Transmitting system and transmitting method for array type launching unmanned aerial vehicle cluster

Abstract

The invention relates to the technical field of unmanned aerial vehicles, and particularly discloses an array launching unmanned aerial vehicle cluster launching system and method, and the method comprises the steps: carrying out the state detection and in-place detection of each unit launcher after a cluster launcher main power supply is turned on, and resetting a launcher pitching mechanism after confirmation, and the angle is adjusted to 90 degrees. Then, the unmanned aerial vehicles are powered on, the working states of the unmanned aerial vehicles are started, the launching angles of the unmanned aerial vehicles are set and adjusted after the self-inspection process of the unmanned aerial vehicles is completed, and all the unmanned aerial vehicles to be launched are displayed on a launching display terminal; afterwards, the safety of the unmanned aerial vehicle is relieved, the unmanned aerial vehicle is launched to lift off through an ignition device and enters a hovering mode to start a task, collected data are transmitted to a ground station, the battery state of the unmanned aerial vehicle is further analyzed, and corresponding battery management is carried out. In this way, efficient launching of the unmanned aerial vehicle cluster is achieved, and the cooperative control capability and operation safety of the unmanned aerial vehicle cluster are improved.

Description

Technical Field

[0001] The present application relates to the field of UAV technology, and more specifically, to a launching system and launching method for launching a cluster of UAVs in an array. Background Art

[0002] With the development of science and technology, drone technology has rapidly developed and popularized. Among them, small drones have shown great potential in reconnaissance and surveillance, aerial photography and film and television production, agricultural applications, logistics and distribution, scientific research, entertainment and sports, education and training, infrastructure inspection and public safety due to their small size, low cost and easy operation.

[0003] However, the main take-off methods for small and medium-sized drones on the market currently include ground taxiing, hand throwing, ground vertical take-off, rocket boost, etc. These take-off methods have high site requirements, slow take-off speed, inconvenient drone transportation, long preparation time before take-off, cumbersome operation, and high requirements for users. In addition, these launch methods are usually only suitable for the launch of a single drone or a small number of drones, and cannot achieve rapid cluster release, multi-drone coordinated control and monitoring.

[0004] Therefore, an optimized launch system and launch method for array-type launch of drone clusters is desired. Summary of the invention

[0005] In order to solve the above technical problems, the present application is proposed. The embodiment of the present application provides a system and a method for launching a drone cluster in an array, which includes: after turning on the main power supply of the cluster transmitter, performing status detection and in-situ detection on each unit transmitter, and after confirming that there is no error, resetting the pitch mechanism of the transmitter and adjusting its angle to 90 degrees. Then, the drone is powered on and started in its working state. After its self-check process is completed, the launch angle of the drone is set and adjusted, and all drones ready to be launched are displayed on the launch display terminal. Thereafter, the drone insurance is released, the drone is launched into the air through the ignition device, enters the hovering mode to start the mission, and the collected data is transmitted to the ground station, and the battery status of the drone is further analyzed according to the collected data and corresponding battery management is performed. In this way, it is helpful to achieve efficient launch of drone clusters and improve the collaborative control capability and operational safety of drone clusters.

[0006] According to one aspect of the present application, a method for launching a drone cluster in an array is provided, which includes: Step 1: Turn on the main power of the cluster transmitter; Step 2: Perform status detection on each transmitter in the cluster transmitter to obtain a status detection result. If the status detection result is normal, execute step 3; if the status detection result is abnormal, end the process and perform manual troubleshooting and maintenance; Step 3: Perform an in-place detection on the drones to determine whether all drones are correctly installed in the launch tube of the launcher to obtain an in-place detection result. If the in-place detection result is normal, execute step 4. If the in-place detection result is abnormal, mark the abnormal drone number on the launch display terminal; Step 4: resetting the pitch mechanism of each transmitter; Step 5: Start the pitch mechanism of each transmitter until the pitch angle is equal to 90°. If the pitch angle reaches 90°, execute step 6; if the pitch angle does not reach 90°, execute step 4 and then step 5 until the pitch angle is equal to 90°; Step 6: Power on each of the drones so that each of the drones is in working state; Step 7: Perform a self-test on the flight control system and / or the electronic speed governor of each UAV to obtain a self-test result. If the self-test result is passed, execute step 8. If the self-test result is not passed, mark the abnormal UAV number on the transmitting display terminal; Step 8: Setting the launch angle of each drone, and controlling the pitch mechanism of each launcher to adjust the launch angle of each drone; Step 9: Displaying the UAV to be launched on the launch display terminal; Step 10: releasing the safety of each of the drones to be launched, and igniting and launching the drones to be launched into the sky, wherein the drones launched into the sky enter a hovering mode; Step 11: The drone transmits the collected data back.

[0007] In the above-mentioned method for launching a cluster of array-type drones, the step 11 includes: receiving a time queue of ambient temperature collected by an ambient temperature sensor and a time queue of battery discharge temperature collected by a battery temperature sensor; performing timing analysis on the time queue of ambient temperature and the time queue of battery discharge temperature to obtain an ambient temperature timing-associated implicit coding vector and a battery discharge temperature timing-associated implicit coding vector; generating an auxiliary decision result for drone battery management based on the timing-associated interaction characteristics between the ambient temperature timing-associated implicit coding vector and the battery discharge temperature timing-associated implicit coding vector; if the auxiliary decision result is to turn on the battery self-heating function, the auxiliary decision result is popped up, and it is asked whether to confirm turning on the battery self-heating function; after receiving the user's confirmation indication, a command to start the battery self-heating function is sent to the drone.

