UAV data transmission system and method based on high-orbit and low-orbit satellite fusion communication

Abstract

The present invention discloses a data transmission system and method for an unmanned aerial vehicle based on high-orbit and low-orbit satellite fusion communication, and specifically relates to the technical field of data transmission, including obtaining task data types of the unmanned aerial vehicle in a basic communication link, dividing the task data types, obtaining data division results, obtaining communication characteristic data of the basic communication link, generating a basic connection coefficient based on the communication characteristic data, analyzing the basic connection coefficient, and judging whether to generate a link adjustment instruction; receiving the link adjustment instruction, and obtaining dynamic fusion training data of the unmanned aerial vehicle based on the basic connection coefficient; the present invention establishes optimization models of bandwidth utilization and data rate for low-orbit real-time transmission links and high-orbit non-real-time transmission links, respectively, to ensure the adaptive adjustment capability of the transmission link under various environments, and to help achieve the reliability of data transmission and the high efficiency of task execution.

Description

Technical Field

[0001] The present invention relates to the field of data transmission technology, and more specifically, to a UAV data transmission system and method based on high-orbit and low-orbit satellite fusion communication. Background Art

[0002] Drones are widely used in a variety of scenarios (such as disaster relief, environmental monitoring, logistics and transportation). Their missions usually require real-time transmission of high-quality data (such as video, images, location information, etc.) to ground control centers or the cloud. However, traditional data transmission methods are limited by signal coverage and bandwidth, and are difficult to meet the needs in complex terrain or long-distance scenarios.

[0003] Among the existing methods, for example, the Chinese patent application with publication number CN111600643A discloses a satellite communication resource scheduling method and device storage medium, including: extracting task description parameters from the received task requirement information to generate task description information; based on the task description information, using a resource decision model to determine the task category, wherein the resource decision model is generated based on a training sample set, and the training sample set includes historical tasks and historical resource scheduling plans; searching for a resource scheduling plan corresponding to the task category from a pre-established resource scheduling rule library, wherein the resource scheduling rules in the resource scheduling rule library are determined based on the resource requirement type of the task and the matching degree between the task indicator characteristics and the satellite communication resource characteristics; scheduling satellite communication resources according to the found resource scheduling plan. Although the above method can improve the bandwidth utilization of the satellite network and prevent network congestion, research and application of the above method and the prior art have found that the above method and the prior art have at least the following partial defects:

[0004] Although the above method takes into account the matching degree between mission requirements and satellite resources, the maintenance and update of rules may lag behind the changes in actual requirements. When the rule base cannot quickly adapt to new mission indicators or changes in the communication environment, the scheduling efficiency will be reduced.

[0005] To this end, the present invention provides a UAV data transmission system and method based on high-orbit and low-orbit satellite fusion communication. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art, the present invention provides a UAV data transmission system and method based on high-orbit and low-orbit satellite fusion communications to solve the problems raised in the above-mentioned background technology.

[0007] To achieve the above purpose, the present invention provides the following technical solutions:

[0008] In a first aspect, the present invention provides a method for transmitting data of an unmanned aerial vehicle based on high-orbit and low-orbit satellite fusion communication, comprising:

[0009] Step 1: In the basic communication link, the mission data type of the UAV is obtained, and the mission data type is divided to obtain a data division result, wherein the data division result includes low-orbit real-time data and high-orbit non-real-time data;

[0010] Step 2: Acquire communication characteristic data of the basic communication link, generate a basic connection coefficient based on the communication characteristic data, analyze the basic connection coefficient, and determine whether to generate a link adjustment instruction;

[0011] Step 3: Receive link adjustment instructions and obtain the dynamic fusion training data of the UAV based on the basic connection coefficient;

[0012] Step 4: Based on the dynamic fusion training data, a fusion prediction model for feeding back the fusion coefficient of the UAV is trained;

[0013] Step 5: Analyze the fusion coefficient output by the fusion prediction model to determine the main link instruction, which includes the high-orbit main link instruction, the parallel transmission instruction and the low-orbit main link instruction.

[0014] Furthermore, the basic communication link includes a low-orbit link and a high-orbit link, the mission data type has R types, including video data, sensor data, flight control data and real-time control instructions, the video data includes aerial video and monitoring video; the low-orbit real-time data refers to the data transmission of the UAV through the low-orbit satellite; the high-orbit non-real-time data refers to the data transmission of the UAV through the high-orbit satellite;

[0015] Methods for dividing task data types and obtaining data division results include:

[0016] Step a1: Obtain the task type index of the rth task data type; r = 1, 2, ..., R;

[0017] Step a2: Set N task type index intervals, and set N data partitioning results corresponding to the N task type index intervals; N is equal to 2, and each task type index interval is bound if and only if there is a data partitioning result associated with it;

[0018] Step a3: Compare the task type index of the r-th task data type with each task type index interval to obtain the task type index interval into which the task type index of the r-th task data type falls;

[0019] Step a4: according to the task type index interval into which the task type index of the r-th task data type falls, divide the r-th task data type into the corresponding data division results; and set r=r+1, and jump back to step a1;

[0020] Step a5: repeat the above steps a1 to a4 until the loop ends when r=R, so that each task data type is sequentially divided into N data division results;

[0021] The method for obtaining the task type index includes:

[0022] Type characteristic data of the acquisition task data type, wherein the type characteristic data includes the number of data transmission frames, compression rate and maximum delay time;

[0023] The type feature data is calculated by formula to obtain the task type index, and the calculation formula is:

[0024] Lx r =ln(μ1×Zs r +μ2×Ys r +μ3×Sc r );

[0025] In the formula, Lx r Represents the task type index of the r-th task data type, Zs r Indicates the number of data transmission frames of the rth task data type, Ys r represents the compression ratio of the rth task data type, Sc r represents the maximum delay duration of the r-th task data type, μ1, μ2 and μ3 are weight factors, and ln(·) represents the logarithmic function with e as the base.

