellites or surrounding base stations, using these signals to determine its current location.

[0057] Specifically, the drone flight device includes: a drone body, a base station information and communication module, a distance calculation module, and a positioning module. The drone body is used to fly high into the target area; the base station information and communication module is used to establish a signal connection with surrounding signal base stations and obtain their location information; the distance calculation module determines the distance between the drone body and surrounding base stations based on communication loss with the surrounding base stations; and the positioning module determines the drone body's location based on the distance between the drone body and surrounding base stations.

[0058] Furthermore, the base station information communication module obtains the location information of the base stations whose 3 direction vector differences exceed a preset threshold.

[0059] Specifically: After the drone reaches an altitude of 200-300 meters, it initiates a 360° omnidirectional signal scan, screening valid base stations based on an RSSI strength threshold (≥-90dBm). It then selects three base stations with the best spatial distribution (with an azimuth angle ≥ 120°) and obtains their precise coordinates from the LTE base station database. The distance calculation module uses a dual-frequency signal fusion algorithm.

[0060] The calculation formula for the distance between the drone and the surrounding base stations is as follows:

[0061] ;

[0062] Among them, L0 is the reference distance, n is the environmental attenuation factor, d0 is the path loss, X σ is the shadow fading margin, d represents the actual distance from the UAV to the base station, d0 represents the reference distance, and L represents the path loss.

[0063] In this solution, the distance between the drone and the surrounding base stations is calculated based on the degree of signal attenuation. After obtaining the distance between the drone and the surrounding base stations, the horizontal distance between the drone and each base station can be calculated based on the altitude of the drone's current location, thereby obtaining the actual location of the drone in the map information.

[0064] When calculating the actual distance between a drone and each signal base station, accurate environmental attenuation factors and shadow fading margins are required. To this end, this solution pre-trains a long-short-term memory model. The input data for this model is the effective mountain length between the signal base station and the target, and the output data is the shadow fading margin and environmental attenuation factor.

[0065] Specifically, sufficient sample data is obtained in advance. The sample data includes input data and labels. The labels are shadow fading margin and environmental attenuation factor, and the input data is effective mountain length.

[0066] Over time, the shadow fading margin and environmental attenuation factor have a clear distance relationship with the length of the mountain the signal penetrates. Therefore, this solution uses a long-short memory model to handle the relationship between the two, accurately finding the changing relationship between the shadow fading margin and environmental attenuation factor as the mountain length changes.

[0067] Furthermore, the effective mountain length is the length of the line connecting the drone body and the base station that passes through the mountain. In practice, because the specific location of the drone is unclear, the possible position and direction of the drone can be calculated using the initially set shadow fading margin and environmental attenuation factor. Then, after the drone body flies to a high altitude and stabilizes, the barometer is used to read the current altitude. In this way, the position between the base station and the drone can be roughly determined. Because the drone body is located at a high altitude in the mountainous area, the signal link between the drone body and the base station only needs to pass through the mountain near the base station. When the signal transmission direction is obtained, the distance of the mountain that the signal needs to cross can be calculated based on the elevation information in the GPS satellite map, and this distance is used as the "effective mountain length."

[0068] The long-short memory model includes: an input layer for inputting the effective mountain length;

[0069] LSTM units selectively retain or forget information through a gating mechanism;

[0070] The output layer generates shadow fading margin and environmental attenuation factor based on the information output by the LSTM unit.

[0071] Furthermore, the LSTM unit includes:

[0072] Input gate, used to control the inflow of new information:

[0073] I t =σ(W xi x t +W hi h t-1 +b i );

[0074] Among them, I t表示 The output of the input gate, t is the current time step, W xi Represents the weight matrix input to the input gate, x t represents the input vector, W hi represents the weight matrix from the hidden state to the input gate, b i represents the input gate bias term; σ represents the Sigmoid function, h t-1 Indicates the hidden state at the previous moment;

[0075] The forget gate is used to determine the retention ratio of old information;

[0076] f t =σ(W xf x t +W hf h t-1 +b f );

[0077] Wxf Represents the weight matrix input to the forget gate, W hf The weight matrix from the hidden state to the forget gate, b f represents the forget gate bias term, f t Represents the output of the forget gate;

[0078] Candidate cell states:

[0079] C t =tanh(W xc x t +W hc h t-1 +b c );W xc Represents the weight matrix input to the candidate state, W hc Represents the weight matrix from hidden state to candidate state, b c represents the candidate state bias term, C t Indicates candidate status;

[0080] The cell state update unit is used to integrate the input gate, forget gate, and candidate cell states to generate output information;

[0081] D t =f t ⊙D t-1 +I t ⊙C t , D t表示 Output information of the LSTM unit, ⊙ represents the element-wise product.