[0008] According to another aspect of the present application, a system for launching a cluster of drones in an array is provided, comprising: The drone data receiving module is used to receive the time queue of the ambient temperature collected by the ambient temperature sensor and the time queue of the battery discharge temperature collected by the battery temperature sensor; A temperature data time series analysis module, used for performing time series analysis on the time series of the ambient temperature and the time series of the battery discharge temperature to obtain an ambient temperature time series associated implicit coding vector and a battery discharge temperature time series associated implicit coding vector; A battery management auxiliary decision module, used to generate an auxiliary decision result for the battery management of the drone based on the time series correlation interaction characteristics between the ambient temperature time series correlation implicit coding vector and the battery discharge temperature time series correlation implicit coding vector; A user confirmation module, for, in response to the auxiliary decision result being to enable the battery self-heating function, popping up the auxiliary decision result and asking whether to confirm enabling the battery self-heating function; The battery management control module is used to send a battery self-heating function start-up instruction to the drone after receiving a user confirmation instruction.

[0009] Compared with the prior art, the present application provides a launch system and launch method for array-type launch of drone clusters, which includes: after turning on the main power supply of the cluster transmitter, performing status detection and in-situ detection on each unit transmitter, and after confirmation, resetting the pitch mechanism of the transmitter and adjusting its angle to 90 degrees. Then, the drone is powered on and started in its working state. After its self-check process is completed, the launch angle of the drone is set and adjusted, and all drones ready for launch are displayed on the launch display terminal. Thereafter, the drone insurance is released, the drone is launched into the air through the ignition device, enters the hovering mode to start the mission, and the collected data is transmitted to the ground station, and the battery status of the drone is further analyzed based on the collected data and corresponding battery management is performed. In this way, it is helpful to achieve efficient launch of drone clusters and improve the collaborative control capability and operational safety of drone clusters. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0011] Figure 1 This is a flowchart of a method for launching a drone cluster in an array according to an embodiment of the present application.

[0012] Figure 2 This is a flowchart of step S11 in the method for launching a drone cluster in an array according to an embodiment of the present application.

[0013] Figure 3 This is a data flow diagram of step S11 in the method for array-based launch of a drone cluster according to an embodiment of the present application.

[0014] Figure 4 This is a flowchart of step S130 in the method for launching a drone cluster in an array according to an embodiment of the present application.

[0015] Figure 5 This is a flowchart of step S131 in the method for launching a drone cluster in an array according to an embodiment of the present application.

[0016] Figure 6 Schematic diagram of the architecture of a system for array-based launch of drone clusters according to an embodiment of the present application.

[0017] Figure 7 It is a communication diagram of a system for launching a drone cluster in an array according to an embodiment of the present application.

[0018] Figure 8 This is a schematic diagram of a method for powering on drones in an array-type drone cluster launch system according to an embodiment of the present application.

[0019] In the picture, 1. tube-launched folding drone; 2. cluster launcher; 3. remote control unit; 4. BMS management system; 5. launch display terminal. DETAILED DESCRIPTION

[0020] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0021] As mentioned in the background technology above, the main take-off methods for small and medium-sized drones on the market currently include ground taxiing, hand throwing, ground vertical take-off, rocket boost, etc. These take-off methods have high site requirements, slow take-off speed, inconvenient drone transportation, long preparation time before take-off, cumbersome operation, and high requirements for users. In addition, these launch methods are usually only suitable for the launch of a single drone or a small number of drones, and cannot achieve rapid cluster release, multi-machine collaborative control and monitoring.

[0022] In this regard, the present application proposes a method for launching a cluster of drones in an array, which includes: after turning on the main power supply of the cluster transmitter, performing a status detection on each unit transmitter, and after confirming that the transmitter status is correct, checking whether the drone is correctly installed, and resetting the pitch mechanism of the transmitter, adjusting its angle to 90 degrees. Next, the drone is powered on and started in its working state. After its self-check process is completed, the launch angle of the drone is set and adjusted, and all drones ready to be launched are displayed on the launch display terminal. Thereafter, the drone insurance is released, the drone is launched into the air through the ignition device, enters the hovering mode to start the mission, and the collected data is transmitted to the ground station for further analysis of the drone's battery status and corresponding battery management. In this way, it is helpful to achieve efficient launch of drone clusters and improve the collaborative control capability and operational safety of drone clusters.

[0023] Specifically, Figure 1 FIG. 1 is a flow chart of a method for launching a drone cluster in an array according to an embodiment of the present application. Figure 1As shown, the method for array-type launching of drone clusters described in the present application includes: step 1: turning on the main power supply of the cluster transmitter; step 2: performing a status detection on each transmitter in the cluster transmitter to obtain a status detection result. If the status detection result is normal, executing step 3; if the status detection result is abnormal, ending the process and performing manual troubleshooting and maintenance; step 3: performing an in-situ detection on the drones to determine whether all drones are correctly installed in the launch tube of the transmitter to obtain an in-situ detection result. If the in-situ detection result is normal, executing step 4; if the in-situ detection result is abnormal, marking the abnormal drone number on the launch display terminal; step 4: resetting the pitch mechanism of each transmitter; step 5: starting the pitch mechanism of each transmitter until the pitch angle is equal to 90°. If the pitch angle reaches 90°, executing step 6; if the pitch angle does not reach 9 0°, then execute step 4 and then step 5 until the pitch angle is equal to 90°; step 6: power on each of the drones so that each of the drones is in working condition; step 7: self-check the flight control system and / or electronic speed regulator of each of the drones to obtain the self-check result. If the self-check result is passed, execute step 8; if the self-check result is not passed, mark the abnormal drone number on the launch display terminal; step 8: set the launch angle of each of the drones, and control the pitch mechanism of each transmitter to adjust the launch angle of each drone; step 9: display the drones to be launched on the launch display terminal; step 10: release the insurance of each of the drones to be launched, and ignite and launch to launch the drones to be launched into the sky, wherein the drones launched into the sky enter the hovering mode; step 11: the drones transmit the collected data back.

[0024] Specifically, when preparing to launch a drone cluster, you first need to turn on the main power supply of the cluster transmitter to ensure that all hardware components have the necessary power support. Then, perform a comprehensive status test on each unit transmitter in the cluster transmitter to verify whether each part is working properly. If the test results show that everything is normal, proceed to the next step; if an abnormality is found, the current process needs to be terminated and technicians will conduct manual inspection and repair to avoid potential risks.

