Unmanned aerial vehicle positioning method and apparatus, computing device, storage medium, and program product

Abstract

Embodiments of the present application provide a UAV positioning method, device, computing equipment, storage medium and program product, wherein the UAV positioning method comprises: positioning based on collected current motion state information and / or environment information, determining a current reference position of the UAV; in the case that a signal of a target transponder is received by a transponder receiving device, determining a time point of passing through the target transponder and a target position of the target transponder, wherein the target transponder is any one of each trackside transponder laid along a track in a rail transit system; correcting the current reference position based on the time point of passing through the target transponder and the target position of the target transponder, obtaining target position information of the UAV. By combining the motion state of the UAV, the environment information and the accurate position of the trackside transponder, the position information of the UAV is corrected in real time, the positioning accuracy and reliability are improved, and it is especially suitable for the rail transit tunnel environment in which the positioning signal is unavailable.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of rail transportation technology, and in particular to a method, device, computing device, storage medium and program product for positioning an unmanned aerial vehicle (UAV). Background Art

[0002] With the rapid expansion of urban rail transit networks, ensuring the safety of infrastructure such as subway tunnels has become crucial. Drone inspections are playing an increasingly important role in rail transit maintenance due to their efficiency, low cost, and ability to reach hard-to-reach areas.

[0003] In related technologies, due to the limitation that satellite positioning signals cannot penetrate underground, drones mainly rely on inertial measurement, lidar positioning, visual cameras and other means for positioning. However, these means have limited accuracy and will accumulate errors, affecting inspection results and safety assessments.

[0004] Therefore, a more accurate UAV positioning method is urgently needed. Summary of the Invention

[0005] In view of this, embodiments of the present invention provide a method for positioning a drone. One or more embodiments of the present invention also include a drone positioning device, a computing device, a computer-readable storage medium, and a computer program product to address technical deficiencies in the prior art.

[0006] According to a first aspect of an embodiment of the present invention, a method for positioning a UAV is provided, wherein a transponder receiving device corresponding to a trackside transponder is deployed on the UAV, comprising:

[0007] Perform positioning based on the collected current motion state information and / or environmental information to determine the current reference position of the UAV;

[0008] determining a time point of passing the target transponder and a target position of the target transponder when the transponder receiving device receives a signal from the target transponder, wherein the target transponder is any one of the trackside transponders laid along the track in the rail transit system;

[0009] Based on the time point of passing the target transponder and the target position of the target transponder, the current reference position is corrected to obtain the target position information of the UAV.

[0010] According to a second aspect of an embodiment of the present invention, a UAV positioning device is provided. A transponder receiving device corresponding to a trackside transponder is deployed on the UAV, comprising:

[0011] A positioning module is configured to perform positioning based on the collected current motion state information and / or environmental information to determine the current reference position of the UAV;

[0012] a determination module configured to determine a time point of passing a target transponder and a target position of the target transponder when the transponder receiving device receives a signal from the target transponder, wherein the target transponder is any one of the trackside transponders laid along the track in the rail transit system;

[0013] The correction module is configured to correct the current reference position based on the time point of passing the target transponder and the target position of the target transponder to obtain the target position information of the UAV.

[0014] According to a third aspect of an embodiment of the present invention, there is provided a computing device, including:

[0015] memory and processor;

[0016] The memory is used to store computer programs / instructions, and the processor is used to execute the computer programs / instructions. When the computer programs / instructions are executed by the processor, the steps of the above-mentioned drone positioning method are implemented.

[0017] According to a fourth aspect of an embodiment of the present invention, a computer-readable storage medium is provided, which stores a computer program / instruction, which implements the steps of the above-mentioned drone positioning method when executed by a processor.

[0018] According to a fifth aspect of an embodiment of the present invention, a computer program product is provided, comprising a computer program / instruction, which implements the steps of the above-mentioned drone positioning method when executed by a processor.

