Unmanned aerial vehicle position prediction method and device, equipment and storage medium
By continuously acquiring the location and flight speed of drones, identifying their flight patterns, and utilizing different models and wind speed and direction corrections, the problem of inaccurate drone location prediction has been solved, improving the accuracy and success rate of the countermeasure system.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-01-23
- Publication Date
- 2026-04-14
AI Technical Summary
Existing drone countermeasure systems struggle to accurately predict the location of drones when they are flying at high speeds, leading to the failure of countermeasure operations.
By continuously acquiring the drone's position and flight speed, identifying its flight mode, and predicting its future position based on its current position and speed, the system utilizes different models (uniform speed, uniform acceleration, and other flight modes) for prediction and considers wind speed and direction for correction.
This improves the accuracy of drone location prediction, enhances the effectiveness of the countermeasure system, and ensures the success rate of countermeasure operations.
Smart Images
Figure CN121857741A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of unmanned aerial vehicle (UAV) technology, specifically to a UAV location prediction method, apparatus, device, and storage medium. Background Technology
[0002] To counter drones, directional jamming signals, navigation decoy data, or high-energy lasers can be launched at their location to sever their communication links, disrupt their flight control systems, or physically destroy their key components, thereby achieving the purpose of countermeasures.
[0003] Because drones fly at high speeds, their real-time position may have changed significantly by the time the countermeasure system issues countermeasure commands and transmits signals. Therefore, countermeasures targeting only their real-time position are prone to failure due to system response delays and drone displacement. To successfully counter drones, it is possible to predict their position at a future moment and launch a countermeasure at that predicted position shortly before that moment, thus improving the success rate. Accurately predicting the drone's position is a problem that needs to be solved. Summary of the Invention
[0004] In view of the above problems, this application provides a method, apparatus, device, anti-drone system and storage medium for predicting the location of a drone, in order to solve the problem that the prior art cannot accurately predict the location of a drone.
[0005] According to one aspect of the embodiments of this application, a method for predicting the location of a drone is provided. The method includes: continuously acquiring state data of a first drone, wherein the state data includes the location and flight speed of the first drone; identifying the flight mode of the first drone based on historical state data of the first drone, wherein the historical state data is the state data acquired at a historical time; predicting the location of the first drone at a first moment based on the historical state data, the flight mode, and the current state data, thereby obtaining a predicted location, wherein the current state data is the state data acquired at the current moment, and the first moment is a future moment at a distance of a preset time Δt from the current moment.
[0006] In one optional approach, predicting the position of the first UAV at a first moment based on the historical state data, the flight mode, and the current state data, and obtaining the predicted position, includes: if the flight mode is uniform speed flight, inputting the current state data and the preset duration Δt into a first model to obtain the predicted position output by the first model; if the flight mode is uniform acceleration flight, determining the acceleration of the first UAV based on the historical state data, and inputting the current state data, the acceleration, and the preset duration Δt into a second model to obtain the predicted position output by the second model; if the flight mode is another flight mode besides uniform speed flight and uniform acceleration flight, inputting the historical state data, the current state data, and the preset duration Δt into a third model to obtain the predicted position output by the third model.
[0007] In one alternative approach, the method further includes: obtaining the wind speed and wind direction of the current flight environment of the first UAV; and correcting the predicted position based on the wind speed and the wind direction.
[0008] In an optional embodiment, the method further includes: in response to reaching the first time moment, acquiring first state data of the first UAV, wherein the first state data includes a first position and a first flight speed of the first UAV at the first time moment; determining a first deviation between the first position and a first predicted position, and a second deviation between the first position and a second predicted position, wherein the first predicted position is the predicted position before correction, and the second predicted position is the predicted position after correction; if the first deviation is less than or equal to the second deviation, then the steps of acquiring the wind speed and wind direction of the current flight environment of the first UAV and subsequent steps are not executed.
[0009] In one optional embodiment, the method further includes: controlling a second drone to fly toward the predicted location; determining the relative distance between the second drone and the first drone at the current moment; if the relative distance is less than or equal to a preset distance threshold, sending a shooting command to the second drone, so that after receiving the shooting command, the second drone can shoot the first drone through its camera device; if the relative distance is greater than the preset distance threshold, proceeding to the step of continuously acquiring the status data of the first drone.
[0010] In one alternative approach, determining the relative distance between the second drone and the first drone at the current moment includes: obtaining the second position of the second drone at the current moment; calculating a third distance between the second position and the predicted position as the relative distance; or, determining the relative distance between the second drone and the first drone at the current moment includes: obtaining the second position of the second drone and the third position of the first drone at the current moment; calculating a fourth distance between the second position and the third position as the relative distance.
[0011] According to another aspect of the embodiments of this application, a drone position prediction device is provided. The device includes: an acquisition module, configured to continuously acquire state data of a first drone, wherein the state data includes the position and flight speed of the first drone; an identification module, configured to identify the flight mode of the first drone based on historical state data of the first drone, wherein the historical state data is the state data acquired at a historical time; and a prediction module, configured to predict the position of the first drone at a first moment based on the historical state data, the flight mode, and the current state data, thereby obtaining a predicted position, wherein the current state data is the state data acquired at the current moment, and the first moment is a future moment at a distance of a preset time Δt from the current moment.