[0082] The hidden state output gate is connected to the cell state update unit signal:

[0083] ;

[0084] ;

[0085] Among them, W xo is the weight matrix of the output gate corresponding to the input vector, W ho is the weight matrix of the hidden state corresponding to the output gate, b o is the bias vector of the output gate.

[0086] The above is the general structure of the LSTM unit. In actual use, the long short-term memory model needs to be trained. The specific method is as follows:

[0087] Step 1: Pre-configure a large number of training samples. The input data in the training samples is the effective mountain length, the label is "environmental attenuation factor and shadow fading margin", and the number of samples is 1000.

[0088] Step 2: Divide the samples into training set and validation set, input the training set into the long short-term memory model, make the training set into the input sequence {x1, x2, ...}, and make the label into the output sequence {y1, y2, ...}.

[0089] Step 3: Forward propagation:

[0090] For each time step t, calculate h t and D t , the final output h t Mapped to predicted value y`;

[0091] Step 4: Use mean square error as the loss function and train the long short-term memory model through time back propagation.

[0092] In this way, the length and model can be used to accurately calculate the environmental attenuation factor and shadow fading margin.

[0093] After calculating the distance between the drone and several surrounding base stations, the drone needs to be positioned. The specific method includes the following steps:

[0094] S1: Obtain three base stations that can transmit signals to the drone body and obtain the latitude and longitude of each base station;

[0095] S2: Select a base station as the origin of the plane rectangular coordinate system, and convert the longitude and latitude of each base station into coordinates of the plane rectangular coordinate system.

[0096] S3: Calculate the distance weight information α of each base station to the drone body k , where k is the index of the base station;

[0097] ;

[0098] in, is the distance weight information of the kth base station, Indicates the minimum effective RSSI, Maximum effective RSSI, represents the received signal strength of the kth base station, represents the geometric dilution of precision of the ith emergency department, Indicates the altitude of the drone (calculated by the barometer). represents the altitude of the kth base station, and dk represents the distance from the kth base station to the drone body;

[0099] S4: Calculate the coordinates of the drone body in the plane rectangular coordinate system;

[0100] ; ;

[0101] Among them, Ep is the horizontal coordinate of the drone body in the plane rectangular coordinate system, and Np is the vertical coordinate of the drone body in the plane rectangular coordinate system. is the horizontal coordinate of the kth base station in the plane rectangular coordinate system, is the ordinate of the kth base station in the rectangular coordinate system.

[0102] Because the drone takes off vertically, it is possible to obtain the user's horizontal and vertical coordinates in the plane rectangular coordinate system, and then convert the horizontal and vertical coordinates into longitude and latitude to obtain the user's exact location.

[0103] After obtaining the user's accurate location, the current location can be marked on the GPS satellite map of the target area.

[0104] Because mountainous areas are prone to vegetation, roads can change, necessitating the acquisition of as much terrain as possible. This solution utilizes a high-altitude image acquisition module to acquire high-altitude image data of the target area and build a 3D scene model based on this data. The navigation information generation module determines the drone's position within the 3D scene model based on its location and generates the target route based on this model.

[0105] Specifically, high-altitude image data is acquired directly by taking photos from a drone. After acquiring the high-altitude image data, it is necessary to convert it into 3D point cloud data. Acquiring 3D information from high-altitude image data is a well-known technique, and the following is a possible implementation method.

[0106] The specific steps include:

[0107] St1: The drone collects high-altitude image information in a zigzag pattern above the user.

[0108] St2: Use SIFT algorithm to detect key points in high-altitude image information.

[0109] St3: Quickly match different images in the high-altitude image information based on key points to obtain matching feature points, and generate a matching matrix based on the feature matching points.

[0110] The matching matrix is actually the position information that can be matched with each other in pictures of different perspectives.

[0111] St4: Linear triangulation of matching feature points is performed to generate a sparse point cloud.

[0112] St5: Use multi-view stereo matching to generate dense point clouds, and then generate 3D model information based on the dense point clouds. The above reconstruction of 3D point cloud information is an existing technology and no further parameters are given here.

[0113] Satellite maps generally only show the actual altitude of the nearby ground, so it is actually impossible to accurately find the route using satellite maps. In this solution, three-dimensional model information will be generated. The three-dimensional model information describes the surrounding vegetation coverage, which can better enable navigation.

[0114] The navigation information generation module determines the drone's position within the 3D scene model based on its location and generates a target route based on the 3D scene model. During use, the navigation information generation module will try to select a route with minimal elevation differences for user reference. In practice, users can easily choose an appropriate route based on their current location and satellite maps.

[0115] The above descriptions are merely some preferred embodiments of the present application and illustrate the technical principles employed. Those skilled in the art should understand that the scope of the invention described in the embodiments of the present application is not limited to technical solutions formed by specific combinations of the aforementioned technical features, but also encompasses other technical solutions formed by any combination of the aforementioned technical features or their equivalents without departing from the aforementioned inventive concept. For example, a technical solution formed by replacing the aforementioned features with (but not limited to) technical features with similar functions disclosed in the embodiments of the present application.