[0019] One embodiment of the present invention realizes positioning based on the collected current motion state information and / or environmental information to determine the current reference position of the drone; when the transponder receiving device receives the signal of the target transponder, the time point of passing the target transponder and the target position of the target transponder are determined, wherein the target transponder is any one of the trackside transponders laid along the track in the rail transit system; based on the time point of passing the target transponder and the target position of the target transponder, the current reference position is corrected to obtain the target position information of the drone. The motion state and environmental information collected by the drone are used to determine the preliminary position, and when the trackside transponder signal is received, the position of the drone is corrected according to the precise position and time point of the transponder. This method not only improves the positioning accuracy, but also reduces the positioning error in a GPS-free environment such as a rail transit tunnel, thereby enhancing the safety and efficiency of the inspection task. Through real-time correction, it is ensured that the drone can fly accurately along the predetermined trajectory, thereby improving the reliability of the overall operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1This is a flow chart of a method for positioning a drone provided by one embodiment of the present invention;

[0021] Figure 2 This is a flowchart of a processing process of a drone positioning method provided by one embodiment of the present invention;

[0022] Figure 3 This is a schematic diagram of a scenario of a drone positioning method provided by an embodiment of the present invention;

[0023] Figure 4 This is a schematic structural diagram of a UAV positioning device provided by one embodiment of the present invention;

[0024] Figure 5 This is a structural block diagram of a computing device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0025] The following description sets forth numerous specific details to facilitate a thorough understanding of the present invention. However, the present invention can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without departing from the scope of the present invention. Therefore, the present invention is not limited to the specific implementations disclosed below.

[0026] The terms used in one or more embodiments of the present invention are for the purpose of describing specific embodiments only and are not intended to limit one or more embodiments of the present invention. The singular forms "a", "the" and "the" used in one or more embodiments of the present invention and the appended claims are also intended to include plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in one or more embodiments of the present invention refers to and includes any or all possible combinations of one or more associated listed items.

[0027] It should be understood that although the terms first, second, etc. may be used to describe various information in one or more embodiments of the present invention, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from each other. For example, without departing from the scope of one or more embodiments of the present invention, the first may also be referred to as the second, and similarly, the second may also be referred to as the first. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0028] In addition, it should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in one or more embodiments of the present invention are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation entrances must be provided for users to choose to authorize or refuse.

[0029] First, the terms involved in one or more embodiments of the present invention are explained.

[0030] Extended Kalman Filter (EKF): A recursive estimation algorithm suitable for nonlinear systems. By locally linearizing the nonlinear model at each step, the EKF extends the concept of Kalman filtering to nonlinear scenarios. It is widely used in areas such as drone navigation and target tracking.

[0031] Train transponders are electronic devices installed along the tracks of rail transit systems, providing precise location information and route data to trains or inspection equipment. When a train or drone passes by, the transponder sends a signal to an onboard receiver, helping to correct its position and improve positioning accuracy. This is particularly important in tunnels and complex environments where GPS signals are unavailable.

[0032] Inertial Measurement Unit (IMU): A sensor device used to measure an object's acceleration, angular velocity, and attitude. It typically consists of an accelerometer and gyroscope, and some IMUs also integrate a magnetometer. By monitoring motion in real time, IMUs provide essential navigation and positioning data for drones or robots, playing a particularly important role in GPS-denied environments.

[0033] LiDAR (Light Detection and Ranging): A remote sensing technology that measures the distance and shape of targets by emitting laser pulses and receiving reflected signals. LiDAR generates highly accurate three-dimensional environmental maps and is widely used in areas such as drone navigation, autonomous driving, and terrain mapping. It is particularly well-suited for obstacle detection and localization in complex environments.

[0034] With the rapid development of drone technology, its application in inspection, mapping, rescue, and other fields is becoming increasingly widespread. However, in complex and enclosed environments such as tunnels, drone positioning and navigation face numerous challenges. Due to a lack of significant landmarks, lack of GNSS (Global Navigation Satellite System) signal coverage, and limited wireless communication, traditional positioning technologies struggle to meet the requirements for high accuracy and robustness. To address these issues, researchers have proposed various solutions, but these solutions still face trade-offs between accuracy, cost, and stability.

[0035] Drone positioning in tunnels faces several key challenges: First, SLAM (Simultaneous Localization and Mapping) technology can fail due to the limited nature of the environment; second, the lack of GNSS signals prevents satellite positioning; and third, limited wireless communication makes remote control and positioning unreliable. Furthermore, while INS (Inertial Navigation System) can provide short-term high-precision positioning, long-term use can lead to accumulated errors, compromising overall positioning effectiveness.

[0036] A related technology addresses the positioning problem in tunnels by fusing inertial navigation with QR code visual navigation. This technology uses a camera to detect QR code markers placed within the tunnel, combined with gyroscope and accelerometer data, and Kalman filtering to achieve data fusion and error correction. However, this method is highly dependent on the QR code. If the QR code is damaged or the lighting conditions are poor, recognition performance will be significantly reduced. Furthermore, the spacing between QR codes is fixed. If a drone flies between two QR codes for a long time, inertial navigation errors may accumulate, affecting positioning accuracy.