[0026] Further, the communication characteristic data of the basic communication link includes low-orbit transmission characteristic data of the low-orbit link and high-orbit transmission characteristic data of the high-orbit link, wherein the low-orbit transmission characteristic data includes a first signal strength, a low-orbit data transmission rate, and a first bandwidth utilization rate; the high-orbit transmission characteristic data of the high-orbit link includes a second signal strength, a high-orbit data transmission rate, and a second bandwidth utilization rate;

[0027] The method for acquiring the low-orbit data transmission rate comprises:

[0028] The transmission rate of the current data packet is collected through a communication protocol stack; the communication protocol stack is TCP / IP or UDP;

[0029] Monitor the total amount of data transferred within a preset time;

[0030] Calculate the total data volume and the transmission rate to obtain the low-orbit data transmission rate, and the calculation formula is:

[0031] In the formula, Cs D It represents the data transmission rate between the UAV and the low-orbit satellite, Sz represents the transmission rate of the current data packet, and CL represents the total amount of data transmitted per unit time;

[0032] The method for acquiring the first bandwidth utilization includes:

[0033] The data transmission rate of the UAV is obtained, and the first total bandwidth allocated to the UAV by the low-orbit satellite is obtained, and the data transmission rate and the allocated first total bandwidth are calculated by formula to obtain the first bandwidth utilization rate; the calculation formula is:

[0034] Where Ly1 represents the first bandwidth utilization of the low-orbit link, and Fz represents the first total bandwidth allocated by the low-orbit satellite to the UAV.

[0035] Furthermore, a method for obtaining a basic connection coefficient based on communication characteristic data includes:

[0036] Step b1: obtaining low-orbit transmission characteristic data of the low-orbit link, formulating the low-orbit transmission characteristic data, and obtaining a low-orbit connection evaluation coefficient;

[0037] Step b2: Formulate the high-orbit transmission characteristic data to obtain the high-orbit connection evaluation coefficient;

[0038] Step b3: Fuse the low-orbit connection evaluation coefficient and the high-orbit connection evaluation coefficient to obtain the basic connection coefficient Gx.

[0039] Furthermore, the method of analyzing the basic connection coefficient and determining whether to generate a link adjustment instruction includes:

[0040] Preset a connection coefficient threshold, the connection coefficient threshold includes Lj1 and Lj2, where Lj1>Lj2; compare the basic connection coefficient with the preset connection coefficient threshold;

[0041] If Gx>Lj1 or Gx<Lj2, a link adjustment command is generated, indicating that the UAV has unstable data transmission in the basic communication link;

[0042] If Lj1≥Gx≥Lj2, no link adjustment command is generated, indicating that the data transmission of the UAV in the basic communication link is stable.

[0043] Further, the dynamic fusion training data includes a relationship between each bandwidth value and a first link coefficient difference and a relationship between each transmission rate and a second link coefficient difference;

[0044] The method for obtaining the relationship between each bandwidth value and the first link coefficient difference includes:

[0045] Step c1: in the experimental scenario, obtain the real-time communication link coefficient for measuring the data transmission of the UAV under the hth bandwidth value;

[0046] Step c2: extracting the standard link coefficient pre-stored in the system database and used to measure the data transmission of the UAV under the set standard bandwidth value;

[0047] Step c3: Calculate the difference between the real-time communication link coefficient and the standard link coefficient for measuring the data transmission of the UAV under the h-th bandwidth value, and use the difference between the real-time communication link coefficient and the standard link coefficient for measuring the data transmission of the UAV under the h-th bandwidth value as the first link coefficient difference;

[0048] Step c4: compare the first link coefficient difference with a preset first link coefficient difference threshold. If the first link coefficient difference is less than the preset first link coefficient difference threshold, set h=h+1 and return to step c1. If the first link coefficient difference is greater than or equal to the preset first link coefficient difference threshold, record the relationship between the hth bandwidth value and the first link coefficient difference, set h=h+1 again and return to step c1.

[0049] Step c5: Repeat the above steps c1 to c4 until h is equal to the set bandwidth value H, end the loop, obtain the relationship between each bandwidth value and the first link coefficient difference, and use the relationship between each bandwidth value and the first link coefficient difference as dynamic fusion training data, where h and H are integers greater than zero.

[0050] Furthermore, a method for obtaining the relationship between each transmission rate and the second link coefficient difference includes:

[0051] Step d1: in the experimental scenario, obtain the real-time communication link coefficient for measuring the data transmission of the UAV at the kth transmission rate;

[0052] Step d2: extracting the standard link coefficient pre-stored in the system data and used to measure the data transmission of the UAV at the set standard transmission rate;

[0053] Step d3: Calculate the difference between the real-time communication link coefficient and the standard link coefficient for measuring the data transmission of the UAV at the kth transmission rate, and use the difference between the real-time communication link coefficient and the standard link coefficient of the UAV at the kth transmission rate as the second link coefficient difference;

[0054] Step d4: compare the second link coefficient difference with a preset second link coefficient difference threshold value. If the second link coefficient difference is less than the preset second link coefficient difference threshold value, set k=k+1 and return to step d1; if the second link coefficient difference is greater than or equal to the preset second link coefficient difference threshold value, record the relationship between the kth transmission rate and the second link coefficient difference, set k=k+1 again and return to step d1;

[0055] Step d5: Repeat the above steps d1 to d4 until k is equal to the set transmission rate K, end the loop, obtain the relationship between each transmission rate and the second link coefficient difference, and use the relationship between each transmission rate and the second link coefficient difference as dynamic fusion training data, k and K are integers greater than zero.

[0056] Furthermore, a method for obtaining a real-time communication link coefficient for measuring UAV data transmission under the hth bandwidth value includes:

[0057] When the network bandwidth is the hth bandwidth value, the signal strength, signal-to-noise ratio and packet loss rate are retrieved from the system database;

[0058] The signal strength, signal-to-noise ratio, and packet loss rate are formally calculated to obtain the real-time communication link coefficient, which is calculated as follows:

[0059] SLx=ln(WXh×SNR)-exp(SDL);

[0060] Where SLx is the real-time communication link coefficient, WXh is the signal strength of the drone connection, SNR is the signal-to-noise ratio, SDL is the packet loss rate, ln(·) is the logarithmic function with e as the base, and exp(·) is the exponential function; the signal strength and packet loss rate are normalized values.

