Unmanned aerial vehicle and method based on satellite internet communication

Abstract

The invention discloses an unmanned aerial vehicle and method based on satellite internet communication, and the method comprises the steps: carrying out the correlation interaction analysis of collected signal intensity and a corresponding antenna angle through employing a data processing technology based on deep learning, so as to mine a dynamic correlation mode between the signal intensity and the antenna angle; and intelligent recommendation and optimization control of the antenna angle are carried out based on the above, so that automatic control of the antenna angle can be realized, stable connection with a satellite is maintained, and the communication performance and task execution efficiency of the unmanned aerial vehicle in a complex environment are improved.

Description

Technical Field

[0001] The present application relates to the field of intelligent control, and more specifically, to a UAV and method based on satellite Internet communication. Background Art

[0002] With the rapid development of drone technology, drones are increasingly used in military reconnaissance, environmental monitoring, logistics distribution, agricultural plant protection and other fields. However, the communication capability of drones has always been one of the important factors limiting their performance and application scenarios. Especially in long-distance or complex terrain conditions, the communication stability between drones and ground control stations directly affects the completion and safety of the mission.

[0003] Traditional drone communications mainly rely on ground base stations, but in long-distance or complex terrain environments, the coverage of ground base stations is limited, and signals are easily interfered or attenuated, resulting in unstable communications. In recent years, the rise of satellite Internet technology has provided new possibilities for solving this problem. Satellite Internet can provide global communication coverage, is not restricted by geographical conditions, and is particularly suitable for application scenarios over long distances and in complex terrain.

[0004] However, although satellite communications have the advantage of global coverage, they have high transmission delays and signal attenuation problems due to the long signal transmission distance. In addition, during the flight of drones, the relative position between the antenna and the satellite is constantly changing due to factors such as attitude changes and terrain influences, resulting in large fluctuations in the received signal strength. Traditional antenna angle adjustment mainly relies on preset parameters or manual control, which is usually difficult to adapt to complex flight environments in real time, affecting the stability and reliability of communications.

[0005] Therefore, an optimized UAV based on satellite Internet communication is needed to solve the above technical problems. Summary of the invention

[0006] In order to solve the above technical problems, this application is proposed. The embodiment of the present application provides a UAV and method based on satellite Internet communication, which uses deep learning-based data processing technology to perform correlation and interactive analysis on the collected signal strength and the corresponding antenna angle to dig out the dynamic correlation pattern between the signal strength and the antenna angle, and based on this, intelligently recommend and optimize the antenna angle, which can realize automatic control of the antenna angle to maintain a stable connection with the satellite, thereby improving the communication performance and task execution efficiency of the UAV in complex environments.

[0007] According to one aspect of the present application, a satellite internet communication-based drone is provided, comprising:

[0008] The drone body;

[0009] A satellite communication receiving device installed on the drone body, wherein the satellite communication receiving device includes an antenna, a motor for adjusting the angle of the antenna, a power supply for powering the motor, and a controller, wherein the controller is used to control the motor to adjust the angle of the antenna;

[0010] Wherein, the controller comprises:

[0011] An antenna status monitoring module, used to obtain the signal strength values ​​of the antenna at different angles collected by the signal strength detector to obtain a sequence of antenna angle values ​​and a sequence of signal strength values;

[0012] An inter-sample correlation analysis module, used for performing inter-sample correlation coding on the sequence of antenna angle values ​​and the sequence of signal strength values ​​respectively to obtain an implicit correlation feature vector between antenna angle samples and an implicit correlation feature vector between signal strength samples;

[0013] An interaction analysis module, used for performing feature interaction analysis on the implicit correlation feature vectors between the antenna angle samples and the implicit correlation feature vectors between the signal strength samples to obtain an antenna angle-signal strength enhancement interaction feature vector;

[0014] The motor control instruction generation module is used to generate a motor control instruction based on the antenna angle-signal strength enhancement interaction feature vector.

[0015] According to another aspect of the present application, a method for controlling a drone based on satellite Internet communication is provided, which includes:

[0016] Acquire signal strength values ​​of the antenna at different angles collected by a signal strength detector to obtain a sequence of antenna angle values ​​and a sequence of signal strength values;

[0017] Performing inter-sample correlation coding on the sequence of antenna angle values ​​and the sequence of signal strength values ​​respectively to obtain an implicit correlation feature vector between antenna angle samples and an implicit correlation feature vector between signal strength samples;

[0018] Performing feature interaction analysis on the implicit correlation feature vectors between the antenna angle samples and the implicit correlation feature vectors between the signal strength samples to obtain an antenna angle-signal strength enhancement interaction feature vector;

[0019] Based on the antenna angle-signal strength enhancement interaction feature vector, a motor control command is generated.