[0037] Another related technology utilizes multi-sensor fusion, including binocular vision, IMU, lidar, and UWB (Ultra-Wideband) to achieve drone pose estimation and path planning. This technology uses Kalman filtering for denoising and maximum likelihood estimation to eliminate outliers, thereby improving positioning accuracy. LiDAR is also used to detect obstacles and adjust waypoints to ensure safe flight. However, this solution relies heavily on UWB, which has high deployment and maintenance costs and limited coverage, making it difficult to adapt to the needs of long-distance tunnel inspections.

[0038] In the present invention, a drone positioning method is provided. The present invention also relates to a drone positioning device, a computing device, a computer-readable storage medium, and a computer program product, which are described in detail one by one in the following embodiments.

[0039] See also Figure 1 , Figure 1 A flowchart of a method for positioning a UAV according to an embodiment of the present invention is shown, which specifically includes the following steps.

[0040] Step 102: Positioning is performed based on the collected current motion state information and / or environmental information to determine the current reference position of the UAV.

[0041] It's important to note that current motion information typically includes data such as the drone's speed, acceleration, and attitude, while environmental information refers to surrounding features captured by sensors like LiDAR and cameras. Positioning refers to determining the specific location of a drone within a specific coordinate system.

[0042] In actual implementation, the system first uses the IMU to collect the drone's acceleration and angular velocity information, and then integrates it to obtain velocity and position estimates. Simultaneously, the LiDAR scans the surrounding environment to generate point cloud data, which is used to identify fixed structures or feature points in the environment. The camera captures images, extracts features using a visual SLAM algorithm, and matches historical data to determine relative displacement. Regarding positioning based on the collected information in this step, one implementation method is to directly fuse the inertial navigation data provided by the IMU with the LiDAR and visual information. Another implementation method is to update the position estimate only when sufficiently reliable environmental feature points are detected.

[0043] Specifically, imagine a drone performing an inspection mission inside a tunnel. The IMU continuously provides high-frequency position updates, but its long-term accuracy is limited due to accumulated errors. In this case, when the drone approaches a pre-placed transponder, the system can accurately determine the drone's arrival at a specific location by analyzing the transponder signal and adjust the IMU's calculated position accordingly. Simultaneously, the LiDAR and vision systems continuously work to identify texture changes or other significant features on the tunnel wall, helping the system more accurately track the drone's movement.

[0044] Step 104: When the transponder receiving device receives the signal of the target transponder, determine the time point of passing the target transponder and the target position of the target transponder, wherein the target transponder is any one of the trackside transponders laid along the track in the rail transit system.

[0045] It should be noted that the transponder receiver is a device mounted on a drone that receives signals from trackside transponders. Target transponders are devices installed along the tracks in rail transit systems to identify specific locations. Each transponder has unique coordinates that provide a precise location reference.

[0046] In actual implementation, when a drone approaches a trackside transponder, its onboard transponder receiver can receive the signal emitted by the transponder. The system analyzes the received signal to determine the exact time the drone passed the transponder and retrieves the transponder's specific location information from a pre-stored transponder location database. Regarding the time point of passing the target transponder in the step, one implementation method is to directly record the time of signal reception as the passing time point; another implementation method is to estimate the closest time point by analyzing the changing trend of signal strength, thereby improving time accuracy.

[0047] Specifically, imagine a drone performing a tunnel inspection. When it approaches a trackside transponder, the transponder receiver begins receiving signals from it. The system immediately records the current moment as the time the drone passed the transponder and queries a pre-stored database of transponder locations to obtain the transponder's precise coordinates. This enables the system to obtain highly accurate position updates even in the absence of GNSS signals, ensuring the accuracy and reliability of inspections.

[0048] Step 106: Based on the time point of passing the target transponder and the target position of the target transponder, the current reference position is corrected to obtain the target position information of the UAV.

[0049] It should be noted that the time point of passing the target transponder refers to the exact moment when the drone's receiver receives the signal from a specific trackside transponder, while the target position of the target transponder is the known fixed coordinates of the transponder in the rail transit system. The current reference position is an estimated value of the drone's position obtained by fusing data from multiple sources such as IMU, LiDAR, and vision sensors.

