A method and device for through-ground communication of unmanned aerial vehicles (UAVs)

Abstract

This application provides a method and apparatus for UAV ground-penetrating communication, relating to the field of communication technology. It can acquire communication environment information of the area where the underground communication node is located when the UAV meets the communication mode decision conditions; process the communication environment information using a geological state prediction model to obtain the ground penetration difficulty level; input the ground penetration difficulty level and the communication link parameters of the wireless communication system into the communication decision model to obtain the communication mode decision result output by the communication decision model; and control the UAV to communicate with the underground communication node based on the communication mode decision result. This method can meet the needs of complex and ever-changing communication scenarios and reduce the risk of communication interruption.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to a method and apparatus for unmanned aerial vehicle (UAV) ground-penetrating communication. Background Technology

[0002] Currently, through-ground communication technology, as a core means of connecting the surface and underground spaces, has significant application value in disaster relief, underground resource exploration, and urban underground pipeline monitoring. However, the fixed frequency bands and static parameter configurations of existing through-ground communication systems cannot adapt to dynamic geological changes, often leading to communication interruptions. Summary of the Invention

[0003] The technical problem to be solved by this application is to provide a method and apparatus for UAV ground-penetrating communication, which can reduce the risk of communication interruption. The specific solution is as follows:

[0004] A method for unmanned aerial vehicle (UAV) through-ground communication, applied to a wireless communication system, the wireless communication system further including the UAV and an underground communication node, the method comprising:

[0005] In response to a communication command, the communication status information of the wireless communication system is obtained;

[0006] Based on the communication status information, determine whether the UAV in the wireless communication system meets the communication mode decision conditions;

[0007] If the UAV is detected to meet the communication mode decision conditions, the communication environment information of the area where the underground communication node is located is obtained;

[0008] The communication environment information is processed using a pre-trained geological condition prediction model to obtain the stratum penetration difficulty level corresponding to the underground communication node;

[0009] The stratum penetration difficulty level and the communication link parameters of the wireless communication system are input into a pre-trained communication decision model to obtain the communication mode decision result output by the communication decision model.

[0010] Based on the decision result of the communication mode, the UAV is controlled to communicate with the underground communication node.

[0011] Optionally, the method described above includes obtaining the communication status information of the wireless communication system, including:

[0012] The system acquires at least one communication status information, namely, the signal attenuation rate of the underground communication node in the wireless communication system and the intensity of environmental electromagnetic interference.

[0013] Optionally, in the above method, determining whether the UAV in the wireless communication system meets the communication mode decision conditions based on the communication status information includes:

[0014] If the signal attenuation rate is detected to be greater than the attenuation rate threshold and the environmental electromagnetic interference intensity is less than the preset intensity threshold, it is determined that the UAV in the wireless communication system meets the communication mode decision conditions.

[0015] If the signal attenuation rate is not greater than the attenuation rate threshold or the environmental electromagnetic interference intensity is not less than the preset intensity threshold, it is determined that the UAV in the wireless communication system does not meet the communication mode decision conditions.

[0016] Optionally, the training process of the communication decision model in the above method includes:

[0017] A training sample set and an initial communication decision model are obtained. The training dataset includes multiple training samples and sample labels for each training sample. Each training sample includes historical stratigraphic penetration difficulty level and historical communication link parameters. The sample labels include identifiers used to indicate communication modes.

[0018] The initial decision model is trained based on the training sample set.

[0019] Optionally, in the above method, controlling the UAV to communicate with the underground communication node based on the communication mode decision result includes:

[0020] When the decision result indicates the selection of a low-frequency electromagnetic wave mode, basic beamforming parameters are obtained. If the distribution of the rock dielectric constant determines that the area where the underground communication node is located meets the parameter adjustment conditions, at least one of the amplitude weight and phase offset in the basic beamforming parameters is adjusted to obtain target beamforming parameters. The UAV is then controlled to communicate with the underground communication node based on the target beamforming parameters.

[0021] When the decision result indicates that the magnetic induction communication mode is selected, the drone is controlled to reduce its flight altitude to within a preset threshold, and the multi-input multi-output coil array of the magnetic induction unit is activated to enhance the near-field coupling efficiency, so that the drone can communicate with the underground communication node.

[0022] When the decision result indicates that the acoustic communication mode is selected, the carrier frequency and modulation rate of the acoustic transmitter are adjusted so that the UAV can communicate with the underground communication node;

[0023] When the decision result indicates the selection of a hybrid communication mode, communication time slots for different modes are allocated according to the priority of the data to be transmitted, so that the UAV can communicate with the underground communication node.

[0024] A drone-to-ground communication device is applied to a wireless communication system, the wireless communication system further including a drone and an underground communication node, the device comprising:

[0025] The first acquisition unit is used to acquire the communication status information of the wireless communication system in response to a communication command;

[0026] A determining unit is configured to determine, based on the communication status information, whether the UAV in the wireless communication system meets the communication mode decision conditions;

[0027] The second acquisition unit is used to acquire communication environment information of the area where the underground communication node is located when the UAV is detected to meet the communication mode decision conditions.

[0028] The processing unit is used to process the communication environment information using a pre-trained geological state prediction model to obtain the stratum penetration difficulty level corresponding to the underground communication node.