[0025] After completing the status detection, the drones need to be tested in place to confirm whether all drones to be launched have been correctly installed in the launch tube. In this way, launch failures or safety hazards caused by improper installation can be effectively prevented. Once any problem is detected, the number of the abnormal drone will be marked on the launch display terminal to remind the operator to pay attention to it.

[0026] To ensure the consistency and accuracy of each launch, the pitch mechanism of each launcher needs to be reset to a known starting position, usually a horizontal position. Then the pitch angle is gradually adjusted until it reaches 90 degrees, that is, the vertical direction. This setting helps the drone to take off quickly, reduces the uncertainty caused by lateral movement, and facilitates the smooth implementation of subsequent hovering operations. Throughout the process, it is also necessary to continuously monitor the changes in the pitch angle to ensure that it reaches the predetermined value accurately.

[0027] With the pitch angle set, the drone needs to be powered on and put into standby mode. At this point, all the electronic equipment inside the drone, including the flight control system, sensors, etc., are activated and ready to receive further instructions. Subsequently, the drone's flight control system and electronic speed regulator are self-checked to ensure that these key components can operate normally after launch. If the self-check results indicate a fault, the corresponding drone number will be marked on the launch display terminal so that timely measures can be taken to solve the problem.

[0028] After all preparations are completed, the operator will check and confirm the information of the drone to be launched, such as the number and status, on the launch display terminal. This helps the operator understand the current situation and provides a basis for subsequent decision-making. Once confirmed, the safety lock mechanism of the drone is released and the ignition device is triggered to launch the drone. Once the drone is sent into the air, it will automatically switch to hover mode and wait for the next mission instruction.

[0029] After the drone is successfully launched, it begins its mission phase. During this period, the drone will perform various tasks according to preset programs or real-time instructions and collect relevant data. In particular, considering that when the drone is working at high altitudes, the ambient temperature of the surrounding environment may drop significantly, especially when the drone flies to higher altitudes or cold areas, the discharge efficiency and performance of the battery may be affected. Therefore, in order to avoid the degradation of battery performance due to low temperature, the present application further proposes a battery auxiliary management method. After receiving the ambient temperature data and battery discharge temperature data sent back by the drone, the method performs a time-series correlation interaction analysis on the two to determine whether the battery self-heating function needs to be turned on, thereby providing users with auxiliary decision-making to ensure the normal operation of the drone battery in a low temperature environment.

[0030] Figure 2 This is a flowchart of step S11 in the method for launching a drone cluster in an array according to an embodiment of the present application. Figure 3 FIG. 1 is a schematic diagram of data flow in step S11 of the method for launching a drone cluster in an array according to an embodiment of the present application. Figure 2 and Figure 3As shown, the step S11 includes the steps of: S110, receiving a time queue of the ambient temperature collected by the ambient temperature sensor, and a time queue of the battery discharge temperature collected by the battery temperature sensor; S120, performing a time series analysis on the time queue of the ambient temperature and the time queue of the battery discharge temperature to obtain an ambient temperature time series associated implicit coding vector and a battery discharge temperature time series associated implicit coding vector; S130, generating an auxiliary decision result for the battery management of the drone based on the time series associated interaction characteristics between the ambient temperature time series associated implicit coding vector and the battery discharge temperature time series associated implicit coding vector; S140, if the auxiliary decision result is to turn on the battery self-heating function, then pop up the auxiliary decision result and ask whether to confirm turning on the battery self-heating function; S150, after receiving a user confirmation instruction, send a command to start the battery self-heating function to the drone.

[0031] In the above-mentioned method for launching a cluster of drones in an array, the step S110 receives a time queue of the ambient temperature collected by the ambient temperature sensor, and a time queue of the battery discharge temperature collected by the battery temperature sensor. It should be understood that the temperature of the battery during operation (i.e., the discharge temperature) is an important indicator reflecting the health status and performance of the battery. If the discharge temperature is too low, it may affect the discharge efficiency and life of the battery. Since the battery discharge temperature is affected by the external ambient temperature, in order to accurately determine whether the self-heating function of the battery needs to be activated, the present application hopes to perform a time correlation analysis on the external ambient temperature and the battery discharge temperature to reveal the timing response characteristics of the battery discharge temperature relative to the external ambient temperature, understand the discharge performance of the battery at the ambient temperature, and thus provide accurate battery-assisted management results.

[0032] In the above-mentioned array-type launching method for launching a drone cluster, the step S120 performs a time series analysis on the time queue of the ambient temperature and the time queue of the battery discharge temperature to obtain an ambient temperature time series associated implicit coding vector and a battery discharge temperature time series associated implicit coding vector. In a specific example of the present application, the step S120 includes: inputting the time queue of the ambient temperature and the time queue of the battery discharge temperature into a sequence encoder based on an RNN-LSTM hybrid model to obtain the ambient temperature time series associated implicit coding vector and the battery discharge temperature time series associated implicit coding vector. That is, considering that the ambient temperature and the battery discharge temperature are continuously changing time series data, therefore, in order to fully extract the time series change characteristics of the ambient temperature and the battery discharge temperature, the present application adopts an RNN-LSTM hybrid model to construct a sequence encoder, and performs multi-scale time series analysis on the time queue of the ambient temperature and the time queue of the battery discharge temperature, respectively, to obtain an ambient temperature time series associated implicit coding vector and a battery discharge temperature time series associated implicit coding vector. It should be understood that the RNN model has a memory function by virtue of its internal cyclic connection structure, and can effectively process sequence data. However, the traditional RNN model may encounter the problem of gradient vanishing when processing long time series data, which limits its ability to learn long-distance dependencies. As a variant of the RNN model, the LSTM model can effectively solve the problem of gradient vanishing or gradient exploding when the traditional RNN processes long sequence data by introducing special gating structures (such as forget gate, input gate, and output gate), and better capture the long-term dependencies in time series data. Therefore, by introducing the LSTM model after the RNN model, the model's ability to model and analyze the time series of ambient temperature and battery discharge temperature can be significantly improved.