[0012] According to another aspect of the embodiments of this application, a drone position prediction device is provided, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the drone position prediction method as described above.
[0013] According to another aspect of the embodiments of this application, a drone countermeasure system is provided, including a second drone and a drone position prediction device as described above. The drone position prediction device is used to send a shooting command to the second drone. The second drone is used to receive the shooting command and control its camera device to shoot the first drone based on the received shooting command to obtain a first image, wherein the first image includes a first drone region. The second drone is also used to determine the position information of the first drone region in the first image, and determine whether the first drone region is located in a preset region in the first image based on the position information. If the first drone region is not located in the preset region in the first image, at least one of the heading angle, pitch angle, and flight speed of the second drone is adjusted based on the position information so that in the second image obtained by the camera device shooting the first drone, the first drone region is located in the preset region in the second image.
[0014] According to another aspect of the embodiments of this application, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the UAV position prediction method as described above.
[0015] In this embodiment, by continuously acquiring the position and flight speed of the first UAV, the flight mode of the first UAV is identified based on the flight speed of the first UAV at historical moments, and then the position of the first UAV at a future moment is predicted based on the flight mode, current position, and current flight speed of the first UAV. Compared with the method of predicting the position of the first UAV based solely on the position of the first UAV at historical moments, this application also makes full use of key dynamic information such as the change in the flight speed and flight mode of the first UAV, thereby effectively overcoming the lag and one-sidedness of single position information in reflecting the position change of the first UAV, and improving the accuracy of the predicted position.
[0016] The above description is merely an overview of the technical solutions of the embodiments of this application. In order to better understand the technical means of the embodiments of this application and to implement them in accordance with the contents of the specification, and to make the above and other objects, features and advantages of the embodiments of this application more obvious and understandable, specific implementation methods of this application are described below. Attached Figure Description
[0017] The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 A flowchart illustrating the UAV location prediction method provided in an embodiment of this application is shown. Figure 2 A flowchart illustrating another embodiment of the UAV location prediction method provided in this application is shown. Figure 3 The illustration shows an application scenario provided by an embodiment of this application; Figure 4 A flowchart illustrating another embodiment of the UAV location prediction method provided in this application is shown. Figure 5 A schematic diagram of the structure of the UAV position prediction device provided in an embodiment of this application is shown; Figure 6 A schematic diagram of the structure of the UAV position prediction device provided in an embodiment of this application is shown; Figure 7 A schematic diagram of the structure of the UAV countermeasure system provided in an embodiment of this application is shown; Figure 8 A schematic diagram of the first image, the first drone area, and the preset area provided in this application is shown. Detailed Implementation
[0018] Exemplary embodiments of the present application will now be described in more detail with reference to the accompanying drawings. Although exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application may be implemented in various forms and should not be limited to the embodiments set forth herein.
[0019] With the rapid development of drone technology, the use of drones has become very common. However, drones also bring a series of security risks, such as privacy violations and conflicts over flight areas. Therefore, effective detection and countermeasures against drones are extremely important.
[0020] Existing countermeasures against drones include launching capture nets, anti-aircraft missiles, or using lasers to directly destroy the drone's critical structures or electronic components, causing it to crash. Taking countermeasures using lasers as an example, the laser needs to be focused on the drone's critical structure and emitted at that location to destroy it, thus achieving countermeasures.
[0021] However, the aforementioned drone countermeasures rely on the drone's specific location to ensure their effectiveness. Due to the high speed of drones, their real-time location may have changed significantly during the transmission of countermeasure commands and signals. Therefore, targeting only their real-time location may be ineffective due to system response delays and drone displacement.
[0022] To successfully counter drones, it's possible to predict their location at a future moment and launch a countermeasure close to that predicted location, thus increasing the success rate. Therefore, accurately predicting a drone's location is crucial for successful countermeasures.
[0023] In order to accurately predict the location of a drone, this application proposes a drone location prediction method. By continuously acquiring the drone's location and flight speed, identifying the drone's flight mode based on the acquired flight speed, and then accurately predicting the drone's location at a future moment based on the drone's current location, current flight speed, and flight mode.
[0024] Figure 1This diagram illustrates a flowchart of a UAV location prediction method provided in an embodiment of this application. The method is executed by an electronic device, which may include one or more processors, such as a server, touchscreen phone, smartphone, tablet computer, or other electronic device. The processor may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application; no limitation is made herein. The one or more processors included in the electronic device may be of the same type, such as one or more CPUs; or they may be of different types, such as one or more CPUs and one or more ASICs; no limitation is made herein. Figure 1 As shown, the method includes the following steps 110 to 130.
[0025] Step 110: Continuously acquire the status data of the first UAV.
[0026] The first UAV refers to the UAV that needs to be tracked and countered. The status data of the first UAV includes its position and flight speed. The position of the first UAV can include its longitude, latitude, and altitude; the flight speed can include its vertical speed and horizontal speed.
[0027] Specifically, in this step, the position and flight speed of the first UAV can be collected and determined in real time by UAV detection equipment (such as radar, radio spectrum detector or protocol detector), and the position and flight speed of the first UAV determined at the current moment can be transmitted to the electronic device executing the embodiment of this application.