[0029] The input unit is used to input the stratum penetration difficulty level and the communication link parameters of the wireless communication system into a pre-trained communication decision model to obtain the communication mode decision result output by the communication decision model.

[0030] The control unit is used to control the UAV to communicate with the underground communication node based on the communication mode decision result.

[0031] Optionally, the first acquisition unit of the aforementioned apparatus includes:

[0032] The acquisition subunit is used to acquire at least one communication status information, namely the signal attenuation rate of the underground communication node in the wireless communication system and the intensity of environmental electromagnetic interference.

[0033] Optionally, the determining unit in the aforementioned apparatus includes:

[0034] The first determining subunit is used to determine that the UAV in the wireless communication system meets the communication mode decision conditions when the signal attenuation rate is detected to be greater than the attenuation rate threshold and the environmental electromagnetic interference intensity is less than the preset intensity threshold.

[0035] The second determining subunit is used to determine that the UAV in the wireless communication system does not meet the communication mode decision conditions when the signal attenuation rate is not greater than the attenuation rate threshold or the environmental electromagnetic interference intensity is not less than a preset intensity threshold.

[0036] Optionally, the input unit in the aforementioned apparatus includes:

[0037] The acquisition sub-unit is used to acquire a training sample set and an initial communication decision model. The training dataset includes multiple training samples and sample labels for each training sample. Each training sample includes historical stratigraphic penetration difficulty level and historical communication link parameters. The sample label includes an identifier for indicating the communication mode.

[0038] A training unit is used to train the initial decision model based on the training sample set.

[0039] Optionally, the control unit in the aforementioned device includes:

[0040] The first control subunit is configured to, when the decision result indicates the selection of a low-frequency electromagnetic wave mode, acquire basic beamforming parameters, and, when the area where the underground communication node is located is determined to meet the parameter adjustment conditions based on the rock dielectric constant distribution, adjust at least one of the amplitude weight and phase offset in the basic beamforming parameters to obtain target beamforming parameters; and control the UAV to communicate with the underground communication node according to the target beamforming parameters.

[0041] The second control subunit is used to control the UAV to reduce its flight altitude to within a preset threshold when the decision result indicates that the magnetic induction communication mode is selected, and to activate the multi-input multi-output coil array of the magnetic induction unit to enhance the near-field coupling efficiency, so that the UAV can communicate with the underground communication node.

[0042] The third control subunit is used to adjust the carrier frequency and modulation rate of the acoustic transmitter when the decision result indicates that the acoustic communication mode is selected, so that the UAV can communicate with the underground communication node.

[0043] The fourth control subunit is used to allocate communication time slots of different modes according to the priority of the data to be transmitted when the decision result indicates the selection of the hybrid communication mode, so that the UAV can communicate with the underground communication node.

[0044] The UAV ground-penetrating communication method and apparatus provided in this application can be applied to a wireless communication system, which further includes a UAV and an underground communication node. The method includes: in response to a communication command, acquiring communication status information of the wireless communication system; determining whether the UAV in the wireless communication system meets communication mode decision conditions based on the communication status information; if the UAV meets the communication mode decision conditions, acquiring communication environment information of the area where the underground communication node is located; processing the communication environment information using a pre-trained geological state prediction model to obtain the stratum penetration difficulty level corresponding to the underground communication node; inputting the stratum penetration difficulty level and the communication link parameters of the wireless communication system into a pre-trained communication decision model to obtain the communication mode decision result output by the communication decision model; and controlling the UAV to communicate with the underground communication node based on the communication mode decision result. Applying the method provided in this application can meet the needs of complex and ever-changing communication scenarios and reduce the risk of communication interruption. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0046] Figure 1 A flowchart of a method for UAV through-ground communication provided in this application;

[0047] Figure 2 A flowchart illustrating the training process of a communication decision model provided in this application;

[0048] Figure 3 A schematic diagram of the structure of a wireless communication system provided in this application;

[0049] Figure 4 This is a schematic diagram of the structure of a ground-penetrating communication device for unmanned aerial vehicles (UAVs) provided in this application. Detailed Implementation

[0050] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0051] In this application, the terms "comprising," "including," or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0052] This invention provides a method for unmanned aerial vehicle (UAV) through-ground communication, applied to a wireless communication system. The wireless communication system further includes the UAV and an underground communication node. The flowchart of the method is shown below. Figure 1 As shown, it specifically includes:

[0053] S101: In response to a communication command, obtain the communication status information of the wireless communication system.

[0054] In this embodiment, the communication command refers to the control signal that triggers the UAV to start or adjust the communication connection with the underground communication node. Its sources include, but are not limited to, active commands sent by the ground control center, self-triggered commands generated by the UAV's built-in controller based on preset rules, or request commands initiated by the underground communication node when it detects data to be transmitted.

[0055] Communication status information is a set of quantitative parameters characterizing the current communication quality and link reliability of a wireless communication system. These parameters include, but are not limited to, environmental electromagnetic interference intensity and signal attenuation rate of underground communication nodes. The signal attenuation rate of an underground communication node refers to the proportion of energy loss from the UAV transmitter to the receiver, expressed in dB, calculated as: Attenuation rate = Transmit power - Received power. Environmental electromagnetic interference intensity refers to the energy density of non-target signals within the communication frequency band, such as industrial electromagnetic noise and natural electromagnetic interference.

