Intelligent base station for mooring unmanned aerial vehicle

Abstract

The invention provides a mooring unmanned aerial vehicle intelligent base station, a mooring unmanned aerial vehicle is controlled through a mooring line, the mooring unmanned aerial vehicle intelligent base station comprises a control unit, a mooring line take-up and pay-off unit and an environment analysis unit, and the control unit is used for controlling the unmanned aerial vehicle to carry out a preset hovering test and returning hovering flight parameters; in the hovering test process of the mooring unmanned aerial vehicle, a mooring line take-up and pay-off unit is controlled to collect hovering and pay-off parameters; an environment analysis unit is controlled to obtain environment parameters, the hovering flight parameters, the hovering spreading parameters and the environment parameters are input into a preset compensation calculation model, and drive-by-wire compensation parameters are obtained and uploaded; generating a target spreading parameter according to the drive-by-wire compensation parameter and a preset initial spreading parameter; and when a preset flight control instruction is received, the tethered unmanned aerial vehicle is controlled to fly according to the preset flight control instruction, and the tethered line winding and unwinding unit is controlled to unwind the tethered line according to the target unfolding parameters.

Description

Technical Field

[0001] The present application relates to the technical field of tethered drones, and in particular to a tethered drone intelligent base station. Background Art

[0002] At present, with the rapid development of drone technology, tethered drones have attracted widespread attention due to their ability to hover in the air for a long time to perform tasks. They are widely used in emergency communications, aerial monitoring, lighting support and other fields. However, in existing tethered drone systems, the tension management of the tether line has become one of the key factors affecting the performance and safety of drones. In practical applications, excessive or insufficient tension of the tether line will have an adverse effect on the stability and maneuverability of the drone. Excessive tension may cause damage to the drone structure or even cause the tether line to break; while too little tension will loosen the tether line, affecting the positioning accuracy of the drone. In severe cases, it may also cause the drone to lose control and fall, posing a threat to ground personnel and facilities. . Summary of the invention

[0003] The present application provides a tethered drone intelligent base station, which is used to intelligently manage the tension of the tethering line during the flight of the drone, thereby providing a better operating environment for the tethered drone.

[0004] In a first aspect, an embodiment of the present application provides a tethered drone intelligent base station for controlling a tethered drone through a tethered line, wherein the tethered drone intelligent base station comprises: a control unit, a tethered line retracting unit, and an environment analysis unit, wherein the control unit is used to: Sending a flight control instruction to a tethered drone connected to the tethering line to limit the ascent speed and lateral movement speed of the tethered drone; Upon receiving a preset flight test instruction, a first test instruction is sent to the tethered UAV, and a second test instruction is sent to the tethered line retracting and releasing unit, so that the tethered line retracting and releasing unit releases the tethered line according to preset test release parameters, so that the tethered UAV performs a hovering test within a preset distance in the vertical direction, and returns the hovering flight parameters; During the hovering test of the tethered UAV, controlling the tethered line retracting and releasing unit to collect hovering and releasing parameters; Controlling the environmental analysis unit to obtain environmental parameters, and inputting the hovering flight parameters, the hovering deployment parameters and the environmental parameters into a preset compensation calculation model to obtain and upload a control-by-wire compensation parameter; Generating target deployment parameters according to the control-by-wire compensation parameters and preset initial deployment parameters; When a preset flight control instruction is received, the tethered UAV is controlled to fly according to the preset flight control instruction, and the tethered line retracting and extending unit is controlled to extend the tethered line according to the target extension parameter.

[0005] In some embodiments, the preset compensation calculation model includes: a convolutional network, a long short-term memory network, a fully connected neural network, a coding network and a residual network, and the hovering flight parameters, the hovering deployment parameters and the environmental parameters are input into the preset compensation calculation model to obtain and upload the line control compensation parameters, including: Based on the convolutional network, feature extraction processing is performed on the hovering flight parameters to obtain a flight feature vector, wherein the feature extraction processing includes: performing Fourier transform on the hovering flight parameters to extract frequency domain features of the hovering flight parameters after the Fourier transform, wherein the hovering flight parameters include: attitude angle data, velocity data, and acceleration data; Based on the long short-term memory network, adaptively filtering the hovering deployment parameters according to the flight characteristic vector to obtain filtered deployment parameters; According to the filtered deployment parameters and the environmental parameters, forward propagation calculation is performed on the fully connected neural network to obtain initial compensation parameters, wherein the initial compensation parameters include: a mooring line length adjustment amount, a release speed, and a retraction speed; Based on the encoding network, the initial compensation parameter is processed by a particle swarm optimization algorithm to obtain a first tension compensation parameter; Based on the residual network, residual connection and jump connection are performed on the first tension compensation parameter to obtain a second tension compensation parameter, and the second tension compensation parameter is scaled according to a preset safety factor to obtain the line control compensation parameter.