[0061] Furthermore, the training logic of the fusion prediction model is as follows:

[0062] Extracting the relationship between each bandwidth value and the first link coefficient difference and the relationship between each transmission rate and the second link coefficient difference in the dynamic fusion training data;

[0063] Divide the relationship between each bandwidth value and the first link coefficient difference and the relationship between each transmission rate and the second link coefficient difference into a fusion coefficient training set and a fusion coefficient test set;

[0064] Constructing a neural network model, inputting the relationship between each bandwidth value and the first link coefficient difference and the relationship between each transmission rate and the second link coefficient difference in the fusion coefficient training set into the fusion prediction model for training, and obtaining a trained neural network model; using the fusion coefficient test set to verify the trained neural network model, and outputting a trained neural network model greater than or equal to a preset test accuracy as a fusion prediction model; the fusion prediction model is an RNN neural network, a DNN neural network or a CNN neural network model;

[0065] The method of analyzing the fusion coefficient output by the fusion prediction model and determining the main link instruction includes:

[0066] Preset fusion coefficient thresholds Rh1 and Rh2, Rh1<Rh2, and compare the fusion coefficient with the fusion coefficient threshold;

[0067] If the fusion coefficient is greater than the fusion coefficient threshold Rh2, a high-orbit main link instruction is generated;

[0068] If the fusion coefficient is less than or equal to the fusion coefficient threshold Rh2, and greater than or equal to the fusion coefficient threshold Rh1, a parallel transmission instruction is generated;

[0069] If the fusion coefficient is less than the fusion coefficient threshold Rh1, a low-orbit main link instruction is generated;

[0070] Among them, high-orbit main link instructions > parallel transmission instructions > low-orbit main link instructions.

[0071] In a second aspect, the present invention provides a UAV data transmission system based on high- and low-orbit satellite fusion communication, which is used to implement the above-mentioned UAV data transmission method based on high- and low-orbit satellite fusion communication, including:

[0072] A type classification module is used to obtain the mission data type of the UAV in the basic communication link, classify the mission data type, and obtain data classification results, wherein the data classification results include low-orbit real-time data and high-orbit non-real-time data;

[0073] A judgment module, used to obtain communication characteristic data of a basic communication link, generate a basic connection coefficient based on the communication characteristic data, analyze the basic connection coefficient, and judge whether to generate a link adjustment instruction;

[0074] A data acquisition module, used to receive link adjustment instructions and acquire dynamic fusion training data of the UAV based on the basic connection coefficient;

[0075] A model training module is used to train a fusion prediction model for feeding back the fusion coefficient of the UAV based on dynamic fusion training data;

[0076] The analysis module is used to analyze the fusion coefficient output by the fusion prediction model and determine the main link instruction, which includes the high-orbit main link instruction, the parallel transmission instruction and the low-orbit main link instruction.

[0077] Technical effects and advantages of the present invention:

[0078] 1. The present invention divides the UAV mission data into low-orbit real-time data and high-orbit non-real-time data according to real-time requirements by constructing a fusion communication mechanism based on high- and low-orbit satellites. The characteristics of different types of tasks such as video data, sensor data, and flight control data are classified using the mission type index and communication feature data, and the mission type index is accurately calculated through parameters such as the number of data transmission frames, compression rate, and maximum delay duration, thereby achieving accurate classification and efficient allocation of mission data. This classification method not only reduces transmission delays and resource waste, but also effectively improves the stability and efficiency of data transmission. In addition, the method establishes optimization models for bandwidth utilization and data rate for low-orbit real-time transmission links and high-orbit non-real-time transmission links, respectively, ensuring the adaptive adjustment capability of the transmission link in various environments, and helping to achieve the reliability of data transmission and the efficiency of task execution.

[0079] 2. The present invention innovatively realizes the dynamic adjustment and optimization of the UAV communication link by dynamically fusing training data and training a fusion prediction model based on a neural network. Based on the first link coefficient difference and the second link coefficient difference, a dynamic fusion model with bandwidth value and transmission rate as input is constructed, which effectively captures the dynamic change characteristics of the communication link performance. Through real-time calculation of the fusion coefficient and intelligent analysis of the link status, the present invention can generate instructions for high-orbit, parallel or low-orbit transmission links according to the actual status of the communication link, thereby realizing efficient selection and switching of the UAV communication link, and avoiding mission failures caused by unstable communication links. This method significantly improves the accuracy of data transmission, the stability of the communication link and the intelligent level of UAV mission execution, and lays a solid foundation for the application of high- and low-orbit satellite fusion communication technology in UAV mission scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] Figure 1 This is a flow chart of the UAV data transmission method based on high-orbit and low-orbit satellite fusion communication of Example 1;

[0081] Figure 2 This is a flow chart of a method for obtaining data partitioning results in Example 1;

[0082] Figure 3 This is a flow chart of a method for obtaining a basic connection coefficient based on communication characteristic data in Example 1;

[0083] Figure 4 This is a schematic diagram of the structure of the UAV data transmission system based on high- and low-orbit satellite fusion communications of Example 2. DETAILED DESCRIPTION

[0084] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0085] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0086] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and a similar second unit may be referred to as a first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0087] Example 1

[0088] See also Figure 1 As shown, this embodiment discloses a method for transmitting data of a UAV based on high-orbit and low-orbit satellite fusion communication, and the method includes:

[0089] Step 1: In the basic communication link, the mission data type of the UAV is obtained, and the mission data type is divided to obtain a data division result, wherein the data division result includes low-orbit real-time data and high-orbit non-real-time data;

[0090] It should be understood that the UAV includes multiple mission stages during operation, including the take-off mission stage, the cruise mission stage, and the landing mission stage. In the cruise mission stage, the UAV needs to communicate with high-orbit satellites and low-orbit satellites to achieve data transmission, so the stability of the communication link is particularly important. For this reason, this embodiment takes the cruise mission stage of the UAV as an example for detailed description.

[0091] It should be noted that the basic communication link includes a low-orbit link and a high-orbit link. The basic communication link refers to the communication infrastructure and protocol activation between the high-orbit satellite and the low-orbit satellite that have been completed before the UAV cruise mission. Through the basic communication link, the initial communication channel can be established, thereby supporting the normal operation of data transmission and command interaction.

[0092] There are R types of task data types, including video data, sensor data, flight control data and real-time control instructions. The video data includes aerial video and monitoring video. The sensor data includes data such as ambient temperature, ambient humidity, air pressure, radiation and pollutant index; the flight control data includes flight altitude data, flight speed data, flight direction data and GPS coordinate data; the real-time control instructions include flight route adjustment and obstacle avoidance instructions, etc.

[0093] The low-orbit real-time data refers to the data transmission of the UAV via the low-orbit satellite; the high-orbit non-real-time data refers to the data transmission of the UAV via the high-orbit satellite.