[0033] In the specific implementation, taking the time series of the ambient temperature as an example, the time series of the ambient temperature is first input into the RNN layer of the RNN-LSTM hybrid model, and is preliminarily processed through its cyclic connection structure to extract the local time series correlation features of the ambient temperature, and further pass it to the LSTM layer, which further analyzes the data to capture deeper time series dependencies. In this way, the short-term memory capacity of the RNN model and the ability of the LSTM model to capture long-term dependencies can be comprehensively utilized, thereby improving the accuracy and efficiency of time series analysis.

[0034] In the above-mentioned method for array-based launch of drone clusters, the step S130 generates an auxiliary decision result for drone battery management based on the time series correlation interaction characteristics between the ambient temperature time series correlation implicit coding vector and the battery discharge temperature time series correlation implicit coding vector. Figure 4FIG. 1 is a flow chart of step S130 in the method for launching a drone cluster in an array according to an embodiment of the present application. Figure 4 As shown, the step S130 includes: S131, performing fine-grained timing interaction analysis on the ambient temperature timing-associated implicit coding vector and the battery discharge temperature timing-associated implicit coding vector to obtain an ambient temperature-battery discharge temperature timing fine-grained interaction coding vector; S132, inputting the ambient temperature-battery discharge temperature timing fine-grained interaction coding vector into a classifier-based battery auxiliary management module to obtain the auxiliary decision result, and the auxiliary decision result is used to indicate whether to turn on the battery self-heating function.

[0035] Specifically, in step S131, the step performs fine-grained timing interaction analysis on the ambient temperature timing-associated implicit coding vector and the battery discharge temperature timing-associated implicit coding vector to obtain an ambient temperature-battery discharge temperature timing fine-grained interaction coding vector. That is, by performing timing interaction analysis on the ambient temperature timing-associated implicit coding vector and the battery discharge temperature timing-associated implicit coding vector, the timing response characteristics of the battery discharge temperature relative to the external ambient temperature are revealed, thereby determining whether the battery's self-heating function needs to be activated. In particular, in order to improve the accuracy of battery management decisions, the present application further introduces external knowledge, such as the physical and chemical principles of batteries, engineering specifications and standards, etc., in the timing interaction analysis process of the ambient temperature timing-associated implicit coding vector and the battery discharge temperature timing-associated implicit coding vector, so as to enhance the model's understanding of the complex relationship between ambient temperature and battery discharge temperature, and make more accurate battery management decisions accordingly.

[0036] Figure 5 FIG. 1 is a flow chart of step S131 in the method for launching a drone cluster in an array according to an embodiment of the present application. Figure 5 As shown, the step S131 includes: S1311, based on external knowledge, performing feature modulation optimization on the ambient temperature time series associated implicit coding vector and the battery discharge temperature time series associated implicit coding vector to obtain an optimized ambient temperature time series associated implicit coding vector and an optimized battery discharge temperature time series associated implicit coding vector; S1312, performing position-by-position timing interaction coding on the optimized ambient temperature time series associated implicit coding vector and the optimized battery discharge temperature time series associated implicit coding vector to obtain the ambient temperature-battery discharge temperature time series fine-grained interaction coding vector.

[0037] More specifically, the step S1311 includes: first, inputting the ambient temperature time series associated implicit coding vector and the battery discharge temperature time series associated implicit coding vector into a fine-grained feature interaction network to obtain an ambient temperature-battery discharge temperature time series interaction feature matrix, which is expressed as:

[0038] in, represents the ambient temperature time series associated implicit coding vector, represents the battery discharge temperature time series associated implicit coding vector, represents the transpose of a vector, represents the matrix multiplication operation, Represents the ambient temperature-battery discharge temperature time series interaction characteristic matrix.

[0039] That is, first, by performing fine-grained feature interaction on the ambient temperature time series associated implicit coding vector and the battery discharge temperature time series associated implicit coding vector, the correlation between the two can be understood from a microscopic level, the fine-grained interaction information between the two can be captured, and the ambient temperature-battery discharge temperature time series interaction feature matrix can be generated.

[0040] Next, the ambient temperature-battery discharge temperature timing interaction feature matrix is ​​input into the attention unit based on external knowledge to obtain the external knowledge optimized ambient temperature-battery discharge temperature timing interaction feature matrix, which is expressed as:

[0041] in, and represents the learnable memory parameter matrix of the external knowledge-based attention unit, represents the normalization function, Represents the external knowledge optimization of the ambient temperature-battery discharge temperature timing interaction feature matrix.

[0042] That is, external knowledge is used to further optimize the ambient temperature-battery discharge temperature timing interaction feature matrix to enhance the model's ability to understand the battery's discharge performance under the ambient temperature condition. In the specific implementation, the basic knowledge such as the technical parameters provided by the manufacturer, the working mechanism of the battery, the thermodynamic characteristics, and the effect of temperature on the internal chemical reaction rate of the battery are used as external knowledge to train the parameter matrix of the attention unit, and guide the attention unit to weight the ambient temperature-battery discharge temperature timing interaction feature matrix, so as to effectively identify and strengthen the important correlation interaction features between the ambient temperature and the battery discharge temperature, and capture the potential connection between the two in a larger range.

[0043] Then, based on the external knowledge, the ambient temperature-battery discharge temperature time series interaction feature matrix is ​​optimized, and the ambient temperature time series associated implicit coding vector and the battery discharge temperature time series associated implicit coding vector are respectively optimized by feature attention to obtain the optimized ambient temperature time series associated implicit coding vector and the optimized battery discharge temperature time series associated implicit coding vector, which are expressed as follows:

[0044]

[0045]

[0046]

[0047]

[0048]

[0049] in, and denote the first query embedding matrix and the first value embedding matrix respectively, and denote the first query feature vector and the first value feature vector respectively, and denote the second query embedding matrix and the second value embedding matrix respectively, and denote the second query feature vector and the second value feature vector respectively, , , and Represent different bias terms, represents the normalized exponential function, represents the characteristic scale value of the external knowledge optimized fine-grained feature interaction matrix, and They respectively represent the optimized ambient temperature time series associated implicit coding vector and the optimized battery discharge temperature time series associated implicit coding vector.