[0056] In this embodiment, in response to communication commands actively sent by the ground control center, the UAV's built-in controller, or the underground communication node, the UAV or the underground communication node's built-in sensor module collects and processes the communication status information of the wireless communication system in real time. The communication status information can be structured data after analog-to-digital conversion and filtering calibration.

[0057] Optionally, underground communication nodes are fixed or semi-fixed communication devices deployed in underground environments such as mines, tunnels, and deep geological monitoring points. They can establish two-way communication links with ground / air drones to achieve data transmission, such as transmitting environmental parameters and equipment status information collected by underground sensors, or receiving control commands forwarded by drones.

[0058] S102: Determine whether the UAV in the wireless communication system meets the communication mode decision conditions based on the communication status information.

[0059] In this embodiment, the drone is an airborne communication terminal in a wireless communication system.

[0060] Optionally, the communication mode decision condition is a pre-set threshold condition that triggers the communication mode selection logic. Specifically, when the signal attenuation rate is greater than a preset attenuation rate threshold and the environmental electromagnetic interference intensity is less than a preset intensity threshold, the UAV is determined to meet the communication mode decision condition, that is, the current communication environment allows for further optimization of the communication mode; otherwise, when the signal attenuation rate is ≤ attenuation rate threshold or the environmental electromagnetic interference intensity is ≥ intensity threshold, the UAV is determined not to meet the decision condition and must maintain the default communication mode or terminate communication.

[0061] Optionally, a logic judgment unit integrated into the UAV controller or underground communication node processor can compare parameters such as signal attenuation rate and environmental electromagnetic interference intensity in the communication status information with corresponding preset thresholds in real time. If both the signal attenuation rate and the environmental electromagnetic interference intensity are satisfied, a judgment result indicating that the communication mode decision condition is met is output; otherwise, a judgment result indicating that the condition is not met is output. This ensures that subsequent decision-making processes are initiated only when the communication environment allows for optimization, thereby improving the efficiency and reliability of the method.

[0062] S103: If the UAV is detected to meet the communication mode decision conditions, obtain the communication environment information of the area where the underground communication node is located.

[0063] In this embodiment, the area where the underground communication node is located refers to the geological area within a preset range around the deployment location of the underground communication node. Its geological characteristics directly affect the communication penetration capability between the UAV and the underground communication node. The communication environment information is a set of quantitative parameters reflecting the geological characteristics of the area, including but not limited to: the resistivity of the stratum, which characterizes the ability of the stratum to impede current; the dielectric constant, which characterizes the response characteristics of the stratum to the electric field; and parameters that affect the penetration loss of electromagnetic waves / sound waves, such as water content and rock density.

[0064] Optionally, information about the communication environment can be collected by sensors such as resistivity logging tools, dielectric constant measurement modules, or ground-penetrating radar (GPR) carried by UAVs deployed at underground communication nodes.

[0065] In this embodiment, a resistivity logging tool transmits a current signal to the formation and measures the voltage difference, and calculates the resistivity based on Ohm's law; a dielectric constant measurement module transmits a high-frequency electromagnetic wave and measures the phase difference of the reflected wave, and calculates the dielectric constant based on Maxwell's equations; a humidity sensor measures the formation water content, and a densitometer measures the rock density.

[0066] Optionally, if a ground-penetrating radar mounted on a drone is used, it can transmit high-frequency electromagnetic pulses into the ground through an antenna, receive the echo signals reflected from the stratum interface, record the time delay and amplitude attenuation, and combine the electromagnetic wave propagation speed to calculate parameters such as stratum thickness and resistivity as communication environment information.

[0067] In this embodiment, the raw communication environment information collected by the sensor is filtered and calibrated to convert it into structured communication environment information.

[0068] Optionally, after determining that the UAV meets the communication mode decision conditions, a multi-sensor collaborative acquisition process is initiated to obtain communication environment information of the area where the underground communication node is located. Specifically, this includes: generating a three-dimensional distribution map of rock dielectric constant by scanning with an airborne ground-penetrating radar array, measuring the polarization angle of the stratum stress field in real time via a strain sensor network, and calling up stratum deformation time series data from a historical database.

[0069] In one embodiment provided in this application, based on the above-described scheme, optionally, obtaining the communication status information of the wireless communication system includes:

[0070] Acquire at least one communication status information from underground communication nodes in a wireless communication system, namely, the signal attenuation rate and the intensity of environmental electromagnetic interference.

[0071] S104: The communication environment information is processed using a pre-trained geological condition prediction model to obtain the stratum penetration difficulty level corresponding to the underground communication node.

[0072] In this embodiment, a geological state prediction model trained offline is used to perform real-time reasoning on the collected communication environment information and output the stratum penetration difficulty level of the location of the underground communication node. The communication environment information includes the spatial distribution tensor of rock dielectric constant, the polarization angle matrix of stress field and the time series sequence of historical stratum deformation. The model extracts spatial features by fusing three-dimensional convolutional layers and captures temporal dependencies by long short-term memory networks, and finally maps them into discrete stratum penetration difficulty levels through fully connected layers.

[0073] Optionally, a pre-trained geological condition prediction model refers to a machine learning model trained based on historical stratigraphic parameters and communication penetration loss data.