[0028] It is worth noting that, since the UAV detection equipment continuously collects and determines the position and flight speed of the first UAV and transmits the position and flight speed of the first UAV to the electronic device, in order to facilitate the electronic device to distinguish the position and flight speed of the first UAV at different times, in this application, preferably, the position and flight speed of the first UAV transmitted by the UAV detection equipment to the electronic device carries a timestamp. The timestamp is used to mark the collection time corresponding to each set of position and flight speed data, so that the electronic device can arrange and synchronize the received data sequence in chronological order according to the timestamp, thereby avoiding the problem of timing deviation that may occur during data transmission.
[0029] Step 120: Identify the flight mode of the first UAV based on its historical status data.
[0030] For ease of distinction, the moment when step 110 was most recently executed is referred to as the current moment, and moments earlier than the current moment are referred to as historical moments. In this step, the historical state data of the first UAV is the state data of the first UAV obtained through step 110 at the historical moment.
[0031] Since a drone may fly at a constant speed, accelerate at a constant speed, or use other flight modes, this application categorizes the flight modes of the first drone into three types: constant speed flight, accelerating at a constant speed flight, and other flight modes besides constant speed flight and accelerating at a constant speed flight. The flight speed characteristics of the drone corresponding to each of these three flight modes are predefined. This step determines the flight mode of the first drone from the aforementioned three flight modes.
[0032] Specifically, in this step, the historical state data of the first UAV is statistically analyzed to determine whether the flight speed of the first UAV in the historical period closest to the current moment conforms to the characteristics of uniform flight. If it does, the flight mode of the first UAV is determined to be uniform flight. If it does not, it is determined whether the flight speed of the first UAV in the historical period conforms to the characteristics of uniform acceleration flight. If it does, the flight mode of the first UAV is determined to be uniform acceleration flight. If it does not, the flight mode of the first UAV is determined to be another flight mode besides uniform flight and uniform acceleration flight.
[0033] Step 130: Based on historical status data, flight mode and current status data, predict the position of the first UAV at the first moment to obtain the predicted position.
[0034] The current state data refers to the state data acquired at the current moment, and the first moment refers to a future moment that is a preset time interval Δt away from the current moment.
[0035] In this step, specifically, if the flight mode is uniform speed flight, the current state data and the preset duration Δt are input into the first model to obtain the predicted position output by the first model. The current state data includes the current position P0 and current flight speed V0 of the first UAV. After inputting P0, V0, and Δt into the first model, the first model predicts the position P of the first UAV at the first moment using the following formula (1). pred .
[0036] P pred =P0+V0×Δt(1) If the flight mode is uniform acceleration, the acceleration of the first UAV is determined based on historical state data. The current state data, acceleration, and preset time Δt are input into the second model to obtain the predicted position output by the second model. Specifically, the horizontal and vertical velocity sequences of the first UAV at different historical moments within the preset time window are extracted from the historical state data. Then, the trend of the above velocity sequence relative to time is fitted using a linear fitting algorithm (such as the least squares method). The slope obtained by fitting is used as the acceleration A0 of the first UAV, or the velocity sequence is directly subjected to difference operation to calculate the rate of change of velocity at adjacent moments and the average value is taken, and the average value is used as the acceleration A0 of the first UAV. After inputting P0, V0, A0 and Δt into the second model, the second model predicts the position P of the first UAV at the first moment using the following formula (2). pred .
[0037] P pred =P0+V0×Δt+(1 / 2)×A0×Δt 2 (2) If the flight mode is any other than uniform speed flight mode and uniform acceleration flight mode, input the historical state data, current state data, and preset duration Δt into the third model to obtain the predicted position output by the third model. The third model can be a machine learning prediction model. After inputting the historical state data, current state data, and preset duration Δt into the third model, the third model predicts the position P of the first UAV at the first moment using the following formula (3). pred .
[0038] P pred =f(P) -n ..., P0, V -n ,……V0,Δt)(3) Where f is the trained neural network function, P -n ..., P0 represents the position of the first UAV in the historical and current state data; V -n ...V0 represents the flight speed of the first UAV in the historical and current state data.
[0039] In this embodiment, by continuously acquiring the position and flight speed of the first UAV, the flight mode of the first UAV is identified based on the flight speed of the first UAV at historical moments, and then the position of the first UAV at a future moment is predicted based on the flight mode, current position, and current flight speed of the first UAV. Compared with the method of predicting the position of the first UAV based solely on the position of the first UAV at historical moments, this application also makes full use of key dynamic information such as the change in the flight speed and flight mode of the first UAV, thereby effectively overcoming the lag and one-sidedness of single position information in reflecting the position change of the first UAV, and improving the accuracy of the predicted position.
[0040] Furthermore, this application adaptively matches different position prediction models (i.e., the first model, the second model, and the third model) based on the flight mode of the first UAV. For different flight modes, different models are used to predict the position of the first UAV, overcoming the limitations of a single prediction model under different flight modes and further improving the accuracy of the position of the first UAV predicted by the model.
[0041] Figure 2 A flowchart illustrating a UAV location prediction method according to another embodiment of this application is shown. Figure 2 As shown, the method includes the following steps 110 to 150.
[0042] Step 110: Continuously acquire the status data of the first UAV.