[0050] In actual implementation, once the drone detects the target transponder signal through its transponder receiving device and determines the time point and specific location of the transponder passing through the transponder, the system can use this information to correct the current reference position. The system will match the most recent position update based on the timestamp and correct the drone's position estimate in combination with the absolute coordinates of the transponder. For the position correction based on the information passing through the target transponder in the step, one implementation method is to directly replace the current position estimate with the coordinates provided by the transponder; another implementation method is to use the extended Kalman filter (EKF) algorithm to fuse the newly acquired precise position information with the existing position estimate to optimize the overall position estimate of the drone.

[0051] Specifically, in a tunnel inspection scenario, suppose a drone approaches a trackside transponder during its mission and successfully receives its signal. The system then records the time the signal was received and queries the database to obtain the transponder's precise coordinates. The system then applies the EKF algorithm, combining the newly acquired transponder position information with previously collected data from the IMU, LiDAR, and camera, to perform real-time corrections to the drone's position. This not only improves the accuracy of position estimation but also effectively reduces the potential cumulative errors that can occur during long-term flights.

[0052] In an optional implementation of this embodiment, the target position information includes the current corrected position of the UAV; based on the time point of passing the target transponder and the target position of the target transponder, the current reference position is corrected to obtain the target position information of the UAV, including:

[0053] Based on the time point of passing the target transponder, the current motion state information and the current reference position, the current reference pose is constructed. The current reference pose is used to describe the position coordinates and motion state of the UAV at that point in time.

[0054] Based on the target position of the target transponder and the current reference pose, the current reference pose is corrected to obtain the current corrected pose.

[0055] It should be noted that the current corrected pose refers to the drone's state matrix after correction, including its position coordinates and motion state (such as speed, angular velocity, etc.), while the current reference pose is the drone's position coordinates and motion state at a specific point in time estimated based on IMU and other sensor data. The target position of the target transponder refers to the known fixed coordinates of the transponder in the rail transit system.

[0056] In actual implementation, after the UAV passes by the target transponder and receives its signal, the system first constructs the current reference pose based on the time point of receiving the signal, combined with the current motion state information (such as speed, angular velocity, attitude, acceleration, etc.) and the previous reference position information. In this step, the system uses the high-frequency data provided by the IMU to infer the position coordinates and motion state of the UAV at that point in time. Then, the system will correct the current reference pose based on the precise position information of the target transponder to obtain a more accurate current corrected pose. For the pose correction based on the information passing through the target transponder in the step, one way to implement it is to directly use the transponder position to update the current position; another way is to fuse multi-source data through the extended Kalman filter (EKF) algorithm to optimize the UAV's position estimate.

[0057] Specifically, in a tunnel inspection mission scenario, suppose a drone approaches and successfully receives a signal from a trackside transponder. The system records the exact time the signal is received and constructs the current reference pose based on the speed, angular velocity and other parameters provided by the IMU and the previous position estimate. Subsequently, the system adjusts the current reference pose based on the precise position information of the transponder to obtain a more accurate current corrected pose. For example, if a drone moves at a certain speed and direction and passes a transponder, the system will correct the drone's position and motion state based on the specific location of the transponder to ensure the accuracy of subsequent navigation.

[0058] In the embodiments of this specification, by combining the absolute position information provided by the transponder with the drone's own sensor data, the system can effectively and accurately correct the drone's position and posture, significantly improving the accuracy and stability of positioning. This method not only improves the drone's autonomous navigation capabilities in complex environments, but also reduces the cumulative errors that may occur during long-term flights, ensuring the safe and efficient execution of inspection tasks. At the same time, this strategy enhances the robustness of the system, allowing the drone to maintain high positioning accuracy in changing environments, further improving the quality and efficiency of task completion.

[0059] In an optional implementation of this embodiment, constructing a current reference pose based on the time point of passing the target transponder, the current motion state information, and the current reference position includes:

[0060] Determining a historical time point adjacent to the time point based on the time point of passing the target transponder;

[0061] Get the historical pose corresponding to the historical time point;

[0062] Obtain state transition information, control input information, and process noise information. State transition information is used to describe the posture change from the current time point to the subsequent time point. Control input information is the attitude control information input to the UAV. Process noise information is used to quantify the uncertainty of the posture change process.

[0063] Based on the current motion state information, state transition information, control input information and process noise information, the current reference pose is predicted.