[0006] In some embodiments, the fully connected neural network is forward propagated and calculated according to the filtered deployment parameters and the environmental parameters to obtain initial compensation parameters, and the fully connected neural network layer includes an input layer, a first hidden layer, a second hidden layer, a third hidden layer and an output layer, and each hidden layer uses a ReLU activation function, including: The filtered deployment parameters and the environmental parameters are concatenated to obtain an input vector; According to the input vector, perform matrix multiplication and bias addition operations on the first hidden layer to obtain a first hidden output, and apply a ReLU activation function to the first hidden output; According to the first hidden output, sequentially perform matrix multiplication, bias addition, and ReLU activation function operations on the second hidden layer and the third hidden layer to obtain a second hidden output and a third hidden output; According to the third hidden output, matrix multiplication and bias addition operations are performed on the output layer to obtain an original output vector, and a preset Tanh activation function is used to perform range shrinkage processing on the original output vector to obtain an initial compensation parameter.

[0007] In some embodiments, the initial compensation parameter is processed by a particle swarm optimization algorithm based on the encoding network to obtain a first tension compensation parameter, including: Mapping the initial compensation parameters to a high-dimensional space to generate a particle swarm; Performing encoding processing on each particle in the particle swarm, calculating the distance between particles according to a preset cosine similarity function, and obtaining a particle position vector; Determine the tension fluctuation value according to the particle position vector, calculate the fluctuation amplitude value of the tension fluctuation value, and use the fluctuation amplitude value as the fitness value; Iteratively updating the particle swarm according to the fitness value to obtain an updated particle swarm; The updated particle swarm is judged for convergence until the convergence condition is met, and the global optimal particle is output as the optimized compensation parameter.

[0008] In some embodiments, the residual connection and jump connection are performed on the first tension compensation parameter based on the residual network to obtain the second tension compensation parameter, and the second tension compensation parameter is scaled according to a preset safety factor to obtain the line control compensation parameter, including: According to a preset wavelet transform algorithm, the optimized compensation parameters are subjected to residual decomposition processing to obtain a basic compensation component and a residual component; According to a preset piecewise linear function and the preset safety factor, a linear scaling process is performed on the basic compensation component to obtain a scaled basic compensation component; Performing nonlinear mapping processing on the residual component to obtain a mapped residual component; According to the scaled basic compensation component and the mapped residual component, a weighted fusion process is performed on the two to obtain the line control compensation parameter.

[0009] In some embodiments, the tethered drone intelligent base station further includes: a first power supply unit and a second power supply unit, and the control unit is used to: When it is detected that the power value of the first power supply unit is less than a preset threshold, the second power supply unit is started to supplement the first power supply unit.

[0010] In some embodiments, the tethered drone intelligent base station further includes: A communication unit, the communication unit being arranged in the body of the tethered drone intelligent base station; An antenna unit is provided on the tethered drone, and the antenna unit is communicatively connected with the communication unit via the tethering line.

[0011] In some embodiments, the tethered drone intelligent base station further includes: A remote control unit, the remote control unit being wirelessly connected to the tethered drone intelligent base station; A short-range control unit is arranged on the body of the tethered UAV intelligent base station. When the remote control unit is disconnected, the flight test instruction and the flight control instruction are input through the short-range control unit.

[0012] In some embodiments, the tethered drone performs a first height hovering test, a second height hovering test, and a third height hovering test during the hovering test, and the tethered line retracting and releasing unit is used to: During the hovering test at the first height, collecting the first deployment parameters; During the second height hover test, collect the second deployment parameters During the hover test at the third height, collect the third deployment parameters The hovering deployment parameter is generated according to the first deployment parameter, the second deployment parameter and the third deployment parameter.

[0013] In some embodiments, the tethered drone intelligent base station also includes: a tethered drone.