[0043] Step 120: Identify the flight mode of the first UAV based on its historical status data.
[0044] Step 130: Based on historical status data, flight mode and current status data, predict the position of the first UAV at the first moment to obtain the predicted position.
[0045] The principles and specific implementation methods of steps 110 to 130 can be found in [reference needed]. Figure 1 The provided embodiments will not be described in detail here.
[0046] Step 140: Obtain the wind speed and direction of the current flight environment of the first UAV.
[0047] It is understandable that wind speed in the flight environment of the first UAV will affect its actual flight speed. Therefore, in this step, the wind speed and direction of the first UAV's flight environment are obtained so that the predicted position can be corrected based on the wind speed and direction, thereby improving the accuracy of the predicted position of the first UAV based on its flight speed.
[0048] Specifically, in this step, meteorological sensors deployed in the flight area or meteorological measuring instruments integrated into the UAV detection equipment can be used to collect environmental airflow parameters in real time. Then, the raw airflow data in the sensor's local coordinate system is converted into horizontal and vertical wind speeds in the geographic coordinate system using a coordinate transformation algorithm. Vector synthesis and trigonometric functions are then used to calculate the current environmental wind speed and wind direction angle relative to true north. The determined wind speed and direction are then transmitted to the electronic device executing the embodiments of this application. The meteorological sensor can be an ultrasonic anemometer, a mechanical cup anemometer, or a thermal anemometer.
[0049] Step 150: Adjust the predicted location based on wind speed and direction.
[0050] Specifically, based on the obtained wind speed and direction, the horizontal and vertical wind speed components generated by the wind field on the first UAV are calculated using a vector synthesis and decomposition algorithm, thereby determining the position offset vector caused by the wind force acting on the first UAV at the first moment; then, the position offset vector is superimposed on the predicted position obtained in the aforementioned step 130, that is, the wind-induced displacement is superimposed on the latitude and longitude coordinates of the original predicted position, thereby realizing dynamic compensation and correction of the predicted position.
[0051] Because drones are directly affected by ambient airflow during flight, the aerodynamic force generated by the wind causes a deviation between their actual ground speed (speed relative to the ground) and air speed (speed relative to the air), resulting in their actual trajectory deviating from the theoretical flight path in windless conditions. Secondly, the superposition of different wind speeds and directions will change the drone's flight attitude and displacement rate, especially under crosswind or headwind conditions, where the drone's position drift is more significant.
[0052] Therefore, for Figure 2 The provided embodiment obtains the wind speed and wind direction of the current flight environment of the first UAV, introduces the environmental wind field as a key variable into the prediction process, and performs wind field compensation correction on the UAV's position change trend, thereby eliminating the prediction position error caused by meteorological interference and improving the accuracy and robustness of the prediction position.
[0053] for Figure 2In the provided embodiments, if the wind speed data collected by the wind speed sensor contains noise, or if overfitting occurs when correcting the predicted position, additional calculation errors may be introduced, causing the corrected predicted position to deviate more from the true value than the original predicted position. Therefore, in some embodiments, by determining whether the corrected predicted position deviates more from the actual value than the original predicted position, it is determined whether the predicted position should be corrected based on wind speed and direction when subsequently predicting the position of the first UAV, thereby improving the accuracy of the final predicted position. Specifically, in this embodiment, after step 150, the UAV position prediction method further includes the following steps a1 to a3.
[0054] Step a1: In response to the arrival of the first moment, acquire the first state data of the first UAV.
[0055] The first state data includes the first position and first flight speed of the first UAV at the first moment.
[0056] For ease of distinction, the predicted position obtained in step 130 is referred to as the first predicted position, and the predicted position corrected in step 150 is referred to as the second predicted position.
[0057] Step a2: Determine the first deviation between the first position and the first predicted position, and the second deviation between the first position and the second predicted position.
[0058] Specifically, the Euclidean distance between the first position and the first predicted position can be calculated to obtain the first Euclidean distance, and the first Euclidean distance can be determined as the first deviation; the Euclidean distance between the first position and the second predicted position can be calculated to obtain the second Euclidean distance, and the second Euclidean distance can be determined as the second deviation.
[0059] Step a3: If the first deviation is less than or equal to the second deviation, then step 140 and subsequent steps will not be executed.
[0060] In this step, the calculated first deviation and second deviation are compared numerically. If it is determined that the value of the first deviation is less than or equal to the second deviation, it means that the predicted position after correction based on wind speed and direction (i.e., the second predicted position) is not more accurate than the first predicted position. Therefore, step 140 and subsequent steps are not executed, meaning that the predicted position is no longer corrected based on wind speed and direction. If it is determined that the value of the first deviation is greater than the second deviation, it means that the predicted position after correction based on wind speed and direction (i.e., the second predicted position) is more accurate than the first predicted position. Therefore, step 140 and subsequent steps are executed, meaning that the predicted position obtained in step 130 is further corrected based on wind speed and direction to improve the accuracy of the final predicted position.
[0061] Besides launching capture nets, anti-aircraft missiles, and lasers at drones to counter them, friendly drones can also be used for escort countermeasures when dealing with high-threat drones that possess strong anti-jamming capabilities and operate in complex environments. Specifically, friendly drones can closely follow the target drone to achieve escort. If the target drone is detected to be performing illegal tasks, friendly drones can use close-range electronic jamming or navigation deception to counter it.