[0074] Optionally, the geological state prediction model takes communication environment information as input and outputs the difficulty level of strata penetration. The geological state prediction model includes, but is not limited to, convolutional neural networks, random forests, or support vector machines. For example, the geological state prediction model uses a three-dimensional convolutional layer (kernel size 5×5×5) to process the spatial distribution tensor of dielectric constant to extract geological structural features, then fuses the temporal variation trend of stress field polarization angle through a long short-term memory network (LSTM with 128 units), and finally maps it to a discrete difficulty level by a fully connected layer.

[0075] Optionally, the training data for the geological condition prediction model can be derived from: 1. Stratigraphic parameters recorded in historical ground-penetrating communication experiments collected by geological sensors or ground-penetrating radar; 2. Actual penetration loss values ​​in the corresponding experimental scenarios, such as measured values ​​of signal attenuation rate.

[0076] In this embodiment, the model is trained by minimizing the mean square error (MSE) between the predicted penetration loss and the actual penetration loss, and finally forms a mapping relationship that can quantify the difficulty of formation penetration.

[0077] In this embodiment, the formation penetration difficulty level refers to the discretized evaluation result based on the penetration loss prediction value output by the model. Specifically, it is represented by a preset level label, for example, Level 1 - low difficulty, corresponding to penetration loss ≤ 50dB; Level 2 - medium difficulty, corresponding to 50dB < penetration loss ≤ 80dB; Level 3 - high difficulty, corresponding to penetration loss > 80dB. Alternatively, the formation penetration difficulty level can also be a continuous quantitative score, such as 0-10 points, with higher scores indicating greater penetration difficulty.

[0078] Optionally, the stratum penetration difficulty level is used to intuitively characterize how easy it is for the UAV signal to penetrate the stratum, and the communication mode and power configuration can be dynamically selected based on the stratum penetration difficulty level.

[0079] In this embodiment, the acquired communication environment information is standardized to eliminate the influence of dimensional differences on model predictions; for example, the formation resistivity ρ = 150 Ω·m, dielectric constant ε = 8, water content ω = 12%, and rock density ρ_r = 2.5 g / cm³ are standardized. 3 By using Z-score normalization, each parameter is mapped to the [0, 1] interval to complete the standardization process.

[0080] In this embodiment, the predicted penetration loss value output by the model can be mapped to the corresponding formation penetration difficulty level according to a preset penetration loss-difficulty level comparison table.

[0081] S105: Input the stratum penetration difficulty level and the communication link parameters of the wireless communication system into the pre-trained communication decision model to obtain the communication mode decision result output by the communication decision model.

[0082] In this embodiment, the communication link parameters include a set of quantitative indicators reflecting the current communication link quality in the wireless communication system, specifically including but not limited to: quality factor, polarization mismatch loss, link bandwidth, signal-to-noise ratio, bit error rate, delay, and available power margin, etc. These parameters are acquired through physical layer monitoring modules such as spectrum analyzers and bit error rate testers on UAVs or underground communication nodes and uploaded to the decision unit. The quality factor reflects the resonant efficiency of the magnetic induction link; the polarization mismatch loss characterizes the degree of matching between the electromagnetic wave polarization direction and the receiver.

[0083] Optionally, the pre-trained communication decision model refers to an intelligent decision model trained based on the mapping relationship between historical geological difficulty level, link parameters and optimal communication mode.

[0084] Optionally, the communication decision model takes the ground penetration difficulty level and multi-dimensional communication link parameters as input and outputs the communication mode decision result. The model type includes, but is not limited to, deep neural network (DNN), decision tree, or reinforcement learning (RL) models. Its training data comes from: the ground penetration difficulty level recorded in historical ground penetration communication experiments; the real-time communication link parameters in the corresponding experimental scenario; and the optimal communication mode marked by manual annotation.

[0085] The communication mode decision result is the optimal communication mode identifier output by the model, which is adapted to the current geological and link status. This identifier can correspond to a single mode such as low-frequency electromagnetic wave mode, magnetic induction communication mode or sound wave, or a mixed mode such as low-frequency electromagnetic wave + sound wave.

[0086] In this embodiment, the difficulty level of ground penetration and the communication link parameters of the wireless communication system are input into a pre-trained communication decision model. The communication decision model adopts a multi-layer fully connected neural network with 8 neurons in the input layer, 128 neurons in the hidden layer (with ReLU activation function), and 4 neurons in the output layer corresponding to the probability values ​​of 4 communication modes. The communication decision model calculates the fit probability of each communication mode through linear transformation and nonlinear activation of the weight matrix and bias parameters. For example, the probability of low-frequency electromagnetic wave mode is 0.7, magnetic induction mode is 0.2, sound wave mode is 0.05, and mixed mode is 0.05.

[0087] Optionally, based on the probability distribution output by the model and combined with preset decision rules, a communication mode decision result can be generated. If a single-mode selection rule is used, the mode with the highest probability is selected; if a multi-mode combination rule is used, the combination of modes whose sum of probabilities exceeds a threshold is selected; if priority sorting is required, a list of modes is output in descending order of probability.

[0088] S106: Control the UAV to communicate with the underground communication node based on the communication mode decision result.

[0089] In this embodiment, the communication mode decision result is the optimal communication mode identifier output by the communication decision model, corresponding to low-frequency electromagnetic waves, magnetic induction, sound waves, or a hybrid mode, which is used to guide the communication control between the UAV and the underground node.