[0014] The embodiment of the present application provides a tethered drone intelligent base station, which is used to control a tethered drone through a tethered line. The tethered drone intelligent base station includes: a control unit, a tethered line retracting and releasing unit, and an environmental analysis unit. The control unit is used to: send a flight control instruction to a tethered drone connected to the tethered line to limit the ascending speed and lateral speed of the tethered drone; upon receiving a preset flight test instruction, send a first test instruction to the tethered drone, and send a second test instruction to the tethered line retracting and releasing unit, so that the tethered line retracting and releasing unit releases the tethered line according to the preset test release parameters, so that the tethered drone is moved in the vertical direction. The method comprises the steps of: performing a hovering test within a preset distance on the tethered UAV and transmitting back the hovering flight parameters; controlling the tethered line retraction and extension unit to collect the hovering extension parameters during the hovering test of the tethered UAV; controlling the environmental analysis unit to obtain the environmental parameters, and inputting the hovering flight parameters, the hovering extension parameters and the environmental parameters into a preset compensation calculation model to obtain and upload the line-control compensation parameters; generating the target extension parameters according to the line-control compensation parameters and the preset initial extension parameters; and controlling the tethered UAV to fly according to the preset flight control instructions upon receiving the preset flight control instructions, and controlling the tethered line retraction and extension unit to extend the tethered line according to the target extension parameters. Through the above method, the flight control instructions are used to effectively limit the ascent speed and lateral movement speed of the tethered UAV to ensure operation within a safe range. Through the flight test instructions, the tethered UAV is guided to complete the hovering test, and the flight parameters and tether line deployment parameters during the hovering process are collected. The introduction of the environmental analysis unit enables the system to obtain the surrounding environmental parameters in real time, such as wind speed, temperature, etc., further improving the system's adaptability to complex environments. By inputting the hovering flight parameters, hovering deployment parameters and environmental parameters into the preset compensation calculation model, the tethered UAV intelligent base station can calculate more accurate target deployment parameters, which improves the stability and controllability of the tethered UAV under different environmental conditions, and also significantly enhances the reliability and safety of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying any creative work.

[0016] Figure 1 A schematic block diagram of a tethered drone intelligent base station provided in an embodiment of the present application; Figure 2 A schematic flow chart of a control method for a tethered drone provided in an embodiment of the present application. DETAILED DESCRIPTION

[0017] 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 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.

[0018] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0019] It should also be understood that the terms used in this application specification are only for the purpose of describing specific embodiments and are not intended to limit the application. As used in this application specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0020] It should be further understood that the term “and / or” used in the specification and appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0021] See also Figure 1 , Figure 1 The embodiment of the present application also provides a schematic block diagram of a tethered drone intelligent base station. Figure 1 As shown, the tethered drone intelligent base station 100 includes: a control unit 11, a tethered line retracting and releasing unit 12, an environment analysis unit 13 and a tethered line 14. The tethered drone intelligent base station 100 is connected to a tethered drone 15 via the tethered line 14.

[0022] In some embodiments, the tethered drone 15 may also serve as a part of the tethered drone smart base station 100 , for example, as a flight unit of the tethered drone smart base station 100 .

[0023] See also Figure 2 , Figure 2 The embodiment of the present application also provides a schematic flow chart of a control method for a tethered drone. Figure 1 As shown, the control method of the tethered drone is executed by the control unit 11, and the specific steps include: S101-S106.

[0024] S101. Send a flight control command to a tethered UAV connected to a tethering line to limit an ascending speed and a lateral moving speed of the tethered UAV.

[0025] For example, after the tethered drone is connected to the tethering line, the tethered drone is set to novice mode. In the novice mode, the ascending speed and lateral speed of the tethered drone are limited, which can effectively prevent the tethering line from being overstretched or loosened. This method ensures that the tethered drone responds quickly to flight commands while ensuring safety.

[0026] S102. When receiving a preset flight test instruction, a first test instruction is sent to the tethered UAV, and a second test instruction is sent to the tethered line retraction and extension unit, so that the tethered line retraction and extension unit extends the tethered line according to the preset test extension parameters, so that the tethered UAV performs a hovering test within a preset distance in the vertical direction and sends back the hovering flight parameters.

[0027] Exemplarily, after receiving a preset flight test instruction, the control unit needs to send a first test instruction to the tethered drone, instructing the drone to enter the hovering test mode. At the same time, a second test instruction is sent to the tethered line retraction unit to control the tethered line to perform a deployment operation according to the preset test deployment parameters. The test deployment parameters include key parameters such as deployment speed and deployment length to ensure that the drone can perform a vertical hovering test within a preset distance. During the hovering test, the drone will send back the hovering flight parameters in real time. These parameters include attitude angle, vertical height, speed, acceleration and other information. The control unit evaluates the stability and control accuracy of the drone in the hovering state by receiving and recording these flight parameters. This process helps to verify the flight performance of the drone and provide necessary data support for subsequent compensation calculations.

[0028] S103, during the hovering test of the tethered UAV, controlling the tethering line retraction and extension unit to collect hovering extension parameters.

[0029] For example, during the hovering test of the drone, the control unit also needs to command the tethered line retraction and extension unit to collect data, specifically including collecting hovering deployment parameters. These parameters usually include the deployment length, deployment speed, pulling force, tension, etc. of the tethered line. The tethered line retraction and extension unit monitors and records these parameters in real time through built-in sensors and data acquisition modules. The collection of hovering deployment parameters is crucial for evaluating the working status of the tethered line and the hovering performance of the drone. For example, through a tension sensor, the tension changes of the tethered line can be monitored to determine whether the drone is in a stable hovering state. In addition, changes in deployment speed and length can also provide dynamic information of the tethered line at different deployment stages. These data are not only important inputs for subsequent compensation calculations, but also provide a theoretical basis for further optimizing the tethered line retraction and extension control strategy.