[0062] Figure 3 A schematic diagram illustrating an application scenario provided by an embodiment of this application is shown. For ease of explanation, this section uses an electronic device that executes the UAV position prediction method provided by an embodiment of this application as an example for description. Figure 3 As shown, the drone detection device 11 detects the first drone 12 in real time, collects its status data, and transmits it to the drone position prediction device 13 in real time. The drone position prediction device 13 predicts the position of the first drone 12 at a future moment, and then controls the second drone 14 (its own drone) to fly towards the predicted position, thus achieving escort flight of the first drone 12. Because the second drone 14 is close to the first drone 12, when the first drone 12 is detected to be performing an illegal mission or posing a threat to the control area, the second drone 14 can be used to counteract the first drone 12.
[0063] for Figure 3 Application scenarios provided Figure 4 A flowchart illustrating a UAV location prediction method according to another embodiment of this application is shown. Figure 4 As shown, the method is executed by the UAV position prediction device 13 and includes the following steps 210 to 270.
[0064] Step 210: Continuously acquire the status data of the first UAV 12.
[0065] Step 220: Identify the flight mode of the first UAV 12 based on the historical status data of the first UAV 12.
[0066] Step 230: Based on historical status data, flight mode and current status data, predict the position of the first UAV 12 at the first moment to obtain the predicted position.
[0067] Steps 210 to 230 are the same as steps 110 to 130. Therefore, the principle and specific implementation of steps 210 to 230 can be referred to steps 110 to 130, and will not be repeated here.
[0068] Step 240: Control the second UAV 14 to fly towards the predicted location.
[0069] The drone position prediction device 13 is communicatively connected to the second drone 14. The drone position prediction device 13 sends a flight command including the predicted position to the second drone 14, so that the second drone 14 flies to the predicted position after receiving the flight command.
[0070] Step 250: Determine the relative distance between the second UAV 14 and the first UAV 12 at the current moment.
[0071] Specifically, step 250 can be achieved through the following steps b1 to b2.
[0072] Step b1: Obtain the second position of the second drone 14 at the current moment.
[0073] Among them, after the second position of the second drone 14 is determined by the drone detection device 11, the second position is transmitted to the drone position prediction device 13.
[0074] Step b2: Calculate the third distance between the second position and the predicted position, and use the third distance as the relative distance.
[0075] Specifically, by calculating the third distance between the second position and the predicted position, it is possible to determine whether the first drone 12 can accompany the second drone 14 based on the third distance, and whether the first drone 12 is within the countermeasure range of the second drone 14. This allows for the determination of whether the second drone 14 can effectively counter the first drone 12 when it performs illegal tasks or poses a threat to the control area.
[0076] In some other embodiments, step 250 can also be implemented by the following steps c1 to c2.
[0077] Step c1: Obtain the second position of the second UAV 14 and the third position of the first UAV 12 at the current moment.
[0078] Specifically, after determining the second position of the second drone 14 and the third position of the first drone 12 through the drone detection device 11, the second and third positions can be transmitted to the drone position prediction device 13.
[0079] Step c2: Calculate the fourth distance between the second and third positions, and use the fourth distance as the relative distance.
[0080] Since the first UAV 12 and the second UAV 14 are in flight, their actual spatial positions change dynamically in real time. The actual flight trajectory of the first UAV 12 may deviate from the predicted trajectory, causing the relative distance calculated based on the predicted position to fail to accurately reflect the true relative spatial relationship between the two. Therefore, in this step, by calculating the fourth distance between the third position (i.e., the actual position) of the first UAV 12 and the second position of the second UAV 14, the model prediction error can be eliminated, thus enabling the relative distance to accurately reflect the current physical spatial interval between the two.
[0081] Step 260: Determine whether the relative distance is less than or equal to a preset distance threshold. If yes, proceed to step 270; otherwise, proceed to step 210.
[0082] If the relative distance is less than or equal to a preset distance threshold, it indicates that the distance between the second drone 14 and the first drone 12 is relatively close. In this case, the second drone 14 can accompany the first drone 12, and the process proceeds to step 270. If the relative distance is greater than the preset distance threshold, it indicates that the distance between the second drone 14 and the first drone 12 is relatively far. In this case, the second drone 14 cannot accompany the first drone, and the process proceeds to step 210. The flight speed of the first drone 12 is then acquired to re-predict its future position. The second drone 14 is then controlled to fly towards the latest predicted position of the first drone 12 until the relative distance between the second drone 14 and the first drone 12 is less than or equal to the preset distance threshold. The preset distance threshold can be determined as needed.
[0083] Step 270: Send a shooting command to the second drone 14.
[0084] When the second UAV 14 is escorting the first UAV 12, even if the relative distance between the second UAV 14 and the first UAV 12 is less than or equal to a preset distance threshold, if the relative distance between the second UAV 14 and the first UAV 12 is close to the preset distance threshold, the first UAV 12 may deviate from the optimal observation field of the second UAV 14, or the first UAV 12 may be located at the edge of the countermeasure range of the second UAV 14, which may easily cause the second UAV 14 to lose track of the first UAV 12, or the second UAV 14 to be unable to effectively counter the first UAV 12.