[0090] In this embodiment, based on the communication mode decision result, a corresponding physical layer control command is generated, and the UAV is controlled to communicate with the underground communication node according to the command. The physical layer control command can be a low-level signal that directly operates the communication hardware, including mode identifier, hardware configuration parameters, and time slot allocation information, used to drive the communication module to perform specific actions.

[0091] In this embodiment, the control flow is described by mode through the collaborative interface (such as SPI, I2C, GPIO) between the UAV flight control system and the communication hardware:

[0092] In low-frequency electromagnetic wave mode, control commands are sent to the ADAR1000 chip via the SPI bus to drive the FPGA to generate antenna array control signals; the numerically controlled attenuator adjusts the signal amplitude (in 0.1dB steps) based on the amplitude weight in the target beamforming parameters in the command, and the 6-phase shifter adjusts the signal phase based on the phase offset in the target beamforming parameters, ultimately focusing the electromagnetic beam onto the underground node coordinates, achieving highly directional signal transmission.

[0093] In magnetic induction communication mode, the motor is driven through the PWM interface, and the drone is controlled to descend to a preset height above the ground based on the PID algorithm. At the same time, the relay matrix (4×4 planar spiral coil) of the MIMO coil array is switched through the GPIO interface to activate the resonant circuit, and the near-field coupling efficiency is optimized through the impedance matching network.

[0094] In acoustic communication mode, the AD9834 DDS chip is configured via the I2C bus, and the carrier frequency and modulation rate are set according to the carrier frequency parameters in the instruction. After the output signal is amplified by the high voltage amplifier, it drives the PZT-8 piezoelectric transducer to excite mechanical waves that resonate with the rock strata, thereby achieving effective penetration of the acoustic signal.

[0095] In hybrid mode, the hardware timer is synchronized based on the IEEE 1588 protocol, and the communication time slots are allocated according to the time slot allocation table in the instruction. For example, the low-frequency channel occupies 70% of the time slots to transmit critical data, while the magnetic induction and acoustic channels share the remaining 30% of the time slots in a 2:1 ratio. The time slot switching is controlled by the time slot allocation table stored in the dual-port RAM in a cyclic call, ensuring the timing synchronization of multi-mode communication.

[0096] In one embodiment provided in this application, based on the above-described solution, optionally, determining whether the UAV in the wireless communication system meets the communication mode decision conditions based on the communication status information includes:

[0097] If the signal attenuation rate is detected to be greater than the attenuation rate threshold and the environmental electromagnetic interference intensity is less than the preset intensity threshold, it is determined that the UAV in the wireless communication system meets the communication mode decision conditions.

[0098] If the signal attenuation rate is not greater than the attenuation rate threshold or the environmental electromagnetic interference intensity is not less than the preset intensity threshold, it is determined that the UAV in the wireless communication system does not meet the communication mode decision conditions.

[0099] Optionally, the attenuation rate threshold is a preset critical value for signal attenuation rate, defined as the maximum non-interference attenuation level acceptable to the system. When this value is exceeded, communication mode adjustment needs to be triggered to compensate for the attenuation. The intensity threshold is a preset critical value for electromagnetic interference intensity, defined as the maximum interference level that the system can tolerate. When this value is exceeded, interference will dominate signal distortion, and adjusting the communication mode will not effectively improve performance.

[0100] The method provided in the embodiments of this application can meet the needs of complex and ever-changing communication scenarios and reduce the risk of communication interruption.

[0101] In one embodiment provided in this application, based on the above-described scheme, optionally, the training process of the communication decision model is as follows: Figure 2 As shown, it includes:

[0102] S201: Obtain a training sample set and an initial communication decision model. The training dataset includes multiple training samples and sample labels for each training sample. Each training sample includes historical stratigraphic penetration difficulty level and historical communication link parameters. The sample label includes an identifier used to indicate the communication mode.

[0103] S202: Train the initial decision model based on the training sample set.

[0104] In this embodiment, the training sample set consists of multiple training samples and their sample labels. The training samples include historical stratigraphic penetration difficulty levels and historical communication link parameters. The sample labels are communication mode identifiers, corresponding to low-frequency electromagnetic waves, magnetic induction, sound waves, and mixed modes, respectively. The labeling is based on the optimal communication performance mode in historical experiments, and the optimal communication performance mode is evaluated by quantifiable indicators such as transmission rate and bit error rate.

[0105] In this embodiment, the initial communication decision model adopts a three-layer fully connected network architecture, specifically including an input layer, two hidden layers and an output layer. The initial weights are initialized using a normal distribution, and the bias is initialized to 0.

[0106] Optionally, based on historical ground-penetrating communication experiments (covering geological scenarios such as clay layers and iron ore layers), the physical layer monitoring module collects the stratum penetration difficulty level and communication link parameters, and records the optimal communication mode in the experiment as sample labels. Outliers are eliminated through statistical testing, and an oversampling algorithm is used to balance the sample distribution for minority class samples, ensuring that the number of samples in each class is balanced (accounting for 20-30%). Parameters such as stratum penetration difficulty level, quality factor, and polarization loss are uniformly mapped to the [0,1] interval, consistent with the preprocessing rules in the inference stage. Training parameters such as batch size, learning rate, and maximum number of iterations are set, and the initial model is loaded. The normalized features are input into the model, and after calculation by a fully connected layer and activation function, a four-dimensional probability vector is output. The cross-entropy loss function is used to measure the error between the predicted probability and the true label, and the gradient is calculated by the optimizer to update the model weights, gradually optimizing the model performance. The model accuracy is evaluated using a validation set at regular intervals. If there is no significant improvement for several consecutive cycles, training is terminated early to avoid overfitting. After training, the model is quantized and compressed, and the model computation graph is converted into a hardware description language and deployed to a dedicated chip (such as an FPGA). At the same time, the trained weights are stored in non-volatile memory to achieve inference latency that meets real-time requirements.