[0094] See also Figure 2 As shown, in implementation, the task data type is divided, and the method for obtaining the data division result includes:

[0095] Step a1: Obtain the task type index of the rth task data type; r = 1, 2, ..., R;

[0096] It should be noted that the methods for obtaining the task type index include:

[0097] Type characteristic data of the acquisition task data type, wherein the type characteristic data includes the number of data transmission frames, compression rate and maximum delay time;

[0098] The type feature data is calculated by formula to obtain the task type index, and the calculation formula is:

[0099] Lx r =ln(μ1×Zs r +μ2×Ys r +μ3×Sc r );

[0100] In the formula, Lx r Represents the task type index of the r-th task data type, Zs r Indicates the number of data transmission frames of the rth task data type, Ys r represents the compression ratio of the rth task data type, Sc rrepresents the maximum delay time of the rth task data type, μ1, μ2 and μ3 are weight factors, and ln(·) represents a logarithmic function with e as the base. The weight factors are all set by those skilled in the art based on experience.

[0101] Step a2: setting N task type index intervals, and setting N data division results corresponding to the N task type index intervals;

[0102] It should be understood that: N is equal to 2, and each task type index interval is bound if and only if there is a data partition result associated with it;

[0103] Step a3: Compare the task type index of the r-th task data type with each task type index interval to obtain the task type index interval into which the task type index of the r-th task data type falls;

[0104] Step a4: according to the task type index interval into which the task type index of the r-th task data type falls, divide the r-th task data type into the corresponding data division results; and set r=r+1, and jump back to step a1;

[0105] Step a5: Repeat the above steps a1 to a4 until the loop ends when r=R, so that each of the task data types is divided into N data division results in turn.

[0106] Step 2: Acquire communication characteristic data of the basic communication link, generate a basic connection coefficient based on the communication characteristic data, analyze the basic connection coefficient, and determine whether to generate a link adjustment instruction;

[0107] It should be noted that: the communication characteristic data of the basic communication link includes the low-orbit transmission characteristic data of the low-orbit link and the high-orbit transmission characteristic data of the high-orbit link, the low-orbit transmission characteristic data includes the first signal strength, the low-orbit data transmission rate and the first bandwidth utilization; the high-orbit transmission characteristic data of the high-orbit link includes the second signal strength, the high-orbit data transmission rate and the second bandwidth utilization;

[0108] Among them, the first signal strength represents the signal strength of the low-orbit satellite signal received by the drone, which is collected by the signal receiver of the communication module on the drone; the low-orbit data transmission rate represents the rate of data transmission through the low-orbit satellite; the first bandwidth utilization represents the proportion of the bandwidth used by the drone to the total available bandwidth of the low-orbit satellite.

[0109] Specifically, the method for acquiring the low-orbit data transmission rate includes:

[0110] The transmission rate of the current data packet is collected through a communication protocol stack; the communication protocol stack is TCP / IP or UDP;

[0111] The total amount of data transmitted within a preset time is monitored; the preset time is set by those skilled in the art based on experience and is not specifically limited here. The unit of the preset time can be milliseconds or seconds.

[0112] Calculate the total data volume and the transmission rate to obtain the low-orbit data transmission rate, and the calculation formula is:

[0113] In the formula, Cs D It represents the data transmission rate between the UAV and the low-orbit satellite, Sz represents the transmission rate of the current data packet, and CL represents the total amount of data transmitted per unit time.

[0114] Specifically, the method for obtaining the first bandwidth utilization includes:

[0115] The data transmission rate of the UAV is obtained, and the first total bandwidth allocated to the UAV by the low-orbit satellite is obtained, and the data transmission rate and the allocated first total bandwidth are calculated by formula to obtain the first bandwidth utilization rate; the calculation formula is:

[0116] Where Ly1 represents the first bandwidth utilization of the low-orbit link, and Fz represents the first total bandwidth allocated by the low-orbit satellite to the UAV.

[0117] It is worth noting that the second signal strength, high-orbit data transmission rate and second bandwidth utilization are similar to the method of obtaining the low-orbit transmission characteristic data of the low-orbit link. The difference is that the characteristic data of the high-orbit link between the UAV and the high-orbit satellite is obtained, which will not be repeated here.

[0118] See also Figure 3 As shown, in implementation, the method for obtaining the basic connection coefficient based on the communication characteristic data includes:

[0119] Step b1: Obtain the low-orbit transmission characteristic data of the low-orbit link, perform formula processing on the low-orbit transmission characteristic data, and obtain the low-orbit connection evaluation coefficient, the calculation formula of which is:

[0120] X D =ln(1+Xh1)×ω1+exp(Cs D )×ω2+Ly1×ω3

[0121] Where, X D represents the low-orbit connection evaluation coefficient, Xh1 represents the first signal strength, ω1, ω2 and ω3 are all weight factors, and ω1+ω2+ω3=1; ln(·) represents a logarithmic function with e as the base, and exp(·) represents an exponential function. The weight factors are all set by those skilled in the art based on experience.

[0122] Step b2: Formulate the high-orbit transmission characteristic data to obtain the high-orbit connection evaluation coefficient, which is calculated as follows:

[0123] X G =ln(1+Xh2)×θ1+exp(Cs G )×θ2+Ly2×θ3

[0124] Where, X G represents the high-rail connection evaluation coefficient, Xh2 represents the second signal strength, Cs G represents the high-orbit data transmission rate, Ly2 represents the second bandwidth utilization, θ1, θ2 and θ3 are all weight factors, and θ1+θ2+θ3=1; ln(·) represents a logarithmic function with e as the base, and exp(·) represents an exponential function. The weight factors are all set by those skilled in the art based on experience.

[0125] Step b3: Fusing the low-orbit connection evaluation coefficient and the high-orbit connection evaluation coefficient to obtain the basic connection coefficient;

[0126] Gx = exp[α1×ln(X D )+α2×ln(X G )];

[0127] In the formula, Gx represents the basic connection coefficient, α1 and α1 are weight factors, and ln(·) represents a logarithmic function with e as the base. The weight factors are all set by those skilled in the art based on experience.

[0128] In the above formula, logarithms and exponents are used to balance data of different magnitudes, avoid the dominant role of larger or smaller values, and smoothly integrate the data of low-orbit satellites and high-orbit satellites.

[0129] In implementation, the method of analyzing the basic connection coefficient and determining whether to generate a link adjustment instruction includes:

[0130] Preset a connection coefficient threshold, the connection coefficient threshold includes Lj1 and Lj2, where Lj1>Lj2; compare the basic connection coefficient with the preset connection coefficient threshold;

[0131] If Gx>Lj1 or Gx<Lj2, a link adjustment instruction is generated, indicating that the data transmission of the UAV in the basic communication link is unstable, and the basic communication link of the UAV needs to be dynamically adjusted to ensure the stability of the data transmission of the UAV;

[0132] If Lj1≥Gx≥Lj2, no link adjustment instruction is generated, indicating that the data transmission of the UAV in the basic communication link is stable and there is no need to adjust the basic communication link of the UAV.