[0030] S104, controlling the environment analysis unit to obtain environmental parameters, and inputting the hovering flight parameters, the hovering deployment parameters and the environmental parameters into a preset compensation calculation model to obtain and upload the control-by-wire compensation parameters.

[0031] Exemplarily, the control unit also needs to control the environmental analysis unit to obtain environmental parameters. These environmental parameters include but are not limited to wind speed, wind direction, temperature, humidity, air pressure, etc. These parameters have a direct impact on the hovering stability and flight performance of the drone. The environmental analysis unit usually collects environmental data in real time through a variety of sensors and transmits it to the control unit. Subsequently, the control unit inputs the hovering flight parameters, hovering deployment parameters and environmental parameters into a preset compensation calculation model. The compensation calculation model is usually based on a complex algorithm model, such as a neural network or other machine learning model, which can comprehensively analyze the input multi-dimensional data and calculate the line control compensation parameters. The line control compensation parameters are used to adjust the control strategy of the drone to adapt to the current flight state and environmental conditions. Finally, the line control compensation parameters will be uploaded to the drone and the tethered line retraction and release unit to ensure that the flight of the drone and the deployment of the tethered line can be optimized and controlled.

[0032] S105 , generating target deployment parameters according to the control-by-wire compensation parameters and the preset initial deployment parameters.

[0033] Exemplarily, after obtaining the control-by-wire compensation parameters, the control unit needs to generate target deployment parameters in combination with the preset initial deployment parameters. The initial deployment parameters are set based on theoretical calculations and empirical data, including basic parameters such as deployment length and deployment speed. The control-by-wire compensation parameters are adjustment parameters calculated under actual flight and environmental conditions. Combining the two, more accurate and adaptive target deployment parameters can be generated. These target deployment parameters not only take into account the actual flight state of the tethered drone, but also incorporate the influence of current environmental conditions. In this way, the target deployment parameters can more effectively guide the operation of the tethered line retraction unit, ensuring that the drone can maintain a stable and safe flight state in various complex environments. At the same time, the target deployment parameters also provide a reliable basis for subsequent flight control.

[0034] It should be noted that the preset initial line control parameters are determined according to the type of tethering line and the type of tethered drone.

[0035] S106, when receiving a preset flight control instruction, controlling the tethered UAV to fly according to the preset flight control instruction, and controlling the tethered line retracting and extending unit to extend the tethered line according to the target extension parameter.

[0036] Upon receiving the preset flight control instructions, the control unit needs to perform a series of operations to achieve precise control of the tethered drone and the tethered line retraction and release unit. First, the control unit adjusts the flight state of the drone according to the flight control instructions, such as ascending, descending, advancing, retreating, etc. At the same time, the control unit also needs to control the tethered line retraction and release unit to perform corresponding deployment operations according to the target deployment parameters generated previously. Specifically, the tethered line retraction and release unit will be adjusted according to the deployment length and speed in the target deployment parameters to match the flight action of the drone. This process requires a high degree of synchronization and coordination to ensure that the flight trajectory of the drone and the tension of the tethered line remain within a safe range. Through this dual control, the tethered drone can remain stable and reliable when performing complex flight actions, avoiding safety hazards caused by improper deployment. At the same time, this also provides a guarantee for the flexible application of drones under various environmental conditions.

[0037] The embodiment of the present application provides a tethered drone intelligent base station, which is used to control a tethered drone through a tethered line. The tethered drone intelligent base station includes: a control unit, a tethered line retracting and releasing unit, and an environmental analysis unit. The control unit is used to: send a flight control instruction to a tethered drone connected to the tethered line to limit the ascending speed and lateral speed of the tethered drone; upon receiving a preset flight test instruction, send a first test instruction to the tethered drone, and send a second test instruction to the tethered line retracting and releasing unit, so that the tethered line retracting and releasing unit releases the tethered line according to the preset test release parameters, so that the tethered drone is moved in the vertical direction. The method comprises the steps of: performing a hovering test within a preset distance on the tethered UAV and transmitting back the hovering flight parameters; controlling the tethered line retraction and extension unit to collect the hovering extension parameters during the hovering test of the tethered UAV; controlling the environmental analysis unit to obtain the environmental parameters, and inputting the hovering flight parameters, the hovering extension parameters and the environmental parameters into a preset compensation calculation model to obtain and upload the line-control compensation parameters; generating the target extension parameters according to the line-control compensation parameters and the preset initial extension parameters; and controlling the tethered UAV to fly according to the preset flight control instructions upon receiving the preset flight control instructions, and controlling the tethered line retraction and extension unit to extend the tethered line according to the target extension parameters. Through the above method, the flight control instructions are used to effectively limit the ascent speed and lateral movement speed of the tethered UAV to ensure operation within a safe range. Through the flight test instructions, the tethered UAV is guided to complete the hovering test, and the flight parameters and tether line deployment parameters during the hovering process are collected. The introduction of the environmental analysis unit enables the system to obtain the surrounding environmental parameters in real time, such as wind speed, temperature, etc., further improving the system's adaptability to complex environments. By inputting the hovering flight parameters, hovering deployment parameters and environmental parameters into the preset compensation calculation model, the tethered UAV intelligent base station can calculate more accurate target deployment parameters, which improves the stability and controllability of the tethered UAV under different environmental conditions, and also significantly enhances the reliability and safety of the system.