[0085] Therefore, in this application, when the relative distance between the second UAV 14 and the first UAV 12 is less than or equal to a preset distance threshold, the UAV position prediction device 13 sends a shooting command to the second UAV 14. Upon receiving the shooting command, the second UAV 14 uses its camera to capture a picture of the first UAV 12, obtaining a first image including the area of the first UAV. Based on the position information of the first UAV area in the first image, the second UAV 14 adjusts at least one of its heading angle, pitch angle, and flight speed to maintain an appropriate distance between the second UAV 14 and the first UAV 12, ensuring that the first UAV 12 is within its optimal observation range, i.e., within its optimal countermeasure range. The following will describe in detail how the second UAV 14 adjusts at least one of its heading angle, pitch angle, and flight speed based on the position information of the first UAV area in the first image in other embodiments.
[0086] Figure 5 A schematic diagram of the structure of the UAV position prediction device provided in an embodiment of this application is shown. Figure 5 As shown, the device 300 includes: an acquisition module 301, an identification module 302, and a prediction module 303.
[0087] The acquisition module 301 continuously acquires the status data of the first UAV, including its position and flight speed. The identification module 302 identifies the flight mode of the first UAV based on its historical status data, where the historical status data refers to the status data acquired at historical times. The prediction module 303 predicts the position of the first UAV at a given moment based on the historical status data, flight mode, and current status data, obtaining the predicted position. The current status data refers to the status data acquired at the current moment, and the first moment is a future moment at a preset time interval Δt from the current moment.
[0088] The UAV position prediction device 300 provided in this embodiment is used to execute the technical solution of the UAV position prediction method in the aforementioned method embodiment. Its implementation principle and technical effect are similar, and will not be described again here.
[0089] It is worth noting that the UAV position prediction device 300 provided in this embodiment also includes other modules for performing the steps of the above-described UAV position prediction method embodiment, which will not be described in detail here.
[0090] Figure 6 The diagram shows a structural schematic of the UAV position prediction device provided in the embodiment of this application. The specific embodiments of this application do not limit the specific implementation of the UAV position prediction device.
[0091] like Figure 6As shown, the drone location prediction device 13 may include a processor 132 and a memory 134.
[0092] The memory 134 is used to store the computer program 136. The memory 134 may include high-speed RAM, and may also include non-volatile memory, such as at least one disk storage device. The computer program 136 may include computer-executable instructions.
[0093] The processor 132 is used to execute the computer program 136 to implement the above-described UAV position prediction method embodiment.
[0094] Processor 132 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application. The one or more processors included in the UAV position prediction device 13 may be processors of the same type, such as one or more CPUs; or they may be processors of different types, such as one or more CPUs and one or more ASICs.
[0095] Figure 7 A schematic diagram of the structure of the drone countermeasure system provided in an embodiment of this application is shown. Figure 7 As shown, the drone countermeasure system 1 includes a drone position prediction device 13 and a second drone 14. The drone position prediction device 13 performs... Figure 4 In the provided embodiment, after sending a shooting command to the second drone 14, the second drone 14 receives the shooting command and uses its camera device to capture a picture of the first drone 12, thereby obtaining a first image. As described above, this first image includes the area of the first drone.
[0096] Furthermore, after obtaining the first image, the second drone 14 determines the position information of the first drone region within the first image. Based on this position information, it determines whether the first drone region is located within a preset area in the first image. If the first drone region is not located within the preset area, it adjusts at least one of the heading angle, pitch angle, and flight speed of the second drone based on the position information, so that in the second image subsequently captured by the camera device, the first drone region is located within the preset area in the second image. The preset area can be set as needed; preferably, it can be set as the center area of the image.
[0097] Figure 8A schematic diagram showing the first image, the first drone area, and the preset area provided in this application is illustrated. (See attached diagram.) Figure 8 As shown, the first image 2 includes a first drone region 21, and a preset region 22 is the central region of the first image 2.
[0098] Specifically, the second drone 14 can identify the first image 2 using a target detection algorithm (such as YOLO or SSD algorithm), thereby determining the first drone region 21 and obtaining the position information of the first drone region 21 in the first image 2. This position information can be the coordinate information of one or more pixels belonging to the first drone region 21 in the first image 2.
[0099] Specifically, a reference coordinate system can be established with the upper left corner point A of the first image 2 as the origin, and the coordinates of one or more pixels belonging to the first UAV region 21 in the reference coordinate system can be determined, thereby obtaining the position information of the first UAV region 21 in the first image 2.
[0100] For ease of explanation, the coordinates of the center point B of the first UAV region 21 in this reference coordinate system are used as the position information of the first UAV region 21 in the first image 2. To determine whether the first UAV region 21 is located within the preset region 22 in the first image 2 based on this position information, the coordinates of the center point C of the preset region 22 in a reference coordinate system established with point A as the origin can be determined, and the difference between the coordinates of point B and point C in this reference coordinate system can be determined. This difference can then be used to determine whether the first UAV region 21 is located within the preset region 22 in the first image 2. If the difference is zero, it means that the coordinates of point B and point C are the same, indicating that the first UAV region 21 is located within the preset region 22 in the first image 2. Alternatively, if the difference is small (e.g., less than a difference threshold), it also indicates that the first UAV region 21 is located within the preset region 22 in the first image 2.