[0133] Step 3: Receive link adjustment instructions and obtain the dynamic fusion training data of the UAV based on the basic connection coefficient;

[0134] The dynamic fusion training data includes the relationship between each bandwidth value and the first link coefficient difference and the relationship between each transmission rate and the second link coefficient difference.

[0135] It should be noted that: each bandwidth value is monitored by the drone through its own hardware equipment (such as a network adapter, a satellite terminal) and pre-stored in the database, and each transmission rate is monitored by its own hardware equipment (such as a network adapter, a satellite terminal) on the amount of transmitted data and pre-stored in the system database.

[0136] In implementation, the method for obtaining the relationship between each bandwidth value and the first link coefficient difference includes:

[0137] Step c1: in the experimental scenario, obtain the real-time communication link coefficient for measuring the data transmission of the UAV under the hth bandwidth value;

[0138] It should be noted that the method for obtaining the real-time communication link coefficient for measuring the data transmission of the UAV under the h-th bandwidth value includes:

[0139] When the network bandwidth is the hth bandwidth value, the signal strength, signal-to-noise ratio and packet loss rate are retrieved from the system database;

[0140] The signal strength, signal-to-noise ratio, and packet loss rate are formally calculated to obtain the real-time communication link coefficient, which is calculated as follows:

[0141] SLx=ln(WXh×SNR)-exp(SDL);

[0142] Where SLx is the real-time communication link coefficient, WXh is the signal strength of the drone connection, SNR is the signal-to-noise ratio, SDL is the packet loss rate, ln(·) is the logarithmic function with e as the base, and exp(·) is the exponential function.

[0143] It is additionally noted that the signal strength and packet loss rate are both normalized values.

[0144] Step c2: extracting the standard link coefficient pre-stored in the system database and used to measure the data transmission of the UAV under the set standard bandwidth value;

[0145] It should be understood that the standard link coefficient used to measure drone data transmission under a set standard bandwidth value is obtained through prior experimental settings.

[0146] Step c3: Calculate the difference between the real-time communication link coefficient and the standard link coefficient for measuring the data transmission of the UAV under the h-th bandwidth value, and use the difference between the real-time communication link coefficient and the standard link coefficient for measuring the data transmission of the UAV under the h-th bandwidth value as the first link coefficient difference;

[0147] Step c4: compare the first link coefficient difference with a preset first link coefficient difference threshold. If the first link coefficient difference is less than the preset first link coefficient difference threshold, set h=h+1 and return to step c1. If the first link coefficient difference is greater than or equal to the preset first link coefficient difference threshold, record the relationship between the hth bandwidth value and the first link coefficient difference, set h=h+1 again and return to step c1.

[0148] Step c5: Repeat the above steps c1 to c4 until h is equal to the set bandwidth value H, end the loop, obtain the relationship between each bandwidth value and the first link coefficient difference, and use the relationship between each bandwidth value and the first link coefficient difference as dynamic fusion training data, where h and H are integers greater than zero.

[0149] In implementation, a method for obtaining the relationship between each transmission rate and the second link coefficient difference includes:

[0150] Step d1: in the experimental scenario, obtain the real-time communication link coefficient for measuring the data transmission of the UAV at the kth transmission rate;

[0151] It should be noted that the method for obtaining the real-time communication link coefficient for measuring drone data transmission at the kth transmission rate is the same as the method for obtaining the real-time communication link coefficient for measuring drone data transmission at the hth bandwidth value mentioned above. For details, refer to the above.

[0152] Step d2: extracting the standard link coefficient pre-stored in the system data and used to measure the data transmission of the UAV at the set standard transmission rate;

[0153] It should be understood that: the standard link coefficient for measuring drone data transmission under the above-mentioned set standard bandwidth value, the standard link coefficient for measuring drone data transmission under the set standard transmission rate is obtained by prior experimental setting.

[0154] Step d3: Calculate the difference between the real-time communication link coefficient and the standard link coefficient for measuring the data transmission of the UAV at the kth transmission rate, and use the difference between the real-time communication link coefficient and the standard link coefficient of the UAV at the kth transmission rate as the second link coefficient difference;

[0155] Step d4: compare the second link coefficient difference with a preset second link coefficient difference threshold value. If the second link coefficient difference is less than the preset second link coefficient difference threshold value, set k=k+1 and return to step d1; if the second link coefficient difference is greater than or equal to the preset second link coefficient difference threshold value, record the relationship between the kth transmission rate and the second link coefficient difference, set k=k+1 again and return to step d1;

[0156] Step d5: Repeat the above steps d1 to d4 until k is equal to the set transmission rate K, end the loop, obtain the relationship between each transmission rate and the second link coefficient difference, and use the relationship between each transmission rate and the second link coefficient difference as dynamic fusion training data, k and K are integers greater than zero.

[0157] This step is used to evaluate the communication performance of the UAV under different bandwidth conditions or different transmission rate conditions, and can provide dynamic parameter support for the integrated communication of high and low orbit satellites.

[0158] Step 4: Based on the dynamic fusion training data, a fusion prediction model is trained to provide feedback on the fusion coefficient of the UAV.

[0159] The training logic of the fusion prediction model is as follows:

[0160] Extracting the relationship between each bandwidth value and the first link coefficient difference and the relationship between each transmission rate and the second link coefficient difference in the dynamic fusion training data;

[0161] Divide the relationship between each bandwidth value and the first link coefficient difference and the relationship between each transmission rate and the second link coefficient difference into a fusion coefficient training set and a fusion coefficient test set;

[0162] A neural network model is constructed, and the relationship between each bandwidth value and the first link coefficient difference and the relationship between each transmission rate and the second link coefficient difference in the fusion coefficient training set are input into the fusion prediction model for training to obtain a trained neural network model; the trained neural network model is verified using the fusion coefficient test set, and a trained neural network model greater than or equal to a preset test accuracy is output as a fusion prediction model; the fusion prediction model is an RNN neural network, a DNN neural network or a CNN neural network model.