[0020] The present application has at least the following technical effects: Compared with the prior art, the present application provides a UAV and method based on satellite Internet communication, which utilizes deep learning-based data processing technology to perform correlation and interactive analysis on the collected signal strength and the corresponding antenna angle, so as to dig out the dynamic correlation pattern between the signal strength and the antenna angle, and based on this, make intelligent recommendations and optimize the control of the antenna angle, which can realize automatic control of the antenna angle to maintain a stable connection with the satellite, thereby improving the communication performance and mission execution efficiency of the UAV in complex environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0022] Figure 1 is a block diagram of a UAV based on satellite Internet communication according to an embodiment of the present application;

[0023] Figure 2 A schematic diagram of data flow of a UAV based on satellite Internet communication according to an embodiment of the present application;

[0024] Figure 3 is a block diagram of a controller in a UAV based on satellite Internet communication according to an embodiment of the present application;

[0025] Figure 4 Flow chart of a method for controlling a UAV based on satellite Internet communication according to an embodiment of the present application. DETAILED DESCRIPTION

[0026] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0027] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0028] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0029] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. Meanwhile, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0030] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0031] Traditional UAV communication mainly relies on ground base stations, but in long-distance or complex terrain environments, the coverage of ground base stations is limited, and the signal is easily interfered or attenuated, resulting in unstable communication. In recent years, the rise of satellite Internet technology has provided new possibilities for solving this problem. Satellite Internet can provide global communication coverage, not restricted by geographical conditions, and is particularly suitable for application scenarios in long distances and complex terrain. However, although satellite communication has the advantage of global coverage, due to the long signal transmission distance, there are high transmission delays and signal attenuation problems. In addition, during the flight of the UAV, due to factors such as attitude changes and terrain influences, the relative position between the antenna and the satellite is constantly changing, resulting in large fluctuations in the received signal strength. The traditional antenna angle adjustment mainly relies on preset parameters or manual control, which is usually difficult to adapt to complex flight environments in real time, affecting the stability and reliability of communication. Therefore, an optimized UAV based on satellite Internet communication is expected.

[0032] In the technical solution of the present application, a drone based on satellite Internet communication is proposed. Figure 1 4 is a block diagram of a UAV based on satellite Internet communication according to an embodiment of the present application. Figure 2 FIG. 1 is a schematic diagram of data flow of a UAV based on satellite Internet communication according to an embodiment of the present application. Figure 1 and Figure 2As shown, according to an embodiment of the present application, a drone 300 based on satellite Internet communication includes: a drone body 310; a satellite communication receiving device 320 installed on the drone body, wherein the satellite communication receiving device includes an antenna, a motor for adjusting the angle of the antenna, and a power supply for powering the motor; and a controller 330, the controller is used to control the motor to adjust the angle of the antenna.

[0033] In particular, the drone body 310. The drone body mainly includes a mechanical structure, a power system, a control system and a communication system.

[0034] In particular, the satellite communication receiving device 320 installed on the drone body, wherein the satellite communication receiving device includes an antenna, a motor for adjusting the angle of the antenna, and a power supply for powering the motor. The antenna is a key component for receiving and sending satellite signals. It is responsible for capturing the downlink signal from the satellite and sending the uplink signal of the drone to the satellite; the motor is used to adjust the angle of the antenna to ensure that the antenna is always pointing to the satellite to maintain the best communication effect; the power supply provides the necessary electrical energy for the motor to ensure that the motor can work normally.

[0035] In particular, the controller 330 is used to control the motor to adjust the angle of the antenna. Figure 3 As shown, the controller 330 includes: an antenna state monitoring module 331, which is used to obtain the signal strength values ​​of the antenna at different angles collected by the signal strength detector to obtain a sequence of antenna angle values ​​and a sequence of signal strength values; an inter-sample correlation analysis module 332, which is used to perform inter-sample correlation encoding on the sequence of antenna angle values ​​and the sequence of signal strength values ​​respectively to obtain an implicit correlation feature vector between antenna angle samples and an implicit correlation feature vector between signal strength samples; an interaction analysis module 333, which is used to perform feature interaction analysis on the implicit correlation feature vector between antenna angle samples and the implicit correlation feature vector between signal strength samples to obtain an antenna angle-signal strength enhancement interaction feature vector; and a motor control instruction generation module 334, which is used to generate a motor control instruction based on the antenna angle-signal strength enhancement interaction feature vector.

[0036] Specifically, the antenna state monitoring module 331 is used to obtain the signal strength values ​​of the antenna at different angles collected by the signal strength detector to obtain a sequence of antenna angle values ​​and a sequence of signal strength values. It should be understood that the antenna angle is a key factor affecting the quality of signal reception. At different angles, the signal strength received by the antenna will be different. Therefore, in the technical solution of the present application, it is expected to determine the optimal angle of the antenna under specific flight conditions by analyzing the response of the signal strength when the antenna angle changes, thereby achieving the optimal communication connection between the drone and the satellite.