[0038] In order to more clearly introduce the technical solution of the present application, the technical solution of the present application will be introduced through specific embodiments below. It should be noted that the specific embodiments are used to expand the technical solution of the present application, but are not intended to limit the present application.

[0039] In some embodiments, the preset compensation calculation model includes: a convolutional network, a long short-term memory network, a fully connected neural network, a coding network and a residual network. The hovering flight parameters, the hovering deployment parameters and the environmental parameters are input into the preset compensation calculation model to obtain and upload the line control compensation parameters. The specific steps include: S1041-S1045.

[0040] S1041. Based on the convolutional network, feature extraction processing is performed on the hovering flight parameters to obtain a flight feature vector. The feature extraction processing includes: performing Fourier transform on the hovering flight parameters, and extracting frequency domain features of the hovering flight parameters after the Fourier transform. The hovering flight parameters include: attitude angle data, velocity data, and acceleration data.

[0041] Exemplarily, the hovering flight parameters are Fourier transformed to obtain the input data of the convolutional network. The convolutional network scans the input data through multiple convolution kernels to extract local features. For the hovering flight parameters, convolution kernels of different sizes can be used to capture patterns of different scales. For example, a small convolution kernel (such as 3x3) can capture subtle posture changes, while a large convolution kernel (such as 7x7) can capture a wider range of motion trends. In addition, by stacking multiple layers of convolution and pooling operations, more advanced features can be gradually extracted. In the time-frequency domain analysis, one-dimensional convolution can be applied to time series data, or two-dimensional convolution can be used to process time-frequency graphs.

[0042] S1042. Based on the long short-term memory network, adaptive filtering is performed on the hovering deployment parameters according to the flight characteristic vector to obtain filtered deployment parameters.

[0043] For example, the long short-term memory network (LSTM) can learn long-term dependencies and effectively capture the dynamic changes of hover deployment parameters. The gating mechanism of LSTM (input gate, forget gate, and output gate) enables it to selectively memorize or forget information, which is crucial for filtering out noise and retaining important information. By taking the flight feature vector as an additional input, LSTM can dynamically adjust its internal state and output according to the current flight status. This adaptive capability enables the filtering process to better adapt to different flight conditions and environmental changes, improving the accuracy and robustness of the filtering.

[0044] S1043. Perform forward propagation calculation on the fully connected neural network according to the filtered deployment parameters and environmental parameters to obtain initial compensation parameters. The initial compensation parameters include: mooring line length adjustment amount, release speed, and retraction speed.

[0045] In some embodiments, the fully connected neural network layer includes an input layer, a first hidden layer, a second hidden layer, a third hidden layer and an output layer. According to the filtered expansion parameters and environmental parameters, the fully connected neural network is forward propagated to obtain initial compensation parameters, including: concatenating the filtered expansion parameters and environmental parameters to obtain an input vector; according to the input vector, performing matrix multiplication and bias addition operations on the first hidden layer to obtain a first hidden output, and applying a ReLU activation function to the first hidden output; according to the first hidden output, performing matrix multiplication, bias addition and ReLU activation function operations on the second hidden layer and the third hidden layer in turn to obtain a second hidden output and a third hidden output; according to the third hidden output, performing matrix multiplication and bias addition operations on the output layer to obtain an original output vector, and using a preset Tanh activation function to perform range shrinkage processing on the original output vector to obtain initial compensation parameters.

[0046] For example, the use of the ReLU activation function can introduce nonlinearity while avoiding the gradient vanishing problem, which is conducive to the training of deep networks. The design of three hidden layers allows the model to learn complex nonlinear relationships, and each layer can capture features at different levels. The output layer generates initial compensation parameters. This network structure can be trained end-to-end through the back-propagation algorithm, enabling it to automatically learn the mapping relationship from input to compensation parameters without explicitly programming complex rules.