[0064] It should be noted that historical pose refers to the state matrix of the drone at the previous time point, including position coordinates and motion state. State transition information is used to describe the state transition matrix used to change the pose from the current time point to the subsequent time point. Control input information is the attitude control information based on the IMU, and process noise information quantifies the uncertainty in the pose change process caused by errors in the LiDAR, IMU, and vision sensors.

[0065] In actual implementation, after the drone passes the target transponder and records the exact time point of receiving the signal, the system first determines the historical time point adjacent to this time point. The system then obtains the historical pose corresponding to the historical time point. Next, the system collects state transition information (i.e., the state transition matrix), control input information (i.e., attitude control information provided by the IMU), and process noise information (used to describe sensor errors). For the construction of the current reference pose in step 1, one implementation method is to directly use IMU data to predict the position at the next moment; another implementation method is to combine the extended Kalman filter (EKF) algorithm to comprehensively consider state transitions, control inputs, and process noise to make a more accurate prediction of the drone's pose.

[0066] Specifically, in a tunnel inspection mission scenario, assume that the drone receives a signal from a trackside transponder at a specific point in time. The system finds the most recent historical time point based on that time point and obtains the historical pose at that time point. Next, the system uses the acceleration and angular velocity data provided by the IMU to calculate the state transition information, and combines it with the drone's attitude control instructions as control input information. At the same time, the system evaluates the errors that may be introduced by the LiDAR, IMU, and vision sensors to form process noise information. Based on this information, the system calculates the current reference pose through a prediction model, which reflects the position and motion state of the drone at the moment the transponder signal is received.

[0067] In the embodiments of this specification, by combining the timestamp of the transponder, the historical pose and various information provided by the sensor, the system can construct the current reference pose of the drone.

[0068] In an optional implementation of this embodiment, based on the target position and the current reference pose of the target transponder, correcting the current reference pose to obtain the current corrected pose includes:

[0069] Calculate the difference between the target position of the target transponder and the current reference position;

[0070] Obtaining a relative trust parameter, wherein the relative trust parameter is used to describe the degree of correction of the current reference pose;

[0071] Based on the relative trust parameter and the difference information, first correction information is calculated and obtained, wherein the first correction information is used to describe the correction degree of the current reference posture;

[0072] The current reference posture is corrected based on the first correction information to obtain a current corrected posture.

[0073] It should be noted that the relative trust parameter, also known as the Kalman gain, describes the degree of adjustment to the current reference pose during the correction process. The difference information refers to the difference between the target transponder's target position and the drone's current reference position. The first correction information is a matrix calculated using the Kalman gain and the relative difference. This matrix describes the degree of correction required for the current reference pose. The current corrected pose is the corrected state matrix of the drone at the current time.

[0074] In actual implementation, after determining the exact location of the target transponder and the current reference position of the drone, the system first calculates the difference between the two. The system then obtains the Kalman gain as a relative trust parameter, which determines the extent to which the target transponder position information is used to correct the current reference pose. Based on this relative trust parameter and the difference information, the system calculates the first correction information. Regarding the correction based on the difference information in step 1, one implementation method is to directly apply the calculated first correction information to update the current position estimate; another implementation method may involve more complex filtering algorithms (such as extended Kalman filter EKF) to further optimize the correction results.

[0075] Specifically, in a tunnel inspection mission scenario, assume that the drone has approached and received a signal from a trackside transponder, and the system has established the current reference pose. At this point, the system will calculate the difference between the position of the target transponder and the current drone reference position, and determine how to use this difference information to adjust the drone's position estimate based on the Kalman gain. For example, if there is a certain deviation in the drone's current position estimate, the system can adjust this deviation through the Kalman gain to generate the first correction information, thereby accurately correcting the current reference pose and obtaining a more accurate current corrected pose.

[0076] In the embodiments of this specification, by combining the high-precision position information provided by the target transponder with the current reference position of the UAV, and using relative trust parameters and difference information to calculate the first correction information, the system can effectively and accurately correct the posture of the UAV.

[0077] In an optional implementation of this embodiment, obtaining the relative trust parameter includes:

[0078] Obtaining covariance prediction information corresponding to a time point, where the covariance prediction information is used to describe prediction uncertainty;

[0079] Obtain observation information and observation noise parameters, where the observation information is used to convert pose into position information, and the observation noise parameters are used to describe the uncertainty of the observation information;

[0080] Based on the observation information, the covariance prediction information is mapped to the observation space to obtain the predicted observation covariance information, wherein the predicted observation covariance information is used to describe the result after converting the uncertainty in the state space to the observation space;

[0081] Based on the prediction observation covariance information and observation noise parameters, the adjustment weight calculation is performed to obtain the relative trust parameter.