[0037] Specifically, the inter-sample correlation analysis module 332 is used to perform inter-sample correlation encoding on the sequence of the antenna angle values ​​and the sequence of the signal strength values ​​to obtain the implicit correlation feature vector between the antenna angle samples and the implicit correlation feature vector between the signal strength samples. In a specific example of the present application, the sequence of the antenna angle values ​​and the sequence of the signal strength values ​​are respectively input into a sequence encoder based on a forward LSTM model to obtain the implicit correlation feature vector between the antenna angle samples and the implicit correlation feature vector between the signal strength samples. It should be understood that, considering that in the flight scene of the drone, the changes in the antenna angle and the signal strength are nonlinear, and traditional linear models are usually difficult to accurately capture the dynamic change laws of the two. Therefore, in the technical solution of the present application, a forward LSTM model with strong nonlinear fitting ability in data processing is used to process the sequence of the antenna angle value and the sequence of the signal strength value respectively. Those skilled in the art should know that the LSTM model is a special recurrent neural network that is based on a gating mechanism and can effectively learn and memorize long-term dependencies in sequence data, and mine the complex correlations between parameter samples in the sequence data, thereby capturing the global correlation change characteristics in the antenna angle data and the signal strength data, and generating implicit correlation feature vectors between antenna angle samples and implicit correlation feature vectors between signal strength samples.

[0038] Specifically, the interaction analysis module 333 is used to perform feature interaction analysis on the implicit correlation feature vectors between the antenna angle samples and the implicit correlation feature vectors between the signal strength samples to obtain the antenna angle-signal strength enhancement interaction feature vector. In the technical solution of the present application, in order to learn the influence mode of signal strength being affected by the change of antenna angle, the present application further performs correlation interaction analysis on the implicit correlation feature vectors between the antenna angle samples and the implicit correlation feature vectors between the signal strength samples to mine the dynamic correlation mode between the two. In particular, in view of the difference in data dimension and variation range between signal strength and antenna angle, directly performing feature interaction fusion on the implicit correlation feature vectors between the antenna angle samples and the implicit correlation feature vectors between the signal strength samples may introduce deviations. Therefore, in order to improve the feature interaction effect between multi-source data, the present application proposes an optimized feature interaction method, which enhances the correlation and information transmission efficiency between the implicit correlation feature vectors between the antenna angle samples and the implicit correlation feature vectors between the signal strength samples by co-spatial mapping and mapping deviation compensation, thereby improving the accuracy of feature expression.

[0039] Specifically, in an embodiment of the present application, first, the implicit correlation feature vector between the antenna angle samples and the implicit correlation feature vector between the signal strength samples are co-space mapped and modulated to obtain the implicit correlation feature vector between the modulated antenna angle samples and the implicit correlation feature vector between the modulated signal strength samples. In the technical solution of the present application, the implicit correlation feature vector between the antenna angle samples and the implicit correlation feature vector between the signal strength samples are input into a co-space mapper based on a fully connected layer to obtain the implicit correlation feature vector between the antenna angle samples and the implicit correlation feature vector between the modulated signal strength samples, wherein the implicit correlation feature vector between the antenna angle samples and the implicit correlation feature vector between the modulated signal strength samples have the same characteristic scale. It should be understandable that, given that the implicit correlation feature vector between the antenna angle samples and the implicit correlation feature vector between the signal strength samples are in different feature spaces, cross-space semantic interaction cannot be achieved. This application first uses a fully connected layer to construct a common space mapper, and uses the powerful nonlinear mapping capability of the fully connected layer to perform feature modulation on the two and map them to the same feature space, ensuring that the two have the same feature scale to facilitate subsequent interactive processing.