[0163] The above steps dynamically analyze the data transmission of the UAV under different conditions through real-time sensor data and pre-built models, and automatically adjust the communication link system to achieve precise management of the UAV's data transmission, ensuring that the UAV can achieve the best data transmission efficiency under different flight conditions.

[0164] Step 5: Analyze the fusion coefficient output by the fusion prediction model to determine the main link instruction, which includes the high-orbit main link instruction, the parallel transmission instruction and the low-orbit main link instruction.

[0165] In implementation, the fusion coefficient output by the fusion prediction model is analyzed to determine the main link instruction method including:

[0166] Preset fusion coefficient thresholds Rh1 and Rh2, Rh1<Rh2, and compare the fusion coefficient with the fusion coefficient threshold;

[0167] If the fusion coefficient is greater than the fusion coefficient threshold Rh2, a high-orbit main link instruction is generated. According to the high-orbit main link instruction, the UAV and the high-orbit satellite are used as the main communication connection to increase the data transmission of high-orbit non-real-time data;

[0168] If the fusion coefficient is less than or equal to the fusion coefficient threshold Rh2, and greater than or equal to the fusion coefficient threshold Rh1, a parallel transmission instruction is generated;

[0169] If the fusion coefficient is less than the fusion coefficient threshold Rh1, a low-orbit main link instruction is generated. According to the low-orbit main link instruction, the UAV and the low-orbit satellite are used as the main communication connection to reduce the data transmission of high-orbit non-real-time data and improve the data transmission of low-orbit real-time data.

[0170] Among them, the high-orbit main link instruction>parallel transmission instruction>low-orbit main link instruction; that is, the priority of the high-orbit main link instruction is higher than the priority of the parallel transmission instruction, and the priority of the parallel transmission instruction is higher than the priority of the low-orbit main link instruction.

[0171] It should be noted that: the high-orbit main link instruction indicates that the current UAV is in a non-real-time demand task. Therefore, it is necessary to increase the bandwidth allocation, data caching and transmission confirmation measures of the communication link between the UAV and the high-orbit satellite to ensure the data transmission of high-orbit non-real-time data; the parallel transmission instruction indicates that the current UAV is in a multi-task collaborative state. Therefore, it is necessary to increase the link monitoring mechanism to ensure the high efficiency and stability of the UAV data transmission; the low-orbit main link instruction indicates that the current UAV is in a real-time task state. Therefore, it is necessary to increase the communication link, bandwidth reservation and real-time transmission confirmation protocol between the UAV and the low-orbit satellite to ensure the data transmission of low-orbit real-time data.

[0172] This embodiment greatly improves the data transmission efficiency of UAVs in complex communication environments through the precise division of task data types and the optimization method of fusion transmission. The low-orbit link is used for the transmission of high-real-time data, while the high-orbit link is responsible for the transmission of non-real-time data, realizing the reasonable allocation of communication resources and the balance of link loads. At the same time, the dynamic adjustment mechanism of the basic communication link effectively responds to problems such as link fluctuations and network congestion by real-time analysis of the communication characteristic data (such as signal strength, transmission rate, and bandwidth utilization) of the low-orbit and high-orbit links. Combined with dynamic fusion training data, the real-time prediction and adjustment of the transmission path by the fusion prediction model ensures the stability of the communication link. On this basis, the data transmission of the UAV can be carried out efficiently, especially in environments such as limited bandwidth and limited channel quality, which significantly reduces the data packet loss rate and improves the reliability and efficiency of task completion.

[0173] This embodiment realizes the high intelligence and environmental adaptability of the UAV communication system by integrating the prediction model and the intelligent link command generation mechanism. Based on a variety of feature data and fusion training, the neural network model can accurately predict the fusion coefficient and dynamically select the optimal transmission path (low orbit, high orbit or parallel transmission). This mechanism is not only applicable to conventional data transmission scenarios, but also can be adaptively adjusted when the mission requirements are diversified and the communication environment is complex and changeable. For example, the high-orbit main link adapts to the high capacity requirements of non-real-time tasks, and parallel transmission provides multi-link support under high real-time and high stability requirements, which improves the flexibility of communication. In addition, the dynamic adjustment of the fusion coefficient is combined with threshold analysis to accurately match mission requirements and link characteristics, reduce communication delays, and ensure the reliable data support capabilities of drones in emergency response or long-term missions.

[0174] Example 2

[0175] See also Figure 4 As shown, this embodiment provides a UAV data transmission system based on high-orbit satellite fusion communication, a type classification module, a judgment module, a data acquisition module, a model training module and an analysis module; each module is connected by wire and / or wireless means to realize data transmission between modules; the system includes:

[0176] A type classification module is used to obtain the mission data type of the UAV in the basic communication link, classify the mission data type, and obtain data classification results, wherein the data classification results include low-orbit real-time data and high-orbit non-real-time data;

[0177] A judgment module, used to obtain communication characteristic data of a basic communication link, generate a basic connection coefficient based on the communication characteristic data, analyze the basic connection coefficient, and judge whether to generate a link adjustment instruction;

[0178] A data acquisition module, used to receive link adjustment instructions and acquire dynamic fusion training data of the UAV based on the basic connection coefficient;

[0179] A model training module is used to train a fusion prediction model for feeding back the fusion coefficient of the UAV based on dynamic fusion training data;

[0180] The analysis module is used to analyze the fusion coefficient output by the fusion prediction model and determine the main link instruction, which includes the high-orbit main link instruction, the parallel transmission instruction and the low-orbit main link instruction.

[0181] Example 3

[0182] This embodiment provides an electronic device. The electronic device may include one or more processors and one or more memories. The memories store computer-readable codes, and when the computer-readable codes are executed by the one or more processors, the method for transmitting data of a drone based on high-orbit satellite fusion communication as described above may be executed.

[0183] The system according to the implementation mode of the present application can also be implemented with the help of the architecture of the electronic device described below. The electronic device may include a bus, one or more CPUs, a read-only memory (ROM), a random access memory (RAM), a communication port connected to a network, input / output, a hard disk, etc. A storage device in the electronic device, such as a ROM or a hard disk, can store the drone data transmission method based on high- and low-orbit satellite fusion communication provided in the present application. Furthermore, the electronic device may also include a user interface. Of course, the above architecture is only exemplary. When implementing different devices, one or more components in the above electronic device may be omitted according to actual needs.

[0184] Example 4

[0185] This embodiment provides a computer-readable storage medium storing instructions. When the instructions are executed on a computer, the computer executes the UAV data transmission method based on high- and low-orbit satellite fusion communication of Embodiment 1.