[0047] S1044. Based on the encoding network, the initial compensation parameters are processed by a particle swarm optimization algorithm to obtain a first tension compensation parameter.

[0048] Exemplarily, when optimizing compensation parameters, the encoding network can learn the intrinsic structure and constraints of the parameters. By minimizing the reconstruction error, the encoding network can remove noise, extract the most important features, and generate more stable and consistent compensation parameters. In addition, by adding regularization or sparse constraints in the intermediate layers, the optimization results can be further improved.

[0049] In some embodiments, based on the encoding network, the initial compensation parameters are processed by a particle swarm optimization algorithm to obtain a first tension compensation parameter, including: mapping the initial compensation parameter to a high-dimensional space to generate a particle swarm; encoding each particle in the particle swarm, calculating the distance between particles according to a preset cosine similarity function to obtain a particle position vector; determining the tension fluctuation value according to the particle position vector, and calculating the fluctuation amplitude value of the tension fluctuation value, and using the fluctuation amplitude value as the fitness value; iteratively updating the particle swarm according to the fitness value to obtain an updated particle swarm; performing convergence judgment on the updated particle swarm until the convergence condition is met, and outputting the global optimal particle as the optimized compensation parameter.

[0050] By iteratively updating the particle swarm until the convergence conditions are met, not only is the algorithm's ability to find the global optimal solution guaranteed, but it also speeds up the optimization process and reduces debugging time, providing strong support for achieving efficient and stable tension control in industrial production.

[0051] S1045. Based on the residual network, perform residual connection and jump connection on the first tension compensation parameter to obtain a second tension compensation parameter, and scale the second tension compensation parameter according to a preset safety factor to obtain a line control compensation parameter.

[0052] In some embodiments, based on a residual network, a first tension compensation parameter is residually connected and jump-connected to obtain a second tension compensation parameter, and the second tension compensation parameter is scaled according to a preset safety factor to obtain a line control compensation parameter, including: performing residual decomposition processing on the optimized compensation parameter according to a preset wavelet transform algorithm to obtain a basic compensation component and a residual component; performing linear scaling processing on the basic compensation component according to a preset piecewise linear function and a preset safety factor to obtain a scaled basic compensation component; performing nonlinear mapping processing on the residual component to obtain a mapped residual component, and performing weighted fusion processing on the scaled basic compensation component and the mapped residual component to obtain a line control compensation parameter.

[0053] For example, by incorporating the safety factor into the residual block, the model can dynamically adjust the output according to different safety requirements. This structure not only improves the expressiveness of the model, but also enhances its adaptability to different operating conditions, ensuring the safety and reliability of the final compensation parameters.

[0054] The above compensation calculation model needs to be trained in advance. During the training process, the tethered drone base station and the tethered drone are placed in a closed wind chamber, and environmental factors such as wind direction, wind force, and blowing temperature are set according to requirements to simulate the working conditions of the tethered drone. If necessary, corresponding sensors need to be installed on the tethered drone to obtain more detailed data.

[0055] In some embodiments, the tethered drone intelligent base station also includes: a first power supply unit and a second power supply unit, and the control unit is used to: when it is detected that the power value of the first power supply unit is less than a preset threshold, start the second power supply unit to replenish the first power supply unit.

[0056] Exemplarily, the first power supply unit is an energy storage device such as a battery, and the second power supply unit includes: a generator and a solar panel.

[0057] By setting up the first power supply unit and the second power supply unit in the tethered drone intelligent base station, and the control unit intelligently manages the energy conversion between the two, the operational reliability and continuous operation capability are significantly improved. When it is detected that the power of the main power supply (i.e., the first power supply unit) drops below the preset safety threshold, the tethered drone intelligent base station can automatically activate the backup power supply (the second power supply unit) to replenish the power for the first power supply unit in time, thereby avoiding the risk of flight interruption or equipment damage caused by the exhaustion of the main power supply. This dual power supply configuration not only enhances redundancy and improves the ability to respond to emergencies, but also extends the execution cycle of a single mission and reduces the need for human intervention, which is of great significance for ensuring long-term and continuous aerial operations. In addition, the intelligent energy management system also helps to optimize the overall power efficiency and reduce operating costs.

[0058] In some embodiments, the tethered drone intelligent base station further includes: a communication unit and an antenna unit. The communication unit is disposed in the body of the tethered drone intelligent base station. The antenna unit is disposed in the tethered drone, and the antenna unit is communicatively connected to the communication unit via a tethering line.