[0040] Furthermore, the implicit correlation feature vectors between the modulated antenna angle samples and the implicit correlation feature vectors between the modulated signal strength samples are interactively fused with semantic mapping deviation compensation to obtain the antenna angle-signal strength enhancement interactive feature vector. Considering that part of the original feature information may be lost in the feature space mapping process, the intrinsic connection or covariance relationship between the antenna angle and the signal strength is destroyed, thereby increasing the difficulty of subsequent correlation pattern mining. Therefore, in the technical solution of the present application, after the co-space mapping is performed, the implicit correlation feature vectors between the antenna angle samples and the implicit correlation feature vectors between the signal strength samples after the co-space mapping modulation are further compensated for the feature mapping deviation. Specifically, first, the mapping deviation semantic compensation factors of the implicit correlation feature vector between the modulated antenna angle samples and the implicit correlation feature vector between the modulated signal strength samples relative to the implicit correlation feature vector between the antenna angle samples and the implicit correlation feature vector between the signal strength samples are calculated respectively to obtain the antenna angle mapping deviation semantic compensation factor and the signal strength mapping deviation semantic compensation factor; specifically, the original implicit correlation feature vector between the antenna angle samples and the implicit correlation feature vector between the signal strength samples are interpolated, so that the features after the interpolation processing have the same characteristic scale as the features after co-space modulation, and while ensuring the scale alignment, more important information of the original features is retained, thereby improving the model's perception of detail information. Next, based on the semantic compensation factor of the antenna angle mapping deviation and the semantic compensation factor of the signal strength mapping deviation, the implicit correlation feature vector between the modulated antenna angle samples and the implicit correlation feature vector between the modulated signal strength samples are compensated to obtain the implicit correlation feature vector between the compensated antenna angle samples and the implicit correlation feature vector between the compensated signal strength samples; specifically, the antenna angle feature and signal strength feature after interpolation processing are compared and analyzed with the antenna angle feature and signal strength feature after co-spatial modulation using a feature mapping deviation semantic measurement network to identify the deviation generated in the co-spatial mapping process, and accordingly calculate the corresponding compensation factor to quantify the degree of semantic deviation introduced in the transformation process. Then, based on the calculated compensation factor, the implicit correlation feature vector between the antenna angle samples and the implicit correlation feature vector between the signal strength samples after the co-spatial modulation are compensated and corrected to eliminate the information distortion caused by the feature space transformation and improve the sensitivity of the model to detail information. Finally, the implicit correlation feature vector between the compensated antenna angle samples and the implicit correlation feature vector between the compensated signal strength samples are input into the multi-dimensional significant interaction network between features to obtain the antenna angle-signal strength enhancement interaction feature vector. Through multi-level feature interaction fusion, the correlation pattern between signal strength and antenna angle at different levels is learned, and the correlation response law between the two is fully excavated, thereby generating the antenna angle-signal strength enhancement interaction feature vector.

[0041] The specific process of respectively calculating the mapping deviation semantic compensation factors of the implicit correlation feature vector between the modulated antenna angle samples and the implicit correlation feature vector between the modulated signal strength samples relative to the implicit correlation feature vector between the antenna angle samples and the implicit correlation feature vector between the signal strength samples includes: interpolating the implicit correlation feature vector between the antenna angle samples and the implicit correlation feature vector between the signal strength samples to obtain the implicit correlation feature vector between the antenna angle samples after interpolation and the implicit correlation feature vector between the signal strength samples after interpolation, wherein the implicit correlation feature vector between the antenna angle samples after interpolation is The implicit correlation feature vector between the interpolated signal strength samples, the implicit correlation feature vector between the modulated antenna angle samples and the implicit correlation feature vector between the modulated signal strength samples have the same characteristic scale; the implicit correlation feature vector between the interpolated antenna angle samples and the modulated antenna angle samples, the implicit correlation feature vector between the interpolated signal strength samples and the modulated signal strength samples are respectively input into the feature mapping deviation semantic measurement network to obtain the antenna angle mapping deviation semantic compensation factor and the signal strength mapping deviation semantic compensation factor. More specifically, the implicit correlation feature vector between the antenna angle samples after interpolation processing and the implicit correlation feature vector between the antenna angle samples after modulation, the implicit correlation feature vector between the signal strength samples after interpolation processing and the implicit correlation feature vector between the signal strength samples after modulation are respectively input into the feature mapping deviation semantic measurement network to obtain the antenna angle mapping deviation semantic compensation factor and the signal strength mapping deviation semantic compensation factor, including: calculating the positional dot division vector between the implicit correlation feature vector between the antenna angle samples after modulation and the implicit correlation feature vector between the antenna angle samples after interpolation processing, and calculating the base-two logarithmic value of the absolute value of each eigenvalue in the dot division vector to obtain the antenna angle mapping deviation representation vector; calculating the dot product vector between the antenna angle mapping deviation representation vector and the implicit correlation feature vector between the antenna angle samples after modulation, and calculating the exponential function value with base e and the sum of the eigenvalues ​​of the dot product vector as the exponent to obtain the antenna angle mapping deviation semantic compensation factor.

[0042] More specifically, the specific process of compensating the implicit correlation feature vector between the modulated antenna angle samples and the implicit correlation feature vector between the modulated signal strength samples based on the antenna angle mapping deviation semantic compensation factor and the signal strength mapping deviation semantic compensation factor includes: calculating the sum of the square of the antenna angle mapping deviation semantic compensation factor and itself as a correction factor, and dividing each eigenvalue of the implicit correlation feature vector between the modulated antenna angle samples by the correction factor to obtain the implicit correlation feature vector between the compensated antenna angle samples.