[0186] The formulas involved in the above are all calculated by removing dimensions and taking their numerical values. They are a formula that is closest to the actual situation obtained by collecting a large amount of data and performing software simulation. The weight factors in the formula and the preset thresholds in the analysis process are set by technicians in this field according to actual conditions or obtained by simulating a large amount of data; the size of the weight factor is to quantify each parameter to obtain a specific value for subsequent comparison. The size of the weight factor depends on the amount of sample data and the corresponding processing coefficient initially set by technicians in this field for each group of sample data; as long as it does not affect the proportional relationship between the parameter and the quantized value.

[0187] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0188] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A UAV data transmission method based on high-orbit and low-orbit satellite fusion communication, characterized in that: include: Step 1: In the basic communication link, the mission data type of the UAV is obtained, and the mission data type is divided to obtain a data division result, wherein the data division result includes low-orbit real-time data and high-orbit non-real-time data; Step 2: Acquire communication characteristic data of the basic communication link, generate a basic connection coefficient based on the communication characteristic data, analyze the basic connection coefficient, and determine whether to generate a link adjustment instruction; Step 3: Receive link adjustment instructions and obtain the dynamic fusion training data of the UAV based on the basic connection coefficient; Step 4: Based on the dynamic fusion training data, a fusion prediction model for feeding back the fusion coefficient of the UAV is trained; Step 5: Analyze the fusion coefficient output by the fusion prediction model to determine the main link instruction, which includes the high-orbit main link instruction, the parallel transmission instruction and the low-orbit main link instruction.

2. The method for transmitting data of unmanned aerial vehicle based on high-orbit and low-orbit satellite fusion communication according to claim 1 is characterized in that: The basic communication link includes a low-orbit link and a high-orbit link. The mission data type has R types, including video data, sensor data, flight control data and real-time control instructions. The video data includes aerial video and monitoring video. The low-orbit real-time data refers to the data transmission of the UAV via the low-orbit satellite; the high-orbit non-real-time data refers to the data transmission of the UAV via the high-orbit satellite. Methods for dividing task data types and obtaining data division results include: Step a1: Obtain the task type index of the rth task data type; r = 1, 2, ..., R; Step a2: Set N task type index intervals, and set N data partitioning results corresponding to the N task type index intervals; N is equal to 2, and each task type index interval is bound if and only if there is a data partitioning result associated with it; Step a3: Compare the task type index of the r-th task data type with each task type index interval to obtain the task type index interval into which the task type index of the r-th task data type falls; Step a4: according to the task type index interval into which the task type index of the r-th task data type falls, divide the r-th task data type into the corresponding data division results; and set r=r+1, and jump back to step a1; Step a5: repeat the above steps a1 to a4 until the loop ends when r=R, so that each task data type is sequentially divided into N data division results; The method for obtaining the task type index includes: Type characteristic data of the acquisition task data type, wherein the type characteristic data includes the number of data transmission frames, compression rate and maximum delay time; The type feature data is calculated by formula to obtain the task type index, and the calculation formula is: Lx r =ln(μ1×Zs r +μ2×Ys r +μ3×Sc r ); In the formula, Lx r Represents the task type index of the r-th task data type, Zs r Indicates the number of data transmission frames of the rth task data type, Ys r represents the compression ratio of the rth task data type, Sc r represents the maximum delay duration of the r-th task data type, μ1, μ2 and μ3 are weight factors, and ln(·) represents the logarithmic function with e as the base.

3. The method for transmitting data of unmanned aerial vehicle based on high-orbit and low-orbit satellite fusion communication according to claim 2 is characterized in that: The communication characteristic data of the basic communication link includes low-orbit transmission characteristic data of the low-orbit link and high-orbit transmission characteristic data of the high-orbit link, wherein the low-orbit transmission characteristic data includes a first signal strength, a low-orbit data transmission rate, and a first bandwidth utilization rate; the high-orbit transmission characteristic data of the high-orbit link includes a second signal strength, a high-orbit data transmission rate, and a second bandwidth utilization rate; The method for acquiring the low-orbit data transmission rate comprises: The transmission rate of the current data packet is collected through a communication protocol stack; the communication protocol stack is TCP / IP or UDP; Monitor the total amount of data transferred within a preset time; Calculate the total data volume and the transmission rate to obtain the low-orbit data transmission rate, and the calculation formula is: In the formula, Cs D It represents the data transmission rate between the UAV and the low-orbit satellite, Sz represents the transmission rate of the current data packet, and CL represents the total amount of data transmitted per unit time; The method for acquiring the first bandwidth utilization includes: The data transmission rate of the UAV is obtained, and the first total bandwidth allocated to the UAV by the low-orbit satellite is obtained, and the data transmission rate and the allocated first total bandwidth are calculated by formula to obtain the first bandwidth utilization rate; the calculation formula is: Where Ly1 represents the first bandwidth utilization of the low-orbit link, and Fz represents the first total bandwidth allocated by the low-orbit satellite to the UAV.

4. The method for transmitting data of unmanned aerial vehicle based on high-orbit and low-orbit satellite fusion communication according to claim 3 is characterized in that: The method for obtaining the basic connection coefficient based on the communication characteristic data includes: Step b1: obtaining low-orbit transmission characteristic data of the low-orbit link, formulating the low-orbit transmission characteristic data, and obtaining a low-orbit connection evaluation coefficient; Step b2: Formulate the high-orbit transmission characteristic data to obtain the high-orbit connection evaluation coefficient; Step b3: Fuse the low-orbit connection evaluation coefficient and the high-orbit connection evaluation coefficient to obtain the basic connection coefficient Gx.

5. The method for transmitting data of unmanned aerial vehicle based on high-orbit and low-orbit satellite fusion communication according to claim 4 is characterized in that: The method of analyzing the basic connection coefficient and determining whether to generate a link adjustment instruction includes: Preset a connection coefficient threshold, the connection coefficient threshold includes Lj1 and Lj2, where Lj1>Lj2; compare the basic connection coefficient with the preset connection coefficient threshold; If Gx>Lj1 or Gx<Lj2, a link adjustment command is generated, indicating that the UAV has unstable data transmission in the basic communication link; If Lj1≥Gx≥Lj2, no link adjustment command is generated, indicating that the data transmission of the UAV in the basic communication link is stable.