[0050] That is, the ambient temperature time series associated implicit coding vector is linearly transformed to obtain a first query feature vector and a first value feature vector, and the external knowledge is used to optimize the ambient temperature-battery discharge temperature time series interaction feature matrix as a key matrix, and the first query feature vector, the first value feature vector and the key matrix are input into a fine-grained modulation module based on a Transformer structure to obtain the optimized ambient temperature time series associated implicit coding vector; the battery discharge temperature time series associated implicit coding vector is linearly transformed to obtain a second query feature vector and a second value feature vector, and the external knowledge is used to optimize the ambient temperature-battery discharge temperature time series interaction feature matrix as a key matrix, and the second query feature vector, the second value feature vector and the key matrix are input into the fine-grained modulation module based on a Transformer structure to obtain the optimized battery discharge temperature time series associated implicit coding vector.

[0051] Here, the attention mechanism of the Transformer structure is used to realize the information exchange and integration between the internal and external knowledge of the features of the ambient temperature time series associated implicit coding vector and the battery discharge temperature time series associated implicit coding vector, which can ensure that the ambient temperature time series associated implicit coding vector and the battery discharge temperature time series associated implicit coding vector can benefit from external knowledge, help improve the quality of their feature expression, and enhance the accuracy of battery management decisions.

[0052] More specifically, the step S1312 includes: calculating the point-by-point division between the optimized ambient temperature time series associated implicit coding vector and the optimized battery discharge temperature time series associated implicit coding vector to obtain the ambient temperature-battery discharge temperature time series fine-grained interaction coding vector, which is expressed as:

[0053] in, A fine-grained interaction encoding vector representing the ambient temperature-battery discharge temperature time series.

[0054] That is, the optimized ambient temperature timing-related implicit coding vector and the optimized battery discharge temperature timing-related implicit coding vector are interactively encoded position by position by dividing them by position, so as to reflect the fine-grained timing-related response relationship between the battery discharge performance and the ambient temperature.

[0055] Specifically, in step S132, the ambient temperature-battery discharge temperature time series fine-grained interactive coding vector is input into the battery auxiliary management module based on the classifier to obtain the auxiliary decision result, and the auxiliary decision result is used to indicate whether to turn on the battery self-heating function. In a specific example of the present application, the classifier adopts a multi-layer perceptron structure, which learns the complex dynamic correlation between battery discharge performance and ambient temperature by performing multi-layer nonlinear mapping on the ambient temperature-battery discharge temperature time series fine-grained interactive coding vector. The output layer of the classifier calculates the decision probability of whether to turn on the battery self-heating function through the softmax function based on the learned feature representation, and determines the final decision result based on the size of the probability value, so as to ensure the safety and performance of the battery at different ambient temperatures and achieve accurate management of battery performance.

[0056] More specifically, the step S132 includes: using the fully connected layer of the battery auxiliary management module to fully connect the ambient temperature-battery discharge temperature timing interaction fine-grained interaction encoding vector to obtain an ambient temperature-battery discharge temperature timing interaction fully connected encoding feature vector; inputting the ambient temperature-battery discharge temperature timing interaction fully connected encoding feature vector into the Softmax classification function of the battery auxiliary management module to obtain the probability value of the ambient temperature-battery discharge temperature timing fine-grained interaction encoding vector belonging to each classification label, wherein the classification label includes on and off; and determining the classification label corresponding to the largest of the probability values ​​as the auxiliary decision result.

[0057] In the above-mentioned array-type launching method for launching a drone cluster, in the step S140 and the step S150, if the auxiliary decision result is to turn on the battery self-heating function, the auxiliary decision result is popped up, and it is asked whether to confirm to turn on the battery self-heating function. After receiving the user's confirmation instruction, the instruction to start the battery self-heating function is sent to the drone. That is to say, if the auxiliary decision result shows that the battery self-heating function does not need to be turned on, it means that the current temperature and discharge state of the battery are within the safe range and no additional heating is required. In this case, the battery management system will maintain the current state and continue to monitor the changes in ambient temperature and battery discharge temperature to ensure that the battery operates under safe temperature conditions. On the contrary, if the auxiliary decision result is to turn on the battery self-heating function, the auxiliary decision result is popped up to the user, and it is asked whether to confirm to turn on the battery self-heating function. After receiving the user's confirmation instruction, the instruction to start the battery self-heating function is sent to the drone to turn on the battery self-heating function to ensure that the battery can work normally in a low temperature environment and avoid battery performance degradation or damage caused by too low temperature.

[0058] In summary, a method for launching a drone cluster in an array-type manner based on an embodiment of the present application is explained, which includes: after turning on the main power supply of the cluster transmitter, performing status detection and in-situ detection on each unit transmitter, and after confirming that there is no error, resetting the pitch mechanism of the transmitter and adjusting its angle to 90 degrees. Next, the drone is powered on and started in its working state. After its self-check process is completed, the launch angle of the drone is set and adjusted, and all drones ready for launch are displayed on the launch display terminal. Thereafter, the drone insurance is released, the drone is launched into the air through the ignition device, the hovering mode is entered to start the mission, and the collected data is transmitted to the ground station, and the battery status of the drone is further analyzed based on the collected data and corresponding battery management is performed. In this way, it is helpful to achieve efficient launch of drone clusters and improve the collaborative control capability and operational safety of drone clusters.

[0059] Furthermore, the present application also provides a system for launching a drone cluster in an array. Figure 6 As shown, the array-type launching system for launching drone clusters includes a barrel-launched folding drone 1, which includes: a flight control unit for processing and controlling the drone; a battery unit for providing power to the drone system; a positioning module for positioning the drone during flight; an image transmission module for reconnaissance and returning the drone's image viewing angle; and a data transmission module for returning drone status parameters and flight data. Specifically, the barrel-launched folding drone 1 in this embodiment adopts an X-shaped aerodynamically distributed foldable multi-rotor drone; in other embodiments, a cruise missile folding drone or a coaxial double-propeller folding drone may also be used.