[0043] More specifically, the specific process of inputting the implicit correlation feature vector between the compensated antenna angle samples and the implicit correlation feature vector between the compensated signal strength samples into the multidimensional significant interaction network between features includes: respectively calculating the position point subtraction, position point multiplication, position point addition and position point division between the implicit correlation feature vector between the compensated antenna angle samples and the implicit correlation feature vector between the compensated signal strength samples to obtain the first antenna angle-signal strength interaction representation vector, the second antenna angle-signal strength interaction representation vector, the third antenna angle-signal strength interaction representation vector and the fourth antenna angle-signal strength interaction representation vector; cascading and fusing the first antenna angle-signal strength interaction representation vector, the second antenna angle-signal strength interaction representation vector, the third antenna angle-signal strength interaction representation vector and the fourth antenna angle-signal strength interaction representation vector, and inputting them into a one-dimensional convolutional layer including a maximum pooling layer to obtain the antenna angle-signal strength enhanced interaction feature vector.

[0044] In summary, in the above embodiment, the implicit correlation feature vector between the antenna angle samples and the implicit correlation feature vector between the signal strength samples are subjected to feature interaction analysis to obtain the antenna angle-signal strength enhancement interaction feature vector, including: the implicit correlation feature vector between the antenna angle samples and the implicit correlation feature vector between the signal strength samples are processed by the following timing interaction analysis formula to obtain the antenna angle-signal strength enhancement interaction feature vector, wherein the timing interaction analysis formula is:

[0045] v t1 =Relu(W cm v1+b cm )

[0046] v t2 =Relu(W cm v2+b cm )

[0047] v c1 =L(v1)

[0048] v c2 =L(v2)

[0049]

[0050] Wherein, v1 represents the implicit correlation feature vector between the antenna angle samples, v2 represents the implicit correlation feature vector between the signal strength samples, and W cm and b cm Represent the weight matrix and bias term respectively, Relu represents the activation function, v t1 and v t2They represent the implicit correlation feature vector between the modulated antenna angle samples and the implicit correlation feature vector between the modulated signal strength samples, L(·) represents the interpolation process, and v c1 and v c2 They represent the implicit correlation feature vector between antenna angle samples after interpolation and the implicit correlation feature vector between signal strength samples after interpolation, and denote the eigenvalues ​​of the i-th position in the implicit correlation feature vector between the modulated antenna angle samples, the implicit correlation feature vector between the interpolated antenna angle samples, the implicit correlation feature vector between the modulated signal strength samples, and the implicit correlation feature vector between the interpolated signal strength samples, respectively; log(·) denotes a logarithmic function with base 2; exp(·) denotes a natural exponential function; L denotes a characteristic scale of the implicit correlation feature vector between the modulated antenna angle samples; s1 and s2 denote semantic compensation factors for antenna angle mapping deviation and signal strength mapping deviation, respectively; v b1 and v b2 They represent the implicit correlation feature vector between compensated antenna angle samples and the implicit correlation feature vector between compensated signal strength samples, ⊙ represents the dot product, Indicates point addition, Indicates point reduction, represents point division, conv1D represents a one-dimensional convolution operation, MaxPool represents a maximum pooling operation, and v in Represents the antenna angle-signal strength enhancement interaction feature vector.

[0051] Specifically, the motor control instruction generation module 334 is used to generate motor control instructions based on the antenna angle-signal strength enhancement interaction feature vector. In a specific example of the present application, the antenna angle-signal strength enhancement interaction feature vector is input into the decoder-based antenna angle dynamic optimization module to obtain a recommended antenna angle value. In the technical solution of the present application, the decoder adopts a recursive neural network structure, which can predict and output the optimal antenna angle of the drone under the current flight conditions through layer-by-layer decoding based on the correlation response information between the antenna angle and signal strength contained in the antenna angle-signal strength enhancement interaction feature vector through training optimization. Furthermore, based on the comparison between the current angle value of the antenna and the recommended antenna angle value, a motor control instruction is generated.

[0052] Here, when the implicit correlation feature vector between antenna angle samples and the implicit correlation feature vector between signal strength samples respectively represent the local implicit correlation features between antenna angle bidirectional samples and the local implicit correlation features between signal strength bidirectional samples, when performing feature interaction optimization based on feature mapping deviation compensation, the cross-modal semantic interaction features of the antenna angle-signal strength enhanced interaction feature vector will have probabilistic convergence and divergence relative to the interaction differences of each single modal feature, thereby affecting the accuracy of the decoding results.