[0059] This technical solution integrates a communication unit in the tethered drone intelligent base station, sets an antenna unit on the drone, and uses a tethered line to achieve a stable communication connection between the two, effectively improving the reliability and real-time performance of data transmission. Placing the communication unit inside the base station body not only facilitates centralized management and maintenance, but also provides a more stable signal processing environment; and the antenna unit is directly installed on the drone, which ensures that good signal reception and transmission capabilities can be maintained even in complex flight environments. This design overcomes the problems of traditional wireless communications being susceptible to interference and distance limitations, and provides continuous and uninterrupted data support for applications such as remote control and video surveillance. At the same time, communication methods based on wired connections also enhance the security of information and reduce the risk of data leakage, which is particularly important for applications that perform sensitive tasks or require a high degree of confidentiality.

[0060] In some embodiments, the tethered drone intelligent base station further includes: a remote control unit and a short-range control unit. The remote control unit is wirelessly connected to the tethered drone intelligent base station. The short-range control unit is arranged on the body of the tethered drone intelligent base station. When the remote control unit is disconnected, the flight test instructions and flight control instructions are input through the short-range control unit.

[0061] In some embodiments, the tethered drone performs a first altitude hovering test, a second altitude hovering test, and a third altitude hovering test during a hovering test, and the tethered line retraction and extension unit is used to: collect first extension parameters during the first altitude hovering test; collect second extension parameters during the second altitude hovering test; collect third extension parameters during the third altitude hovering test; and generate hovering extension parameters based on the first extension parameters, the second extension parameters, and the third extension parameters.

[0062] This technical solution provides a more accurate and comprehensive control basis for the tethered UAV intelligent base station by conducting hovering tests at different altitudes and collecting corresponding deployment parameters. When performing hovering tests at the first, second and third altitudes respectively, the tethered line retraction and release unit can dynamically adjust and record key deployment parameters according to actual flight conditions, which not only helps to analyze the stability and response characteristics of the UAV at different altitudes, but also lays a data foundation for the subsequent optimization of hovering performance. The comprehensive hovering deployment parameters generated based on multiple sets of deployment parameters can more accurately reflect the behavioral characteristics of the tethered UAV and the tethered line within the entire working range, thereby achieving smoother and more stable hovering control. In addition, this method also enhances the adaptability of the tethered UAV intelligent base station, enabling it to maintain good operating performance in complex and changing environments, and improves the success rate and safety of mission execution.

[0063] An embodiment of the present application provides a base station device, which includes a memory and a processor; the memory is used to store a computer program; the processor is used to execute the computer program and implement a control method for a tethered drone such as any one of the embodiments of the present application when executing the computer program.

[0064] An embodiment of the present application provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the processor implements a control method for a tethered drone as described in any one of the embodiments of the present application.

[0065] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present application, and these modifications or replacements should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A tethered drone intelligent base station, characterized in that: Used to control a tethered drone through a tethered line, the tethered drone intelligent base station includes: a control unit, a tethered line retracting unit and an environment analysis unit, and the control unit is used to: Sending a flight control instruction to a tethered drone connected to the tethering line to limit the ascent speed and lateral movement speed of the tethered drone; Upon receiving a preset flight test instruction, a first test instruction is sent to the tethered UAV, and a second test instruction is sent to the tethered line retracting and releasing unit, so that the tethered line retracting and releasing unit releases the tethered line according to preset test release parameters, so that the tethered UAV performs a hovering test within a preset distance in the vertical direction, and returns the hovering flight parameters; During the hovering test of the tethered UAV, controlling the tethered line retracting and releasing unit to collect hovering and releasing parameters; Controlling the environmental analysis unit to obtain environmental parameters, and inputting the hovering flight parameters, the hovering deployment parameters and the environmental parameters into a preset compensation calculation model to obtain and upload a control-by-wire compensation parameter; Generating target deployment parameters according to the control-by-wire compensation parameters and preset initial deployment parameters; When a preset flight control instruction is received, the tethered UAV is controlled to fly according to the preset flight control instruction, and the tethered line retracting and extending unit is controlled to extend the tethered line according to the target extension parameter.

2. The tethered drone base station according to claim 1, characterized in that: The preset compensation calculation model includes: a convolutional network, a long short-term memory network, a fully connected neural network, a coding network and a residual network. The hovering flight parameters, the hovering deployment parameters and the environmental parameters are input into the preset compensation calculation model to obtain and upload the line control compensation parameters, including: Based on the convolutional network, feature extraction processing is performed on the hovering flight parameters to obtain a flight feature vector, wherein the feature extraction processing includes: performing Fourier transform on the hovering flight parameters to extract frequency domain features of the hovering flight parameters after the Fourier transform, wherein the hovering flight parameters include: attitude angle data, velocity data, and acceleration data; Based on the long short-term memory network, adaptively filtering the hovering deployment parameters according to the flight characteristic vector to obtain filtered deployment parameters; According to the filtered deployment parameters and the environmental parameters, forward propagation calculation is performed on the fully connected neural network to obtain initial compensation parameters, wherein the initial compensation parameters include: a mooring line length adjustment amount, a release speed, and a retraction speed; Based on the encoding network, the initial compensation parameter is processed by a particle swarm optimization algorithm to obtain a first tension compensation parameter; Based on the residual network, residual connection and jump connection are performed on the first tension compensation parameter to obtain a second tension compensation parameter, and the second tension compensation parameter is scaled according to a preset safety factor to obtain the line control compensation parameter.