[0082] It should be noted that the covariance prediction information refers to the uncertainty description matrix obtained based on the state prediction at the current time point, which is used in the subsequent calculation of the Kalman gain. The observation information, namely the observation matrix, converts the pose into position information. The observation noise parameter describes the uncertainty that may exist in the observation process. The predicted observation covariance information is obtained by mapping the uncertainty in the state space (the covariance prediction matrix) into the observation space. Specifically, it is the measurement matrix multiplied by the covariance prediction matrix, which is then multiplied by the transpose of the observation matrix.

[0083] In actual implementation, after determining the position of the target transponder and the current reference position of the drone, the system first obtains the covariance prediction information corresponding to that point in time, which describes the uncertainty of the current prediction state. Next, the system obtains the observation information (observation matrix) and the observation noise parameter, where the observation information is used to convert the drone's posture into position information that can be compared with the target transponder position, and the observation noise parameter quantifies the uncertainty in this conversion process. Subsequently, the system uses the observation information to map the covariance prediction information to the observation space, thereby obtaining the predicted observation covariance information. Based on this predicted observation covariance information and the observation noise parameter, the system adjusts the weight calculation and finally obtains the relative trust parameter (i.e., Kalman gain), which determines the degree of trust in the new observation data during the correction process.

[0084] Specifically, in a tunnel inspection mission, suppose a drone approaches and receives a signal from a trackside transponder. At this point, the system predicts the drone's state and its covariance prediction matrix based on data from the IMU and other sensors. Next, the system uses the observation matrix to convert the drone's state into position information and evaluates the uncertainty in the conversion process in combination with the observation noise parameter. The system then uses the observation matrix to map the covariance prediction information to the observation space, generating predicted-observation covariance information. Finally, based on this information, the system calculates the Kalman gain, which serves as a relative trust parameter to adjust the drone's pose estimate and ensure the accuracy of the positioning results.

[0085] In the embodiments of this specification, by comprehensively considering the covariance prediction information, observation information and observation noise parameters, the system can accurately calculate the relative trust parameters, thereby effectively correcting the posture of the drone.

[0086] In an optional implementation of this embodiment, obtaining covariance prediction information corresponding to a time point includes:

[0087] Determining a historical time point adjacent to the time point based on the time point of passing the target transponder;

[0088] Obtain state transition information, process noise parameters, and historical covariance information corresponding to historical time points, wherein the process noise parameters are used to describe the uncertainty of state transition information;

[0089] Predictions are made based on historical reference covariance information, state transition information, and process noise parameters to obtain covariance prediction information corresponding to a time point.

[0090] It should be noted that historical covariance information refers to the matrix describing the uncertainty of the drone's state estimate at the previous time point, while state transition information is the matrix used to describe the state change from the previous time point to the current time point. The process noise parameter quantifies the uncertainty introduced by factors such as sensor errors and environmental interference during the state transition process. Covariance prediction information is calculated based on historical covariance information, state transition information, and process noise parameters and is used to describe the uncertainty of the state prediction at the current time point.

[0091] In actual implementation, after the drone passes the target transponder and records the time of receiving the signal, the system first determines the historical time points adjacent to this time point and obtains the corresponding historical covariance information. Next, the system collects state transition information (i.e., the state transition matrix) and process noise parameters, which reflect the state changes and their uncertainty from the historical time point to the current time point. The system then uses the historical covariance information, state transition information, and process noise parameters to make predictions, thereby obtaining the covariance prediction information corresponding to the current time point. For the predicted covariance information in this step, one implementation method is to directly apply the state transition matrix to update the historical covariance information; another implementation method is to combine it with the extended Kalman filter (EKF) algorithm to further optimize the prediction results.

[0092] Specifically, in a tunnel inspection mission scenario, suppose the drone receives a signal from a trackside transponder at a specific time. Based on that time, the system finds the most recent historical time point and obtains the historical covariance information at that time point. Next, the system uses the acceleration and angular velocity data provided by the IMU to calculate state transition information and, combined with process noise parameters, evaluates the uncertainty in the state transition process. Based on this information, the system updates the historical covariance information using the state transition matrix, ultimately obtaining the covariance prediction information for the current time point. This step provides an important uncertainty description for the subsequent calculation of the Kalman gain and correction of the drone's position.