[0053] Preferably, inputting the antenna angle-signal strength enhancement interaction feature vector into a decoder-based antenna angle dynamic optimization module to obtain a recommended antenna angle value comprises:

[0054] Determine the median eigenvalue, maximum eigenvalue and minimum eigenvalue in the antenna angle-signal strength enhancement interaction eigenvalue vector, and divide the median eigenvalue by the difference between the maximum eigenvalue and the minimum eigenvalue to obtain the antenna angle-signal strength enhancement interaction distribution probability value, that is, p=v mid / (v max ,v min ), where v mid 、v max 、v min represent the median eigenvalue, maximum eigenvalue and minimum eigenvalue in the antenna angle-signal strength enhancement interaction eigenvector respectively, and p represents the antenna angle-signal strength enhancement interaction distribution probability value;

[0055] Normalizing the antenna angle-signal strength enhancement interaction feature vector to its maximum value to obtain an antenna angle-signal strength enhancement interaction probability feature vector V;

[0056] The power function of each eigenvalue of the antenna angle-signal strength enhancement interaction probability eigenvector is calculated with one minus the antenna angle-signal strength enhancement interaction distribution probability value as an exponent to obtain the first antenna angle-signal strength enhancement interaction convergence eigenvector V1=V ⊙(1-p) , where V represents the antenna angle-signal strength enhancement interaction probability feature vector, (·) ⊙(1-p) represents a power function calculated by subtracting the antenna angle-signal strength enhancement interaction distribution probability value from one, and V1 represents a first antenna angle-signal strength enhancement interaction convergence eigenvector;

[0057] The power function of each eigenvalue of the point difference eigenvector between the unit eigenvector and the antenna angle-signal strength enhancement interaction probability eigenvector is calculated with the antenna angle-signal strength enhancement interaction distribution probability value as an exponent to obtain the second antenna angle-signal strength enhancement interaction convergence eigenvector V2= in, represents subtraction by position, I represents the unit eigenvector, V represents the antenna angle-signal strength enhancement interaction probability eigenvector, (·) ⊙p The power function is calculated using the antenna angle-signal strength enhancement interaction distribution probability value as an exponent, and V2 represents a second antenna angle-signal strength enhancement interaction convergence feature vector;

[0058] Performing a dot multiplication on the antenna angle-signal strength enhancement interaction probability feature vector and the difference between one and the antenna angle-signal strength enhancement interaction distribution probability value to obtain a first antenna angle-signal strength enhancement interaction limiting feature vector V3=V⊙(1,p), wherein V represents the antenna angle-signal strength enhancement interaction probability feature vector, p represents the antenna angle-signal strength enhancement interaction distribution probability value, ⊙ represents the dot multiplication, and V3 represents the first antenna angle-signal strength enhancement interaction limiting feature vector;

[0059] The point difference feature vector is multiplied by the antenna angle-signal strength enhancement interaction distribution probability value to obtain a second antenna angle-signal strength enhancement interaction limited feature vector Wherein, V represents the antenna angle-signal strength enhancement interaction probability feature vector, represents subtraction by position, I represents the unit eigenvector, p represents the probability value of the antenna angle-signal strength enhancement interaction distribution, ⊙ represents the dot product, and V4 represents the second antenna angle-signal strength enhancement interaction limited eigenvector;

[0060] After multiplying the first antenna angle-signal strength enhancement interactive convergence feature vector by the second antenna angle-signal strength enhancement interactive convergence feature vector, the optimized antenna angle-signal strength enhancement interactive feature vector is obtained by multiplying the first antenna angle-signal strength enhancement interactive limit feature vector by the second antenna angle-signal strength enhancement interactive limit feature vector. Wherein, V1 represents the first antenna angle-signal strength enhancement interactive convergence feature vector, V2 represents the second antenna angle-signal strength enhancement interactive convergence feature vector, V3 represents the first antenna angle-signal strength enhancement interactive limit feature vector, V4 represents the second antenna angle-signal strength enhancement interactive limit feature vector, ⊙ represents the dot product, Indicates adding by position, V ′ represents the optimized antenna angle-signal strength enhancement interaction feature vector;

[0061] The optimized antenna angle-signal strength enhancement interaction feature vector is passed through a decoder-based antenna angle dynamic optimization module to obtain a recommended antenna angle value.

[0062] Based on this, by using the cross-entropy form power series of the distribution analytical probability of the antenna angle-signal strength enhancement interaction feature vector as the probability distribution convergence limit, the regression probability convergence approximation is performed on the basis of the combination of the feature set distribution and the probability density distribution of the antenna angle-signal strength enhancement interaction feature vector, so as to avoid the conservative convergence control of the fatal potential divergence of the antenna angle-signal strength enhancement interaction feature vector, and further use the probability distribution cross entropy of the antenna angle-signal strength enhancement interaction feature vector as the objective function to guide the limited recovery strategy, so as to achieve the common agility of the convergence of the probability convergence and divergence of the antenna angle-signal strength enhancement interaction feature vector to the probability density distribution space, and improve the accuracy of the recommended antenna angle value obtained by the antenna angle-signal strength enhancement interaction feature vector through the decoder-based antenna angle dynamic optimization module.

[0063] As described above, the drone 300 based on satellite Internet communication according to the embodiment of the present application can be implemented in various wireless terminals, such as a server with a drone control algorithm based on satellite Internet communication. In a possible implementation, the drone 300 based on satellite Internet communication according to the embodiment of the present application can be integrated into the wireless terminal as a software module and / or a hardware module. For example, the drone 300 based on satellite Internet communication can be a software module in the operating system of the wireless terminal, or can be an application developed for the wireless terminal; of course, the drone 300 based on satellite Internet communication can also be one of the many hardware modules of the wireless terminal.