[0107] In one embodiment provided in this application, based on the above-described scheme, optionally, controlling the UAV to communicate with the underground communication node according to the communication mode decision result includes:

[0108] When the decision result indicates the selection of a low-frequency electromagnetic wave mode, basic beamforming parameters are obtained. If the distribution of the rock dielectric constant determines that the area where the underground communication node is located meets the parameter adjustment conditions, at least one of the amplitude weight and phase offset in the basic beamforming parameters is adjusted to obtain target beamforming parameters. The UAV is then controlled to communicate with the underground communication node based on the target beamforming parameters.

[0109] When the decision result indicates that the magnetic induction communication mode is selected, the drone is controlled to reduce its flight altitude to within a preset threshold, and the multi-input multi-output coil array of the magnetic induction unit is activated to enhance the near-field coupling efficiency, so that the drone can communicate with the underground communication node.

[0110] When the decision result indicates that the acoustic communication mode is selected, the carrier frequency and modulation rate of the acoustic transmitter are adjusted so that the UAV can communicate with the underground communication node;

[0111] When the decision result indicates the selection of a hybrid communication mode, communication time slots for different modes are allocated according to the priority of the data to be transmitted, so that the UAV can communicate with the underground communication node.

[0112] In this embodiment, the control method selects different communication modes based on the decision results and executes the corresponding control logic, specifically including the following scenarios:

[0113] Optionally, when the decision result indicates that a low-frequency electromagnetic wave mode should be selected, the following steps are performed:

[0114] The system acquires basic beamforming parameters, which are preset initial beamforming configurations used to guide directional electromagnetic wave transmission. It detects the rock dielectric constant distribution in the area where the underground communication node is located to determine if parameter adjustment conditions are met. These conditions can include uneven dielectric constant distribution leading to signal attenuation exceeding a preset level. If the adjustment conditions are met, at least one of the amplitude weights and phase offsets in the basic beamforming parameters is dynamically adjusted to generate target beamforming parameters. The amplitude weights control the power distribution of each array element, and the phase offset controls the phase difference of the signals from each array element. Based on the target beamforming parameters, the system controls the UAV's electromagnetic wave transmission array to adjust the signal transmission direction and energy distribution, achieving directional communication with the underground communication node.

[0115] Optionally, when the decision result indicates that the magnetic induction communication mode should be selected, the following steps are performed:

[0116] The drone is controlled to reduce its flight altitude to within a preset threshold, which is determined based on the effective distance of magnetic induction near-field coupling. The multi-input multi-output (MIMO) coil array of the magnetic induction unit on the drone is activated. The MIMO coil array consists of multiple orthogonally placed transmit / receive coils. The near-field coupling efficiency is enhanced through the coordinated operation of the MIMO coil array, thereby improving the transmission strength and anti-interference capability of the magnetic induction signal, and thus realizing magnetic induction communication between the drone and the underground communication node.

[0117] Optionally, when the decision result indicates that the acoustic communication mode should be selected, the following steps are performed:

[0118] Adjusting the carrier frequency of the drone's acoustic transmitter, such as increasing it from 20kHz to 50kHz, and adjusting the modulation rate of the drone's acoustic transmitter, such as adjusting it from 1kbps to 2kbps, allows the frequency characteristics of the acoustic signal to match the acoustic impedance of underground media such as soil and rock strata, while also adapting to the demodulation capabilities of underground communication nodes, ultimately enabling acoustic communication between the drone and underground communication nodes.

[0119] Optionally, when the decision indicates that a hybrid communication mode should be selected, the following steps are performed:

[0120] Based on the priority of the data to be transmitted, communication time slots corresponding to different priority modes are allocated. Multi-mode collaborative communication is achieved through time-division multiplexing, balancing transmission efficiency and reliability. Different data correspond to different priorities; for example, real-time control commands have high priority, while status monitoring data has low priority. Different priority data also correspond to different time slots: high-priority data is allocated to low-frequency electromagnetic wave mode time slots, and low-priority data is allocated to acoustic wave mode time slots.

[0121] In this embodiment, differentiated control logic is designed for the physical characteristics of different modes to ensure that each mode operates under optimal conditions; the parameter adjustment mechanism of the low-frequency electromagnetic wave mode can dynamically optimize beamforming according to the distribution of rock dielectric constant to compensate for signal attenuation caused by complex geological environment; the time slot allocation strategy of the hybrid mode prioritizes the transmission of high-priority data while using low-priority data to fill idle time slots; the MIMO coil array of the magnetic induction mode is combined with flight altitude control to improve near-field coupling efficiency.

[0122] See Figure 3 This is a schematic diagram of a ground-penetrating communication device for unmanned aerial vehicles (UAVs) provided in an embodiment of this application. The device is applied to a wireless communication system, which further includes a UAV and an underground communication node.