[0101] If the difference between the coordinates of point B and the coordinates of point C is large (greater than or equal to the difference threshold), it indicates that the first UAV region 21 is not within the preset region 22 in the first image 2. In this case, the second UAV 14 adjusts at least one of its heading angle, pitch angle, and flight speed based on the difference, so that in the second image obtained by the subsequent camera device from the first UAV 12, the first UAV region is located within the preset region in the second image.
[0102] Specifically, determine the difference in x-coordinates between point B and point C. Then, the heading angle offset can be determined by the following formula (4). This allows the heading angle of the second UAV 14 to be adjusted based on the heading angle offset.
[0103] (4) in, The focal length of the camera device of the second drone 14, This is the horizontal field of view of the camera device.
[0104] Determine the difference in ordinates between point B and point C. Then, the pitch angle offset can be determined by the following formula (5). This allows the pitch angle of the second UAV 14 to be adjusted based on the pitch angle offset.
[0105] (4) in, This is the vertical field of view of the camera device.
[0106] If the area of the first drone region 21 is larger than the area of the preset region 22, it means that the current distance between the second drone 14 and the first drone 12 is too close. In this case, the flight speed of the second drone 14 can be reduced. Conversely, if the area of the first drone region 21 is much smaller than the area of the preset region 22, it means that the current distance between the second drone 14 and the first drone 12 is too far. In this case, the flight speed of the second drone 14 can be increased.
[0107] It should be noted that after receiving the shooting command, the second drone 14 takes pictures of the first drone 12 at preset intervals (e.g., 10s or 30s). For ease of distinction, in this application, if the current time is t1, the image taken at time t1 is called the first image, and the image taken at time t2 (i.e., a future time with a preset interval from t1) is called the second image. When time t2 is reached, the image taken at that time is the new first image.
[0108] for Figure 7 The provided drone countermeasure system 1 further introduces a visual feedback mechanism when the relative distance between the first drone 12 and the second drone 14 is less than or equal to a preset distance threshold. After the second drone 14 takes a picture of the first drone 12 and obtains a first image 2, the system uses the position information of the first drone region 21 in the first image 2 to adjust the flight attitude and speed of the second drone 14 in real time. This ensures that in the second image obtained by the subsequent camera, the first drone region is located in the preset area of the second image, thereby ensuring that the first drone 12 is always stably located in the center of the field of vision and within the countermeasure range of the second drone 14. This allows the second drone 14 to maintain precise escort of the first drone 12, and when the first drone 12 poses a threat to the control area, the second drone 14 can effectively counter the first drone 12.
[0109] Furthermore, it is understandable that if the second UAV 14 transmits the first image 2 to the UAV position prediction device 13, the UAV position prediction device 13 will analyze the first image 2 and determine the heading angle offset of the second UAV 14. Pitch angle offset After at least one of the following is obtained, the first image 2 is transmitted to the second UAV 14. Due to the high speed of the UAV, there is a communication delay in the transmission of the first image 2 and the return of the command. This can easily cause the adjustment command received by the second UAV 14 to lag behind the actual flight state of the first UAV 12, thus failing to eliminate the position deviation in time. This will reduce the real-time performance and accuracy of the escort flight, and may even lead to the risk of losing the target.
[0110] In this application, the second UAV 14 analyzes the first image 2 and adjusts the flight attitude and speed of the second UAV 14 based on the area of the first UAV region 21 in the first image 2, instead of transmitting the first image 2 to the UAV position prediction device 13 for analysis. This avoids communication delays caused by image data transmission and ensures that the second UAV 14 can respond quickly and make precise adjustments based on the real-time flight status of the first UAV 12. Thus, the second UAV 14 can accurately accompany the first UAV 12, effectively preventing tracking delays or target loss caused by data transmission delays.
[0111] Furthermore, for the drone countermeasure system 1, after the drone position prediction device 13 predicts the position of the first drone 12 at a certain future moment (i.e., the expected interception point), the second drone 14 is controlled to fly towards the predicted position. This allows the second drone 14 to actively intercept the first drone 12, rather than passively tailing it, greatly improving the efficiency of interception and escort. The system also combines the drone position prediction device 13's large-scale, coarse detection and position prediction with the second drone 14's image capture to adjust its flight attitude and speed based on the captured images (i.e., visual servo tracking mode) for small-scale, precise escort of the first drone 12. These two methods complement each other. Even if the signal from the drone position prediction device 13 is temporarily lost, the second drone 14 can still maintain tracking of the first drone 12 through visual lock, improving the robustness of the drone countermeasure system 1. Moreover, the system achieves full automation of the "detection-prediction-guidance-lock-tracking" process without manual intervention, offering a fast response time and making it suitable for security scenarios requiring rapid response. Moreover, the visual servo tracking mode of the second UAV 14 can achieve pixel-level precise control, with a short flight distance and good tracking stability, laying the foundation for subsequent actions such as driving away, forced landing, or capturing the first UAV 12.
[0112] This application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described UAV position prediction method embodiment.
[0113] This application provides a computer program that can be executed by a processor to implement the above-described UAV position prediction method embodiment.