3. The tethered drone base station according to claim 2, characterized in that: The fully connected neural network layer comprises an input layer, a first hidden layer, a second hidden layer, a third hidden layer and an output layer, each hidden layer uses a ReLU activation function, and the fully connected neural network is forward propagated and calculated according to the filtered display parameters and the environmental parameters to obtain initial compensation parameters, including: The filtered deployment parameters and the environmental parameters are concatenated to obtain an input vector; According to the input vector, perform matrix multiplication and bias addition operations on the first hidden layer to obtain a first hidden output, and apply a ReLU activation function to the first hidden output; According to the first hidden output, sequentially perform matrix multiplication, bias addition, and ReLU activation function operations on the second hidden layer and the third hidden layer to obtain a second hidden output and a third hidden output; According to the third hidden output, matrix multiplication and bias addition operations are performed on the output layer to obtain an original output vector, and a preset Tanh activation function is used to perform range shrinkage processing on the original output vector to obtain an initial compensation parameter.

4. The tethered drone base station according to claim 2, characterized in that: The method of performing particle swarm optimization algorithm processing on the initial compensation parameter based on the encoding network to obtain the first tension compensation parameter includes: Mapping the initial compensation parameters to a high-dimensional space to generate a particle swarm; Performing encoding processing on each particle in the particle swarm, calculating the distance between particles according to a preset cosine similarity function, and obtaining a particle position vector; Determine the tension fluctuation value according to the particle position vector, calculate the fluctuation amplitude value of the tension fluctuation value, and use the fluctuation amplitude value as the fitness value; Iteratively updating the particle swarm according to the fitness value to obtain an updated particle swarm; The updated particle swarm is judged for convergence until the convergence condition is met, and the global optimal particle is output as the optimized compensation parameter.

5. The tethered drone base station according to claim 2, characterized in that: The method of performing residual connection and jump connection on the first tension compensation parameter based on the residual network to obtain a second tension compensation parameter, and scaling the second tension compensation parameter according to a preset safety factor to obtain the line control compensation parameter includes: According to a preset wavelet transform algorithm, the optimized compensation parameters are subjected to residual decomposition processing to obtain a basic compensation component and a residual component; According to a preset piecewise linear function and the preset safety factor, a linear scaling process is performed on the basic compensation component to obtain a scaled basic compensation component; Performing nonlinear mapping processing on the residual component to obtain a mapped residual component; According to the scaled basic compensation component and the mapped residual component, a weighted fusion process is performed on the two to obtain the line control compensation parameter.

6. The tethered drone base station according to claim 1, characterized in that: The tethered drone intelligent base station further includes: a first power supply unit and a second power supply unit, and the control unit is used to: When it is detected that the power value of the first power supply unit is less than a preset threshold, the second power supply unit is started to supplement the first power supply unit.

7. The tethered drone base station according to claim 1, characterized in that: The tethered drone intelligent base station also includes: A communication unit, the communication unit being arranged in the body of the tethered drone intelligent base station; An antenna unit is provided on the tethered drone, and the antenna unit is communicatively connected with the communication unit via the tethering line.

8. The tethered drone base station according to claim 1, characterized in that: The tethered drone intelligent base station also includes: A remote control unit, the remote control unit being wirelessly connected to the tethered drone intelligent base station; A short-range control unit is arranged on the body of the tethered UAV intelligent base station. When the remote control unit is disconnected, the flight test instruction and the flight control instruction are input through the short-range control unit.

9. The tethered drone base station according to claim 1, characterized in that: During the hovering test of the tethered drone, the tethered drone performs a first height hovering test, a second height hovering test and a third height hovering test respectively, and the tethered line retracting and releasing unit is used for: During the hovering test at the first height, collecting the first deployment parameters; During the second height hover test, collect the second deployment parameters During the hover test at the third height, collect the third deployment parameters The hovering deployment parameter is generated according to the first deployment parameter, the second deployment parameter and the third deployment parameter.

10. The tethered drone base station according to claim 1, characterized in that: The tethered drone intelligent base station also includes: a tethered drone.

Images