[0060] Cluster launcher 2, the cluster launcher includes: a launch tube, the inner side of the bottom of the launch tube wall is provided with an in-situ detection; a launch unit is installed at the bottom; a power-on unit is provided in the tube; in this embodiment, the launch tube is made of fiberglass; a pitch mechanism unit is used to control the launch angle of the launch tube; in this embodiment, the pitch mechanism unit adopts an electric push rod, and in other embodiments, a two-axis heavy-duty gimbal can be used; a camera is used to observe the state of the drone leaving the tube and the surrounding image environment during launch; an external detection module is used to sense the external environment; in this embodiment, it is specifically an anemometer and a rain gauge to avoid the failure of the drone mission execution due to excessive wind speed and heavy rain; a remote signal transceiver module is used to send and receive instructions; an external power supply is used to supply energy to the entire system; a monitoring module is used to monitor the operation of the system; a cluster launcher processing unit is used for calculation and processing.

[0061] Remote control unit 3, remote control of cluster transmitters.

[0062] The BMS management system (Battery Management System) 4 is used for battery management of the tube-launched folding UAV and cluster launcher. Specifically, the BMS management system includes: Intelligent Battery Indication: Designed with four bright LED indicators, these indicators automatically identify the status and indicate the power when discharging or charging. When the battery is off, short press the button, the LED indicates the battery level for about 2S and then goes out.

[0063] Battery life indication: In the battery off state (long press 5S or more), the power indicator shows the battery life indication. By default, the core can be used up to 500 times, and the LED light of the battery indication will count the number of times the user has used it, which shows that the battery life is still available.

[0064] Charging intelligent alarm: During the charging process, the battery status will be detected in real time, and the charging overvoltage, overcurrent and overheating will be warned. The alarm is displayed in red and white at the same time. If no operation is performed, the indicator light will automatically turn off when entering the sleep state to avoid power consumption. When the key is checked, the warning will be displayed again.

[0065] Low voltage or voltage difference alarm: When the battery is discharged, if a low voltage (less than 3.3V) is detected or the voltage difference of a single battery cell is greater than 50mV, the BMS will flash an alarm.

[0066] Power storage function: When the battery is not used for a long time, the battery is self-consumable. The battery with BMS management system will automatically activate the intelligent storage function and discharge to the storage voltage to ensure the safety of battery storage.

[0067] Smart sleep mode: When the smart battery is not in use, it will automatically enter sleep mode after 10 minutes to save battery power. If the user forgets to charge, when the single voltage is less than 3.58V, it will enter deep energy saving mode, which can ensure the safe storage of the battery for about 5 months. If it exceeds 5 months, for safety reasons, charging will be prohibited when it is not in use and the power supply is not supplemented to ensure safety.

[0068] Data communication module: It has I2C communication function and follows SMBUS protocol, and can obtain real-time battery information through communication. For example, current voltage, current, battery usage, etc. The flight control can also connect to this and communicate.

[0069] Battery logging module: The battery has a special logging function that can record data for the entire life of the battery. The battery log information contains individual voltage, current, battery temperature, number of cycles, number of abnormal states, etc. This log data can be viewed by connecting the battery via the transmitter display terminal.

[0070] Battery self-heating function: When the battery is in use, the BMS will automatically monitor the ambient temperature. When the ambient temperature is lower than the battery's own discharge temperature, it will automatically generate heat inside to bring the battery temperature to the flight temperature.

[0071] The launch display terminal 5 is used to display various information statuses, including: UAV status display, used to display UAV status parameters; cluster transmitter 2 status display, used to display cluster transmitter 2 status parameters; launch monitoring screen display, used to display the launch screen; switch signal indicator light, used to display the ignition status of the cluster transmitter 2 launch tube; UAV reconnaissance screen, used to display the status of the screen sent back by the UAV; launch data recording status, used to record the launch status and form a record.

[0072] In one embodiment of the present application, the tube-launched folding drone 1 is located in a launching tube of a cluster launcher 2; and the launching angles of the plurality of launching tubes are adjusted by a pitch mechanism unit.

[0073] In one embodiment of the present application, Figure 7 As shown, the cluster transmitter uses CAN bus communication.

[0074] In one embodiment of the present application, the power-on unit of the cluster transmitter 2 is adapted to four power-on modes, such as Figure 8 As shown, embodiment 1 of the power-on mode: the power-on unit adopts the point-touch type in the contact power-on mode; a contact point is provided at the bottom of the tube-launched folding UAV 1; a contact module is provided on one side of the launching tube; the tube-launched folding UAV is loaded into the launching tube, and the contact point is connected to the contact module.

[0075] Embodiment 2 of the power-on mode: the power-on unit adopts the wiring type in the contact power-on mode; one end of the wiring extends out from the bottom of the tube-launched folding drone 1; the wiring passes through the bottom launching unit to connect to the external control power-on; when the tube-launched folding drone 1 is launched, the wiring is pulled and retained in the launching tube.

[0076] Embodiment 3 of the power-on mode: the power-on unit adopts a non-inductive coil power-on mode; a large inductor is connected to the switch of the tube-launched folding drone 1; an electromagnetic coil is added to the same plane as the inner wall of the launching tube and the inductor; the horizontal distance between the electromagnetic coil and the inductor is 10 to 20 mm; when the electromagnetic coil is energized, the inductor generates an induced current, the current drives the MOS tube to turn on, and the drone is powered on.

[0077] Embodiment 4 of the power-on mode: the power-on unit adopts a mechanical power-on mode; a square hole is opened on the side of the launching tube; the switch of the tube-launched folding UAV 1 is located in the square hole, and the push rod is driven by the external servo to press the switch to power on the UAV.