[0123] The first acquisition unit 301 is used to acquire the communication status information of the wireless communication system in response to a communication command;

[0124] Determining unit 302 is used to determine whether the UAV in the wireless communication system meets the communication mode decision conditions based on the communication status information;

[0125] The second acquisition unit 303 is used to acquire communication environment information of the area where the underground communication node is located when the UAV is detected to meet the communication mode decision conditions.

[0126] Processing unit 304 is used to process the communication environment information using a pre-trained geological state prediction model to obtain the stratum penetration difficulty level corresponding to the underground communication node.

[0127] The input unit 305 is used to input the stratum penetration difficulty level and the communication link parameters of the wireless communication system into a pre-trained communication decision model to obtain the communication mode decision result output by the communication decision model.

[0128] The control unit 306 is used to control the UAV to communicate with the underground communication node based on the communication mode decision result.

[0129] In one embodiment provided in this application, based on the above-described solution, optionally, the first acquisition unit 301 includes:

[0130] The acquisition subunit is used to acquire at least one communication status information, namely the signal attenuation rate of the underground communication node in the wireless communication system and the intensity of environmental electromagnetic interference.

[0131] In one embodiment provided in this application, based on the above-described solution, optionally, the determining unit 302 includes:

[0132] The first determining subunit is used to determine that the UAV in the wireless communication system meets the communication mode decision conditions when the signal attenuation rate is detected to be greater than the attenuation rate threshold and the environmental electromagnetic interference intensity is less than the preset intensity threshold.

[0133] The second determining subunit is used to determine that the UAV in the wireless communication system does not meet the communication mode decision conditions when the signal attenuation rate is not greater than the attenuation rate threshold or the environmental electromagnetic interference intensity is not less than a preset intensity threshold.

[0134] In one embodiment provided in this application, based on the above-described solution, optionally, the input unit 305 includes:

[0135] The acquisition sub-unit is used to acquire a training sample set and an initial communication decision model. The training dataset includes multiple training samples and sample labels for each training sample. Each training sample includes historical stratigraphic penetration difficulty level and historical communication link parameters. The sample label includes an identifier for indicating the communication mode.

[0136] A training unit is used to train the initial decision model based on the training sample set.

[0137] In one embodiment provided in this application, based on the above-described solution, optionally, the control unit 306 includes:

[0138] The first control subunit is configured to, when the decision result indicates the selection of a low-frequency electromagnetic wave mode, acquire basic beamforming parameters, and, when the area where the underground communication node is located is determined to meet the parameter adjustment conditions based on the rock dielectric constant distribution, adjust at least one of the amplitude weight and phase offset in the basic beamforming parameters to obtain target beamforming parameters; and control the UAV to communicate with the underground communication node according to the target beamforming parameters.

[0139] The second control subunit is used to control the UAV to reduce its flight altitude to within a preset threshold when the decision result indicates that the magnetic induction communication mode is selected, and to activate the multi-input multi-output coil array of the magnetic induction unit to enhance the near-field coupling efficiency, so that the UAV can communicate with the underground communication node.

[0140] The third control subunit is used to adjust the carrier frequency and modulation rate of the acoustic transmitter when the decision result indicates that the acoustic communication mode is selected, so that the UAV can communicate with the underground communication node.

[0141] The fourth control subunit is used to allocate communication time slots of different modes according to the priority of the data to be transmitted when the decision result indicates the selection of the hybrid communication mode, so that the UAV can communicate with the underground communication node.

[0142] This application also provides a storage medium, which includes stored instructions, wherein the instructions, when executed, control the device where the storage medium is located to perform the method described above.

[0143] This application also provides an electronic device, the structural schematic diagram of which is shown below. Figure 4 As shown, it specifically includes a processor 401 and a memory 402 for storing instructions; the processor 401 is configured to execute instructions stored in the memory, causing the electronic device to perform the method described above.

[0144] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0145] Finally, it should be noted that in this paper, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0146] For ease of description, the above system is described by dividing it into various functional units. Of course, in implementing this application, the functions of each unit can be implemented in one or more software and / or hardware.

[0147] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of this application.

[0148] The above provides a detailed description of a UAV ground-penetrating communication method. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for through-ground communication of unmanned aerial vehicles (UAVs), characterized in that, Applied to a wireless communication system, which further includes unmanned aerial vehicles (UAVs) and underground communication nodes, the method includes: In response to a communication command, the communication status information of the wireless communication system is obtained; Based on the communication status information, determine whether the UAV in the wireless communication system meets the communication mode decision conditions; If the UAV is detected to meet the communication mode decision conditions, the communication environment information of the area where the underground communication node is located is obtained; The communication environment information is processed using a pre-trained geological condition prediction model to obtain the stratum penetration difficulty level corresponding to the underground communication node; The stratum penetration difficulty level and the communication link parameters of the wireless communication system are input into a pre-trained communication decision model to obtain the communication mode decision result output by the communication decision model. Based on the decision result of the communication mode, the UAV is controlled to communicate with the underground communication node.

2. The method according to claim 1, characterized in that, Obtaining the communication status information of the wireless communication system includes: The system acquires at least one communication status information, namely, the signal attenuation rate of the underground communication node in the wireless communication system and the intensity of environmental electromagnetic interference.