[0093] In the embodiments of this specification, by combining historical covariance information, state transition information and process noise parameters, the system can accurately predict the covariance prediction information at the current time point, thereby effectively describing the uncertainty of state prediction.

[0094] In an optional implementation of this embodiment, after adjusting the weight calculation based on the prediction observation covariance information and the observation noise parameter to obtain the relative trust parameter, the following is further included:

[0095] Obtaining observation information, and determining second correction information based on the observation information and the relative trust parameter, wherein the second correction information is used to describe a degree of correction of the covariance prediction information;

[0096] The predicted observation covariance information is corrected based on the second correction information to obtain reference covariance information corresponding to the time point.

[0097] It should be noted that the second correction information is calculated as the difference between the identity matrix I and the product of the Kalman gain (i.e., the relative confidence parameter) and the observation matrix. It describes the degree of correction required for the covariance prediction information. The reference covariance information corresponding to a time point is the corrected covariance prediction information and is used to predict the covariance prediction matrix at the next time point, thereby supporting subsequent state estimation and uncertainty management.

[0098] In actual implementation, after obtaining the relative trust parameter (i.e., Kalman gain), the system further obtains observation information (i.e., observation matrix), and determines the second correction information based on this observation information and the relative trust parameter. In this step, the second correction information is obtained by calculating the unit matrix I minus the product of the Kalman gain and the observation matrix, which reflects the degree of correction of the current prediction covariance information. Then, the system uses the second correction information to correct the prediction observation covariance information, and finally obtains the reference covariance information corresponding to the current time point. For the correction based on the second correction information in the step, one implementation method is to directly apply the calculated second correction information to update the current prediction observation covariance information; another implementation method may involve more complex filtering algorithm optimization to ensure the accuracy of the correction result.

[0099] Specifically, in a tunnel inspection mission scenario, assume that the drone has approached and received the signal from the trackside transponder, and the system has calculated the relative trust parameters. At this time, the system obtains observation information based on the data from the IMU and other sensors, and calculates the second correction information based on this observation information and the relative trust parameters. For example, if the current predicted covariance information shows a large uncertainty, the system can adjust it through the second correction information to reduce this uncertainty. Specifically, the system calculates the unit matrix I minus the product of the Kalman gain and the observation matrix to obtain the second correction information, and uses it to correct the predicted observation covariance information, thereby obtaining a more accurate reference covariance information corresponding to the time point.

[0100] In the embodiments of this specification, by combining observation information with relative trust parameters to calculate second correction information, and using it to correct the predicted observation covariance information, the system can effectively reduce the uncertainty in state prediction and improve the accuracy and robustness of the positioning system.

[0101] In an optional implementation of this embodiment, after correcting the current reference position based on the time point of passing the target transponder and the target position of the target transponder to obtain the target position information of the drone, the method further includes:

[0102] Predict the target motion trajectory of the UAV based on the target position information and motion state information;

[0103] The reference motion trajectory is corrected based on the target motion trajectory, wherein the reference motion trajectory is the motion trajectory of the UAV predicted based on the reference position information.

[0104] It's important to note that target position information represents the precise, calibrated coordinates of the drone's position, while motion state information includes parameters such as velocity, angular velocity, attitude, and acceleration. The target trajectory is the predicted future path of the drone based on this latest position and motion state information. The reference trajectory is the predicted trajectory of the drone based on the previously used reference position information.

[0105] In actual implementation, after the current reference position of the UAV is corrected through the position information of the target transponder and the target position information is obtained, the system will combine the target position information and the latest motion state information (such as speed, angular velocity, etc.) to predict the target motion trajectory of the UAV. This process usually involves the use of dynamic models or filtering algorithms (such as extended Kalman filter EKF) to take into account various possible dynamic changes. Then, the system uses the calculated target motion trajectory to correct the reference motion trajectory predicted based on the previous reference position information to ensure that the subsequent flight path is more accurate. For the correction of the reference motion trajectory based on the target motion trajectory in the step, one implementation method is to directly replace the old trajectory prediction; another implementation method may be to fuse the results of the two to smooth the transition and optimize the final predicted trajectory.