[0064] Alternatively, in another example, the satellite Internet communication-based drone 300 and the wireless terminal may also be separate devices, and the satellite Internet communication-based drone 300 may be connected to the wireless terminal via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

[0065] Furthermore, a method for controlling a UAV based on satellite Internet communication is also provided.

[0066] Figure 4 FIG. 1 is a flow chart of a method for controlling a UAV based on satellite Internet communication according to an embodiment of the present application. Figure 4As shown, according to the embodiment of the present application, the drone control method based on satellite Internet communication includes the following steps: S1, obtaining the signal strength values ​​of the antenna at different angles collected by the signal strength detector to obtain a sequence of antenna angle values ​​and a sequence of signal strength values; S2, performing inter-sample correlation coding on the sequence of antenna angle values ​​and the sequence of signal strength values ​​to obtain an implicit correlation feature vector between antenna angle samples and an implicit correlation feature vector between signal strength samples; S3, performing feature interaction analysis on the implicit correlation feature vector between antenna angle samples and the implicit correlation feature vector between signal strength samples to obtain an antenna angle-signal strength enhancement interaction feature vector;

[0067] S4, generating a motor control instruction based on the antenna angle-signal strength enhancement interaction feature vector.

[0068] In summary, according to the embodiment of the present application, a drone control method based on satellite Internet communication is explained, which utilizes deep learning-based data processing technology to perform correlation and interactive analysis on the collected signal strength and the corresponding antenna angle, so as to explore the dynamic correlation pattern between the signal strength and the antenna angle, and based on this, intelligently recommend and optimize the antenna angle, so as to achieve automatic control of the antenna angle to maintain a stable connection with the satellite, thereby improving the communication performance and mission execution efficiency of the drone in complex environments.

[0069] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A UAV based on satellite Internet communication, characterized in that: include: The drone body; A satellite communication receiving device installed on the drone body, wherein the satellite communication receiving device includes an antenna, a motor for adjusting the angle of the antenna, a power supply for powering the motor, and a controller, wherein the controller is used to control the motor to adjust the angle of the antenna; Wherein, the controller comprises: An antenna status monitoring module is used to obtain the signal strength values ​​of the antenna at different angles collected by the signal strength detector to obtain a sequence of antenna angle values ​​and a sequence of signal strength values; An inter-sample correlation analysis module, used for performing inter-sample correlation coding on the sequence of antenna angle values ​​and the sequence of signal strength values ​​respectively to obtain an implicit correlation feature vector between antenna angle samples and an implicit correlation feature vector between signal strength samples; An interaction analysis module, used for performing feature interaction analysis on the implicit correlation feature vectors between the antenna angle samples and the implicit correlation feature vectors between the signal strength samples to obtain an antenna angle-signal strength enhancement interaction feature vector; A motor control instruction generation module, used to generate a motor control instruction based on the antenna angle-signal strength enhancement interaction feature vector; Among them, the interactive analysis module includes: a co-space mapping unit, which is used to perform co-space mapping modulation on the implicit correlation feature vectors between the antenna angle samples and the implicit correlation feature vectors between the signal strength samples to obtain the implicit correlation feature vectors between the modulated antenna angle samples and the implicit correlation feature vectors between the modulated signal strength samples; a compensatory interactive fusion unit, which is used to perform semantic mapping deviation compensation interactive fusion on the implicit correlation feature vectors between the modulated antenna angle samples and the implicit correlation feature vectors between the modulated signal strength samples to obtain the antenna angle-signal strength enhancement interactive feature vector.

2. The UAV based on satellite Internet communication according to claim 1, characterized in that: The sample association analysis module includes: The sequence of antenna angle values ​​and the sequence of signal strength values ​​are respectively input into a sequence encoder based on a forward LSTM model to obtain an implicit correlation feature vector between the antenna angle samples and an implicit correlation feature vector between the signal strength samples.

3. The satellite internet communication-based drone according to claim 2, characterized in that: The common space mapping unit is used for: The implicit correlation feature vector between the antenna angle samples and the implicit correlation feature vector between the signal strength samples are input into a co-space mapper based on a fully connected layer to obtain the implicit correlation feature vector between the modulated antenna angle samples and the implicit correlation feature vector between the modulated signal strength samples, wherein the implicit correlation feature vector between the modulated antenna angle samples and the implicit correlation feature vector between the modulated signal strength samples have the same characteristic scale.