[0078] In one embodiment of the present application, the battery self-heating function of the BMS management system is implemented by the following modules: a drone data receiving module, which is used to receive a time queue of the ambient temperature collected by the ambient temperature sensor and a time queue of the battery discharge temperature collected by the battery temperature sensor; a temperature data timing analysis module, which is used to perform timing analysis on the time queue of the ambient temperature and the time queue of the battery discharge temperature to obtain an ambient temperature timing-associated implicit coding vector and a battery discharge temperature timing-associated implicit coding vector; a battery management auxiliary decision module, which is used to generate an auxiliary decision result for drone battery management based on the timing-associated interaction characteristics between the ambient temperature timing-associated implicit coding vector and the battery discharge temperature timing-associated implicit coding vector; a user confirmation module, which is used to pop up the auxiliary decision result in response to the auxiliary decision result being to turn on the battery self-heating function, and ask whether to confirm turning on the battery self-heating function; a battery management control module, which is used to send an instruction to start the battery self-heating function to the drone after receiving a user confirmation instruction.

[0079] Here, those skilled in the art will appreciate that the specific operations of each module in the array-type launching UAV cluster launching system have been described in detail above. Figures 1 to 5 The description of the method for array-type launching of drone clusters has been introduced in detail, and therefore, its repeated description will be omitted.

[0080] In summary, a cluster launch system for array-type launch of unmanned aerial vehicles has been realized, which can deploy multiple tube-launched folding unmanned aerial vehicles 1 for cluster launch; the tube-launched folding unmanned aerial vehicles 1 and the cluster launcher 2 are equipped with a BMS management system 4 to effectively monitor the battery status and avoid battery failure; the cluster launcher 2 is equipped with a monitoring module and multiple sensors to reduce the launch failure rate; it is equipped with a visual and integrated launch display terminal 5, which can observe the launch status of the cluster launcher 2 and the real-time status of the tube-launched folding unmanned aerial vehicle 1 after power-on in real time. In addition, the remote control unit for the array-type launch of unmanned aerial vehicle cluster launch system has single-control release and cluster diagonal release modes to avoid mutual interference when the unmanned aerial vehicles unfold their arms, and can realize multiple power-on methods for deploying unmanned aerial vehicles in the tube, with the advantages of short take-off and deployment time, simple manual operation, and convenience and speed.

[0081] The basic principle of the present invention is described above in conjunction with specific embodiments. However, it should be pointed out that the advantages, strengths, effects, etc. mentioned in the present invention are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. must be possessed by each embodiment of the present invention. In addition, the specific details of the above embodiments are only for the purpose of illustration and facilitation of understanding, rather than limitation, and the above details do not limit the present invention to be implemented by adopting the above specific details.

Claims

1. A method for launching a cluster of drones in an array, characterized in that: include: Step 1: Turn on the main power of the cluster transmitter; Step 2: Performing status detection on each transmitter in the cluster transmitter to obtain a status detection result. If the status detection result is normal, executing step 3; If the status detection result is abnormal, the process is terminated and manual troubleshooting and repair are performed; Step 3: Perform an in-place detection on the drones to determine whether all drones are correctly installed in the launch tube of the launcher to obtain an in-place detection result. If the in-place detection result is normal, execute step 4. If the in-place detection result is abnormal, mark the abnormal drone number on the launch display terminal; Step 4: resetting the pitch mechanism of each transmitter; Step 5: Start the pitch mechanism of each transmitter until the pitch angle is equal to 90°. If the pitch angle reaches 90°, execute step 6; if the pitch angle does not reach 90°, execute step 4 and then step 5 until the pitch angle is equal to 90°; Step 6: Power on each of the drones so that each of the drones is in working state; Step 7: Perform a self-test on the flight control system and / or the electronic speed governor of each UAV to obtain a self-test result. If the self-test result is passed, execute step 8. If the self-test result is not passed, mark the abnormal UAV number on the transmitting display terminal; Step 8: Setting the launch angle of each drone, and controlling the pitch mechanism of each launcher to adjust the launch angle of each drone; Step 9: Displaying the UAV to be launched on the launch display terminal; Step 10: releasing the safety of each of the drones to be launched, and igniting and launching the drones to be launched into the sky, wherein the drones launched into the sky enter a hovering mode; Step 11: The drone transmits the collected data back.

2. The method for launching a drone cluster in an array according to claim 1, characterized in that: The step 11 comprises: receiving a time queue of an ambient temperature collected by an ambient temperature sensor and a time queue of a battery discharge temperature collected by a battery temperature sensor; Performing time series analysis on the time series of the ambient temperature and the time series of the battery discharge temperature to obtain an ambient temperature time series associated implicit coding vector and a battery discharge temperature time series associated implicit coding vector; Based on the time series correlation interaction characteristics between the ambient temperature time series correlation implicit coding vector and the battery discharge temperature time series correlation implicit coding vector, an auxiliary decision result for the drone battery management is generated; If the auxiliary decision result is to turn on the battery self-heating function, the auxiliary decision result is popped up and a user is asked whether to confirm turning on the battery self-heating function; After receiving the user's confirmation instruction, a battery self-heating function start instruction is sent to the drone.

3. The method for launching a drone cluster in an array according to claim 2, characterized in that: Performing time series analysis on the time series of the ambient temperature and the time series of the battery discharge temperature to obtain an ambient temperature time series associated implicit coding vector and a battery discharge temperature time series associated implicit coding vector, including: The time queue of the ambient temperature and the time queue of the battery discharge temperature are input into a sequence encoder based on the RNN-LSTM hybrid model to obtain an implicit coding vector associated with the ambient temperature time series and an implicit coding vector associated with the battery discharge temperature time series.

4. The method for launching a drone cluster in an array according to claim 3, characterized in that: Based on the time series correlation interaction characteristics between the ambient temperature time series correlation implicit coding vector and the battery discharge temperature time series correlation implicit coding vector, an auxiliary decision result for drone battery management is generated, including: Performing fine-grained time series interaction analysis on the ambient temperature time series associated implicit coding vector and the battery discharge temperature time series associated implicit coding vector to obtain an ambient temperature-battery discharge temperature time series fine-grained interaction coding vector; The ambient temperature-battery discharge temperature time series fine-grained interactive coding vector is input into a classifier-based battery auxiliary management module to obtain the auxiliary decision result, and the auxiliary decision result is used to indicate whether to turn on the battery self-heating function.