3. The method according to claim 2, characterized in that, The step of determining whether the UAV in the wireless communication system meets the communication mode decision conditions based on the communication status information includes: If the signal attenuation rate is detected to be greater than the attenuation rate threshold and the environmental electromagnetic interference intensity is less than the preset intensity threshold, it is determined that the UAV in the wireless communication system meets the communication mode decision conditions. If the signal attenuation rate is not greater than the attenuation rate threshold or the environmental electromagnetic interference intensity is not less than the preset intensity threshold, it is determined that the UAV in the wireless communication system does not meet the communication mode decision conditions.

4. The method according to claim 1, characterized in that, The training process of the communication decision model includes: A training sample set and an initial communication decision model are obtained. The training dataset includes multiple training samples and sample labels for each training sample. Each training sample includes historical stratigraphic penetration difficulty level and historical communication link parameters. The sample labels include identifiers used to indicate communication modes. The initial communication decision model is trained based on the training sample set.

5. The method according to claim 1, characterized in that, The step of controlling the UAV to communicate with the underground communication node based on the communication mode decision result includes: When the decision result indicates the selection of a low-frequency electromagnetic wave mode, basic beamforming parameters are obtained. If the distribution of the rock dielectric constant determines that the area where the underground communication node is located meets the parameter adjustment conditions, at least one of the amplitude weight and phase offset in the basic beamforming parameters is adjusted to obtain target beamforming parameters. The UAV is then controlled to communicate with the underground communication node based on the target beamforming parameters. When the decision result indicates that the magnetic induction communication mode is selected, the drone is controlled to reduce its flight altitude to within a preset threshold, and the multi-input multi-output coil array of the magnetic induction unit is activated to enhance the near-field coupling efficiency, so that the drone can communicate with the underground communication node. When the decision result indicates that the acoustic communication mode is selected, the carrier frequency and modulation rate of the acoustic transmitter are adjusted so that the UAV can communicate with the underground communication node; When the decision result indicates the selection of a hybrid communication mode, communication time slots for different modes are allocated according to the priority of the data to be transmitted, enabling the UAV to communicate with the underground communication node.

6. A ground-penetrating communication device for unmanned aerial vehicles (UAVs), characterized in that, The device is applied to a wireless communication system, which also includes unmanned aerial vehicles (UAVs) and underground communication nodes, and includes: The first acquisition unit is used to acquire the communication status information of the wireless communication system in response to a communication command; A determining unit is configured to determine, based on the communication status information, whether the UAV in the wireless communication system meets the communication mode decision conditions; The second acquisition unit is used to acquire communication environment information of the area where the underground communication node is located when the UAV is detected to meet the communication mode decision conditions. The processing unit is used to process the communication environment information using a pre-trained geological state prediction model to obtain the stratum penetration difficulty level corresponding to the underground communication node. The input unit is used to input the stratum penetration difficulty level and the communication link parameters of the wireless communication system into a pre-trained communication decision model to obtain the communication mode decision result output by the communication decision model. The control unit is used to control the UAV to communicate with the underground communication node based on the communication mode decision result.

7. The apparatus according to claim 6, characterized in that, The first acquisition unit includes: The acquisition subunit is used to acquire at least one communication status information, namely the signal attenuation rate of the underground communication node in the wireless communication system and the intensity of environmental electromagnetic interference.

8. The apparatus according to claim 7, characterized in that, The determining unit includes: The first determining subunit is used to determine that the UAV in the wireless communication system meets the communication mode decision conditions when the signal attenuation rate is detected to be greater than the attenuation rate threshold and the environmental electromagnetic interference intensity is less than the preset intensity threshold. The second determining subunit is used to determine that the UAV in the wireless communication system does not meet the communication mode decision conditions when the signal attenuation rate is not greater than the attenuation rate threshold or the environmental electromagnetic interference intensity is not less than a preset intensity threshold.

9. The apparatus according to claim 6, characterized in that, The input unit includes: The acquisition sub-unit is used to acquire a training sample set and an initial communication decision model. The training dataset includes multiple training samples and sample labels for each training sample. Each training sample includes historical stratigraphic penetration difficulty level and historical communication link parameters. The sample label includes an identifier for indicating the communication mode. The training unit is used to train the initial communication decision model based on the training sample set.

10. The apparatus according to claim 6, characterized in that, The control unit includes: The first control subunit is configured to, when the decision result indicates the selection of a low-frequency electromagnetic wave mode, acquire basic beamforming parameters, and, when the area where the underground communication node is located is determined to meet the parameter adjustment conditions based on the rock dielectric constant distribution, adjust at least one of the amplitude weight and phase offset in the basic beamforming parameters to obtain target beamforming parameters; and control the UAV to communicate with the underground communication node according to the target beamforming parameters. The second control subunit is used to control the UAV to reduce its flight altitude to within a preset threshold when the decision result indicates that the magnetic induction communication mode is selected, and to activate the multi-input multi-output coil array of the magnetic induction unit to enhance the near-field coupling efficiency, so that the UAV can communicate with the underground communication node. The third control subunit is used to adjust the carrier frequency and modulation rate of the acoustic transmitter when the decision result indicates that the acoustic communication mode is selected, so that the UAV can communicate with the underground communication node. The fourth control subunit is used to allocate communication time slots of different modes according to the priority of the data to be transmitted when the decision result indicates the selection of the hybrid communication mode, so that the UAV can communicate with the underground communication node.

Images