4. The UAV based on satellite Internet communication according to claim 3, characterized in that: The compensatory interactive fusion unit comprises: A compensation factor calculation subunit, used to respectively calculate the mapping deviation semantic compensation factors of the implicit correlation feature vector between the modulated antenna angle samples and the implicit correlation feature vector between the modulated signal strength samples relative to the implicit correlation feature vector between the antenna angle samples and the implicit correlation feature vector between the signal strength samples to obtain the antenna angle mapping deviation semantic compensation factor and the signal strength mapping deviation semantic compensation factor; A feature compensation subunit, used to compensate the implicit correlation feature vector between the modulated antenna angle samples and the implicit correlation feature vector between the modulated signal strength samples based on the antenna angle mapping deviation semantic compensation factor and the signal strength mapping deviation semantic compensation factor to obtain the implicit correlation feature vector between the compensated antenna angle samples and the implicit correlation feature vector between the compensated signal strength samples; The multi-dimensional interaction subunit is used to input the implicit correlation feature vector between the compensated antenna angle samples and the implicit correlation feature vector between the compensated signal strength samples into the multi-dimensional significant interaction network between features to obtain the antenna angle-signal strength enhancement interaction feature vector.

5. The UAV based on satellite Internet communication according to claim 4, characterized in that: The compensation factor calculation subunit includes: An interpolation processing secondary subunit is used to perform interpolation processing on the implicit correlation feature vectors between the antenna angle samples and the implicit correlation feature vectors between the signal strength samples to obtain the implicit correlation feature vectors between the antenna angle samples after interpolation processing and the implicit correlation feature vectors between the signal strength samples after interpolation processing, wherein the implicit correlation feature vectors between the antenna angle samples after interpolation processing, the implicit correlation feature vectors between the signal strength samples after interpolation processing, the implicit correlation feature vectors between the modulated antenna angle samples and the implicit correlation feature vectors between the modulated signal strength samples have the same characteristic scale; The mapping deviation measurement secondary subunit is used to input the implicit correlation feature vector between the antenna angle samples after interpolation processing and the implicit correlation feature vector between the antenna angle samples after modulation, the implicit correlation feature vector between the signal strength samples after interpolation processing and the implicit correlation feature vector between the signal strength samples after modulation into the feature mapping deviation semantic measurement network respectively to obtain the antenna angle mapping deviation semantic compensation factor and the signal strength mapping deviation semantic compensation factor.

6. The satellite internet communication based drone according to claim 5, characterized in that: The mapping deviation metric secondary subunit is used to: Calculate the positional dotted division vector between the implicit correlation feature vector between the modulated antenna angle samples and the implicit correlation feature vector between the interpolated antenna angle samples, and calculate the base-two logarithm of the absolute value of each eigenvalue in the dotted division vector to obtain an antenna angle mapping deviation representation vector; Calculate the dot product vector between the antenna angle mapping deviation representation vector and the implicitly associated eigenvector between the modulated antenna angle samples, and calculate the exponential function value with e as the base and the sum of the eigenvalues ​​of the dot product vector as the exponent to obtain the antenna angle mapping deviation semantic compensation factor.

7. The UAV based on satellite Internet communication according to claim 6, characterized in that: The multi-dimensional interaction subunit is used to: Respectively calculating the implicit correlation feature vectors between the compensated antenna angle samples and the implicit correlation feature vectors between the compensated signal strength samples by position point subtraction, position point multiplication, position point addition and position point division to obtain a first antenna angle-signal strength interaction representation vector, a second antenna angle-signal strength interaction representation vector, a third antenna angle-signal strength interaction representation vector and a fourth antenna angle-signal strength interaction representation vector; The first antenna angle-signal strength interaction representation vector, the second antenna angle-signal strength interaction representation vector, the third antenna angle-signal strength interaction representation vector and the fourth antenna angle-signal strength interaction representation vector are cascaded and fused, and then input into a one-dimensional convolutional layer including a maximum pooling layer to obtain the antenna angle-signal strength enhanced interaction feature vector.

8. The UAV based on satellite Internet communication according to claim 7, characterized in that: The motor control instruction generation module is used to: Inputting the antenna angle-signal strength enhancement interaction feature vector into a decoder-based antenna angle dynamic optimization module to obtain a recommended antenna angle value; The motor control command is generated based on a comparison between a current angle value of the antenna and the recommended antenna angle value.

9. A method for controlling a drone based on satellite Internet communication, characterized in that: include: Acquire signal strength values ​​of the antenna at different angles collected by a signal strength detector to obtain a sequence of antenna angle values ​​and a sequence of signal strength values; Performing inter-sample correlation coding on the sequence of antenna angle values ​​and the sequence of signal strength values ​​respectively to obtain an implicit correlation feature vector between antenna angle samples and an implicit correlation feature vector between signal strength samples; Performing feature interaction analysis on the implicit correlation feature vectors between the antenna angle samples and the implicit correlation feature vectors between the signal strength samples to obtain an antenna angle-signal strength enhancement interaction feature vector; Based on the antenna angle-signal strength enhancement interaction feature vector, a motor control command is generated.

Images