6. The method for transmitting data of unmanned aerial vehicle based on high-orbit and low-orbit satellite fusion communication according to claim 5 is characterized in that: The dynamic fusion training data includes the relationship between each bandwidth value and the first link coefficient difference and the relationship between each transmission rate and the second link coefficient difference; The method for obtaining the relationship between each bandwidth value and the first link coefficient difference includes: Step c1: in the experimental scenario, obtain the real-time communication link coefficient for measuring the data transmission of the UAV under the hth bandwidth value; Step c2: extracting the standard link coefficient pre-stored in the system database and used to measure the data transmission of the UAV under the set standard bandwidth value; Step c3: Calculate the difference between the real-time communication link coefficient and the standard link coefficient for measuring the data transmission of the UAV under the h-th bandwidth value, and use the difference between the real-time communication link coefficient and the standard link coefficient for measuring the data transmission of the UAV under the h-th bandwidth value as the first link coefficient difference; Step c4: compare the first link coefficient difference with a preset first link coefficient difference threshold. If the first link coefficient difference is less than the preset first link coefficient difference threshold, set h=h+1 and return to step c1. If the first link coefficient difference is greater than or equal to the preset first link coefficient difference threshold, record the relationship between the hth bandwidth value and the first link coefficient difference, set h=h+1 again and return to step c1. Step c5: Repeat the above steps c1 to c4 until h is equal to the set bandwidth value H, end the loop, obtain the relationship between each bandwidth value and the first link coefficient difference, and use the relationship between each bandwidth value and the first link coefficient difference as dynamic fusion training data, where h and H are integers greater than zero.

7. The method for transmitting data via unmanned aerial vehicle based on high-orbit and low-orbit satellite fusion communication according to claim 6 is characterized in that: The method for obtaining the relationship between each transmission rate and the second link coefficient difference includes: Step d1: in the experimental scenario, obtain the real-time communication link coefficient for measuring the data transmission of the UAV at the kth transmission rate; Step d2: extracting the standard link coefficient pre-stored in the system data and used to measure the data transmission of the UAV at the set standard transmission rate; Step d3: Calculate the difference between the real-time communication link coefficient and the standard link coefficient for measuring the data transmission of the UAV at the kth transmission rate, and use the difference between the real-time communication link coefficient and the standard link coefficient of the UAV at the kth transmission rate as the second link coefficient difference; Step d4: compare the second link coefficient difference with a preset second link coefficient difference threshold value. If the second link coefficient difference is less than the preset second link coefficient difference threshold value, set k=k+1 and return to step d1; if the second link coefficient difference is greater than or equal to the preset second link coefficient difference threshold value, record the relationship between the kth transmission rate and the second link coefficient difference, set k=k+1 again and return to step d1; Step d5: Repeat the above steps d1 to d4 until k is equal to the set transmission rate K, end the loop, obtain the relationship between each transmission rate and the second link coefficient difference, and use the relationship between each transmission rate and the second link coefficient difference as dynamic fusion training data, k and K are integers greater than zero.

8. The method for transmitting data of unmanned aerial vehicle based on high-orbit and low-orbit satellite fusion communication according to claim 7 is characterized in that: The method for obtaining the real-time communication link coefficient for measuring the data transmission of the UAV under the h-th bandwidth value includes: When the network bandwidth is the hth bandwidth value, the signal strength, signal-to-noise ratio and packet loss rate are retrieved from the system database; The signal strength, signal-to-noise ratio, and packet loss rate are formally calculated to obtain the real-time communication link coefficient, which is calculated as follows: SLx=ln(WXh×SNR)-exp(SDL); Where SLx is the real-time communication link coefficient, WXh is the signal strength of the drone connection, SNR is the signal-to-noise ratio, SDL is the packet loss rate, ln(·) is the logarithmic function with e as the base, and exp(·) is the exponential function.

9. The method for transmitting data of unmanned aerial vehicle based on high-orbit and low-orbit satellite fusion communication according to claim 8 is characterized in that: The training logic of the fusion prediction model is as follows: Extracting the relationship between each bandwidth value and the first link coefficient difference and the relationship between each transmission rate and the second link coefficient difference in the dynamic fusion training data; Divide the relationship between each bandwidth value and the first link coefficient difference and the relationship between each transmission rate and the second link coefficient difference into a fusion coefficient training set and a fusion coefficient test set; Constructing a neural network model, inputting the relationship between each bandwidth value and the first link coefficient difference and the relationship between each transmission rate and the second link coefficient difference in the fusion coefficient training set into the fusion prediction model for training, and obtaining a trained neural network model; using the fusion coefficient test set to verify the trained neural network model, and outputting a trained neural network model greater than or equal to a preset test accuracy as a fusion prediction model; the fusion prediction model is an RNN neural network, a DNN neural network or a CNN neural network model; The method of analyzing the fusion coefficient output by the fusion prediction model and determining the main link instruction includes: Preset fusion coefficient thresholds Rh1 and Rh2, Rh1<Rh2, and compare the fusion coefficient with the fusion coefficient threshold; If the fusion coefficient is greater than the fusion coefficient threshold Rh2, a high-orbit main link instruction is generated; If the fusion coefficient is less than or equal to the fusion coefficient threshold Rh2, and greater than or equal to the fusion coefficient threshold Rh1, a parallel transmission instruction is generated; If the fusion coefficient is less than the fusion coefficient threshold Rh1, a low-orbit main link instruction is generated; Among them, high-orbit main link instructions > parallel transmission instructions > low-orbit main link instructions.

10. An unmanned aerial vehicle data transmission system based on high- and low-orbit satellite fusion communication, used to implement the unmanned aerial vehicle data transmission method based on high- and low-orbit satellite fusion communication according to any one of claims 1 to 9, characterized in that: include: A type classification module is used to obtain the mission data type of the UAV in the basic communication link, classify the mission data type, and obtain data classification results, wherein the data classification results include low-orbit real-time data and high-orbit non-real-time data; A judgment module, used to obtain communication characteristic data of a basic communication link, generate a basic connection coefficient based on the communication characteristic data, analyze the basic connection coefficient, and judge whether to generate a link adjustment instruction; A data acquisition module, used to receive link adjustment instructions and acquire dynamic fusion training data of the UAV based on the basic connection coefficient; A model training module is used to train a fusion prediction model for feeding back the fusion coefficient of the UAV based on dynamic fusion training data; The analysis module is used to analyze the fusion coefficient output by the fusion prediction model and determine the main link instruction, which includes the high-orbit main link instruction, the parallel transmission instruction and the low-orbit main link instruction.

Images