Claims

1. A three-dimensional GPS navigation system, characterized in that: include: Map information collection device, obtaining GPS satellite map of the target area; The drone flying device flies to a high altitude in the target area and connects with the signal from the base station or GPS satellite signal to locate the position of the drone at high altitude; High-altitude image acquisition module, which acquires high-altitude image data of the target area and builds a three-dimensional scene model based on the high-altitude image data; A navigation information generation module determines the position of the UAV in the three-dimensional scene model based on the UAV's position and generates a target route based on the three-dimensional scene model; The drone flight device includes: The drone body is used to fly high into the target area; The base station information communication module is used to establish a signal connection with surrounding signal base stations and obtain the location information of surrounding base stations; The distance calculation module determines the distance between the drone and surrounding base stations based on the communication loss with the surrounding base stations; The positioning module determines the location of the drone based on the distance between the drone and surrounding base stations; The calculation formula for the distance between the drone and the surrounding base stations is as follows: Among them, L0 is the reference distance, n is the environmental attenuation factor, d0 is the path loss, X σ is the shadow fading margin, d represents the actual distance from the UAV to the base station, and d0 represents the reference distance; Pre-training a long-short memory model, wherein the input data of the long-short memory model is the effective mountain length between the signal base station and the target line, and the output data is the shadow fading margin and the environmental attenuation factor; The effective mountain length is the length of the line connecting the drone body and the base station passing through the mountain.

2. The three-dimensional GPS navigation system according to claim 1, characterized in that: The GPS satellite map of the target area includes average altitude information of each area.

3. The three-dimensional GPS navigation system according to claim 1, characterized in that: The base station information communication module obtains the location information of the base stations whose directional vector differences exceed a preset threshold.

4. The three-dimensional GPS navigation system according to claim 1, characterized in that: Long-short memory models include: Input layer, used to input effective mountain length; LSTM units selectively retain or forget information through a gating mechanism; The output layer generates shadow fading margin and environmental attenuation factor based on the information output by the LSTM unit.

5. The three-dimensional GPS navigation system according to claim 4, characterized in that: The LSTM unit consists of: Input gate, used to control the inflow of new information: I t =σ(W xi x t +W hi h t-1 +b i ); Among them, I t represents the output of the input gate, t is the current time step, W xi Represents the weight matrix input to the input gate, x t represents the input vector, W hi represents the weight matrix from the hidden state to the input gate, b i represents the input gate bias term; σ represents the Sigmoid function, h t-1 Indicates the hidden state at the previous moment; The forget gate is used to determine the retention ratio of old information; f t =σ(W xf x t +W hf h t-1 +b f ); W xf Represents the weight matrix input to the forget gate, W hf The weight matrix from the hidden state to the forget gate, b f represents the forget gate bias term, f t Represents the output of the forget gate; Candidate cell states: C t =tanh(W xc x t +W hc h t-1 +b c );W xc Represents the weight matrix input to the candidate state, W hc Represents the weight matrix from hidden state to candidate state, b c represents the candidate state bias term, C t Indicates candidate status; The cell state update unit is used to integrate the input gate, forget gate, and candidate cell states to generate output information; D t =f t ⊙D t-1 +I t ⊙C t , D t Represents the output information of the LSTM unit, ⊙ represents the element product; The hidden state output gate is connected to the cell state update unit signal: Among them, W xo is the weight matrix of the output gate corresponding to the input vector, W ho is the weight matrix of the hidden state corresponding to the output gate, b o is the bias vector of the output gate.

6. The three-dimensional GPS navigation system according to claim 1, characterized in that: The positioning method of the drone body includes the following steps: S1: Obtain three base stations that can transmit signals to the drone body and obtain the latitude and longitude of each base station; S2: Select a base station as the origin of the plane rectangular coordinate system and convert the longitude and latitude of each base station into the coordinates of the plane rectangular coordinate system; S3: Calculate the distance weight information of each base station to the drone body , where k is the index of the base station; in, The distance weight information of the kth base station, Indicates the minimum effective RSSI, Maximum effective RSSI, represents the received signal strength of the kth base station, represents the geometric dilution of precision of the kth base station, Indicates the altitude of the drone, h c k represents the altitude of the kth base station, d k represents the distance between the kth base station and the drone body; S4: Calculate the coordinates of the drone body in the plane rectangular coordinate system; Among them, Ep is the horizontal coordinate of the drone body in the plane rectangular coordinate system, and Np is the vertical coordinate of the drone body in the plane rectangular coordinate system. is the horizontal coordinate of the kth base station in the plane rectangular coordinate system, is the ordinate of the kth base station in the plane rectangular coordinate system.

Images