5. The method for launching a cluster of drones in an array according to claim 4, characterized in that: Performing fine-grained time series interaction analysis on the ambient temperature time series associated implicit coding vector and the battery discharge temperature time series associated implicit coding vector to obtain an ambient temperature-battery discharge temperature time series fine-grained interaction coding vector, including: Based on external knowledge, feature modulation optimization is performed on the ambient temperature time series associated implicit coding vector and the battery discharge temperature time series associated implicit coding vector to obtain an optimized ambient temperature time series associated implicit coding vector and an optimized battery discharge temperature time series associated implicit coding vector; The optimized ambient temperature time series associated implicit coding vector and the optimized battery discharge temperature time series associated implicit coding vector are subjected to position-by-position time series interactive coding to obtain the ambient temperature-battery discharge temperature time series fine-grained interactive coding vector.

6. The method for launching a cluster of drones in an array according to claim 5, characterized in that: Based on external knowledge, feature modulation optimization is performed on the ambient temperature time series associated implicit coding vector and the battery discharge temperature time series associated implicit coding vector to obtain an optimized ambient temperature time series associated implicit coding vector and an optimized battery discharge temperature time series associated implicit coding vector, including: Inputting the ambient temperature time series associated implicit coding vector and the battery discharge temperature time series associated implicit coding vector into a fine-grained feature interaction network to obtain an ambient temperature-battery discharge temperature time series interaction feature matrix; Inputting the ambient temperature-battery discharge temperature timing interaction feature matrix into an attention unit based on external knowledge to obtain the external knowledge optimized ambient temperature-battery discharge temperature timing interaction feature matrix; Based on the external knowledge, the ambient temperature-battery discharge temperature timing interaction feature matrix is ​​optimized, and the ambient temperature timing-associated implicit coding vector and the battery discharge temperature timing-associated implicit coding vector are respectively subjected to feature attention optimization to obtain the optimized ambient temperature timing-associated implicit coding vector and the optimized battery discharge temperature timing-associated implicit coding vector.

7. The method for launching a cluster of drones in an array according to claim 6, characterized in that: Optimizing the ambient temperature-battery discharge temperature time series interaction feature matrix based on the external knowledge, and performing feature attention optimization on the ambient temperature time series associated implicit coding vector and the battery discharge temperature time series associated implicit coding vector respectively to obtain the optimized ambient temperature time series associated implicit coding vector and the optimized battery discharge temperature time series associated implicit coding vector, including: Performing a linear transformation on the ambient temperature time series associated implicit coding vector to obtain a first query feature vector and a first value feature vector, and using the external knowledge to optimize the ambient temperature-battery discharge temperature time series interaction feature matrix as a key matrix, inputting the first query feature vector, the first value feature vector and the key matrix into a fine-grained modulation module based on a Transformer structure to obtain the optimized ambient temperature time series associated implicit coding vector; The battery discharge temperature timing associated implicit coding vector is linearly transformed to obtain a second query feature vector and a second value feature vector, and the external knowledge is used to optimize the ambient temperature-battery discharge temperature timing interaction feature matrix as a key matrix, and the second query feature vector, the second value feature vector and the key matrix are input into the Transformer-based fine-grained modulation module to obtain the optimized battery discharge temperature timing associated implicit coding vector.

8. The method for launching a cluster of drones in an array according to claim 7, characterized in that: The optimized ambient temperature time series associated implicit coding vector and the optimized battery discharge temperature time series associated implicit coding vector are subjected to position-by-position time series interactive coding to obtain the ambient temperature-battery discharge temperature time series fine-grained interactive coding vector, including: The optimized ambient temperature time series associated implicit coding vector and the optimized battery discharge temperature time series associated implicit coding vector are calculated and divided by the position points to obtain the ambient temperature-battery discharge temperature time series fine-grained interaction coding vector.

9. The method for launching a cluster of drones in an array according to claim 8, characterized in that: The ambient temperature-battery discharge temperature time series fine-grained interactive coding vector is input into a classifier-based battery auxiliary management module to obtain the auxiliary decision result, which is used to indicate whether to turn on the battery self-heating function, including: Using the fully connected layer of the battery auxiliary management module to perform fully connected encoding on the ambient temperature-battery discharge temperature time series fine-grained interactive encoding vector to obtain an ambient temperature-battery discharge temperature time series interactive fully connected encoding feature vector; Input the fully connected encoding feature vector of the ambient temperature-battery discharge temperature time series interaction into the Softmax classification function of the battery auxiliary management module to obtain the probability value of the ambient temperature-battery discharge temperature time series fine-grained interactive encoding vector belonging to each classification label, wherein the classification label includes on and off; The classification label corresponding to the largest probability value among the probability values ​​is determined as the auxiliary decision result.

10. A system for launching drone clusters in array form, characterized in that: include: The drone data receiving module is used to receive the time queue of the ambient temperature collected by the ambient temperature sensor and the time queue of the battery discharge temperature collected by the battery temperature sensor; A temperature data time series analysis module, used for performing time series analysis on the time series of the ambient temperature and the time series of the battery discharge temperature to obtain an ambient temperature time series associated implicit coding vector and a battery discharge temperature time series associated implicit coding vector; A battery management auxiliary decision module, used to generate an auxiliary decision result for the battery management of the drone based on the time series correlation interaction characteristics between the ambient temperature time series correlation implicit coding vector and the battery discharge temperature time series correlation implicit coding vector; A user confirmation module, for, in response to the auxiliary decision result being to enable the battery self-heating function, popping up the auxiliary decision result and asking whether to confirm enabling the battery self-heating function; The battery management control module is used to send a battery self-heating function start-up instruction to the drone after receiving a user confirmation instruction.

Images