[0114] This application provides a computer program product, which includes a computer program that, when executed by a processor, implements the above-described UAV position prediction method embodiment.
[0115] In the several embodiments provided in this application, any function, if implemented as a software functional module / unit and sold or used as an independent product, can be stored in a computer-readable storage medium. Based on this understanding, all or part of the technical solution of this application can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or other electronic device) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing computer program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0116] The algorithms or displays provided herein are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems can also be used in conjunction with the teachings herein. The required structure for constructing such systems is apparent from the above description. Furthermore, the embodiments of this application are not directed to any particular programming language. It should be understood that the content of this application described herein can be implemented using various programming languages, and the above description of specific languages is for the purpose of disclosing the best mode of implementation of this application.
[0117] It should be noted that the above embodiments are illustrative of this application and not restrictive, and those skilled in the art can devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. This application can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In claims enumerating several means, several units or modules of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names. The steps in the above embodiments, unless otherwise specified, should not be construed as limiting the order of execution.
[0118] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.
Claims
1. A method for predicting the location of a UAV, characterized in that, The method includes: Continuously acquire status data of the first drone, wherein the status data includes the position and flight speed of the first drone; Based on the historical state data of the first UAV, the flight mode of the first UAV is identified, wherein the historical state data is the state data acquired at a historical time. Based on the historical state data, the flight mode, and the current state data, the position of the first UAV at a first moment is predicted to obtain the predicted position. The current state data is the state data acquired at the current moment, and the first moment is a future moment that is a distance of a preset time Δt from the current moment.
2. The method according to claim 1, characterized in that, The step of predicting the position of the first UAV at a first moment based on the historical state data, the flight mode, and the current state data, and obtaining the predicted position, includes: If the flight mode is uniform speed flight mode, the current state data and the preset duration Δt are input into the first model to obtain the predicted position output by the first model; If the flight mode is uniform acceleration flight mode, the acceleration of the first UAV is determined according to the historical state data, and the current state data, the acceleration and the preset duration Δt are input into the second model to obtain the predicted position output by the second model; If the flight mode is any other than the uniform speed flight mode and the uniform acceleration flight mode, the historical state data, the current state data, and the preset duration Δt are input into the third model to obtain the predicted position output by the third model.
3. The method according to claim 1, characterized in that, The method further includes: Obtain the wind speed and direction of the current flight environment of the first UAV; The predicted location is corrected based on the wind speed and wind direction.
4. The method according to claim 3, characterized in that, The method further includes: In response to the arrival of the first moment, first state data of the first drone is acquired, wherein the first state data includes the first position and first flight speed of the first drone at the first moment; Determine a first deviation between the first position and the first predicted position, and a second deviation between the first position and the second predicted position, wherein the first predicted position is the predicted position before correction, and the second predicted position is the predicted position after correction; If the first deviation is less than or equal to the second deviation, then the steps of obtaining the wind speed and wind direction of the current flight environment of the first UAV and subsequent steps will not be executed.
5. The method according to claim 1, characterized in that, The method further includes: Control the second drone to fly towards the predicted location; Determine the relative distance between the second UAV and the first UAV at the current moment; If the relative distance is less than or equal to a preset distance threshold, a shooting command is sent to the second drone, so that after receiving the shooting command, the second drone can shoot the first drone through its camera device; If the relative distance is greater than the preset distance threshold, proceed to the step of continuously acquiring the status data of the first UAV.
6. The method according to claim 5, characterized in that, Determining the relative distance between the second drone and the first drone at the current moment includes: Obtain the second position of the second UAV at the current moment; Calculate a third distance between the second position and the predicted position, as the relative distance; or, Determining the relative distance between the second drone and the first drone at the current moment includes: Obtain the second position of the second drone and the third position of the first drone at the current moment; Calculate a fourth distance between the second position and the third position, as the relative distance.
7. A UAV position prediction device, characterized in that, The device includes: An acquisition module is used to continuously acquire the status data of the first UAV, wherein the status data includes the position and flight speed of the first UAV; The identification module is used to identify the flight mode of the first UAV based on the historical state data of the first UAV, wherein the historical state data is the state data acquired at a historical time. The prediction module is used to predict the position of the first UAV at a first moment based on the historical state data, the flight mode and the current state data, and obtain the predicted position. The current state data is the state data obtained at the current moment, and the first moment is a future moment that is a distance of a preset time Δt from the current moment.
8. A drone position prediction device, comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the UAV position prediction method according to any one of claims 1 to 6.
9. A countermeasure system for unmanned aerial vehicles (UAVs), characterized in that, Includes a second drone and a drone position prediction device as described in claim 8, wherein the drone position prediction device is used to send shooting instructions to the second drone; The second drone is used to receive the shooting command and control the camera device of the second drone to shoot the first drone based on the received shooting command to obtain a first image, wherein the first image includes the area of the first drone; The second drone is further configured to determine the position information of the first drone region in the first image, and determine whether the first drone region is located in a preset area in the first image based on the position information. If the first drone region is not located in the preset area in the first image, at least one of the heading angle, pitch angle and flight speed of the second drone is adjusted based on the position information so that in the second image obtained by the subsequent camera device capturing the first drone, the first drone region is located in the preset area in the second image.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the UAV position prediction method according to any one of claims 1 to 6.