[0106] Specifically, in a tunnel inspection mission scenario, assume that the drone has successfully received the signal from the trackside transponder and completed the correction of its current position. At this point, the system uses a dynamic model to predict the target motion trajectory of the drone in the next period of time based on the updated position and motion state data provided by the IMU. For example, if the drone is executing a specific inspection route, the system will adjust its expected path based on the new position and speed information to ensure that it can accurately reach the next checkpoint. The system then compares this newly predicted target motion trajectory with the reference motion trajectory previously derived based on the reference position information, and corrects it based on the difference. This step not only improves the accuracy of the drone's flight path, but also reduces the deviation caused by cumulative errors and enhances the system's autonomous navigation capabilities.

[0107] In the embodiments of the specification, by combining the target position information and the latest motion state information to predict the target motion trajectory of the drone, and based on this, correcting the reference motion trajectory, the system can significantly improve the navigation accuracy of the drone in complex environments.

[0108] One embodiment of the present invention uses the motion and environmental information collected by a drone to determine a preliminary position. When a trackside transponder signal is received, the drone's position is corrected based on the transponder's precise location and time. This approach not only improves positioning accuracy but also reduces positioning errors in GPS-denied environments, such as in rail transit tunnels, enhancing the safety and efficiency of inspection missions. This real-time correction ensures that the drone accurately follows its intended trajectory, improving overall operational reliability.

[0109] The following combined Figure 2 , taking the application of the drone positioning method provided by the present invention in the rail transit tunnel inspection scene as an example, the drone positioning method is further explained. Figure 2The present invention provides a flowchart of a method for positioning a UAV, which includes the following steps.

[0110] Step 202: Define the state vector of the drone.

[0111] Specifically, under the EKF framework, state variables, prediction equations and observation equations are set to achieve positioning error correction.

[0112] Define the state vector of the drone:

[0113]

[0114] in:

[0115] Indicates the position of the UAV in the orbital coordinate system;

[0116] represents the velocity vector of the UAV;

[0117] The attitude angles of the drone (pitch, roll, and yaw).

[0118] Step 204: The state transition model is based on the UAV motion equations calculated by the inertial measurement system, vision system, and lidar.

[0119] For details, see Figure 3 As shown, Figure 3 This is a scene diagram of a drone positioning method provided by an embodiment of the present invention. The drone inside the tunnel wall is equipped with LiDAR, IMU and visual sensors for autonomous positioning and navigation. In addition, the drone is also equipped with an antenna for receiving signals from the trackside transponder. When the drone approaches the transponder, the antenna sends high-frequency electromagnetic waves to activate the transponder. The transponder transmits pre-stored fixed information (such as mileage points, speed limits, etc.) or dynamic information (such as signal status, switch position, etc.) to the drone through radio induction communication, thereby achieving high-precision position correction and improving the reliability and accuracy of inspection tasks.

[0120] The state transition model is based on the UAV motion equation calculated by IMU+visual sensor+LiDAR:

[0121]

[0122] in:

[0123] F is the state transition matrix, which describes the dynamic model of the UAV;

[0124] G is the control input matrix, which contains the acceleration and angular velocity provided by the IMU;

[0125] is the IMU observation value (acceleration, angular velocity);

[0126] is the process noise, obey , used to describe the errors of LiDAR, IMU and vision, Q is the preset process noise parameter;

[0127] k is the current moment.

[0128] Step 206: When the drone detects the transponder signal, its absolute position can be obtained.

[0129] Specifically, the visual sensor obtains the current position Y of the drone through mechanical observation of the camera device:

[0130]

[0131] in:

[0132] H is the observation matrix, mapping state variables to observation space (position observation);

[0133] is the observation noise, obey , the main source is the measurement error of the visual sensor, and R is the preset observation noise parameter.

[0134] When the UAV detects the transponder signal, the absolute position Zk of the train transponder can be obtained and replaced by Y.

[0135] Step 208: Optimize the position information of the UAV through time update and measurement update.

[0136] Specifically, EKF optimizes the UAV’s position information through two steps: time update (prediction) and measurement update (transponder correction).

[0137] Time update (forecast)

[0138] Compute state prediction: ;

[0139] Compute the covariance forecast: .

[0140] in, is the motion equation to be corrected at time k predicted based on the motion equation at time k-1, is the equation of motion at time k-1, is the covariance matrix at moment k predicted based on the covariance at moment k-1, is the covariance matrix at time k-1.

[0141] Measurement update (balise correction) (performed only if a balise signal is detected)

[0142] Calculate the Kalman gain: ;

[0143] Calculation status correction: ;

[0144] Update the covariance matrix: .

[0145] in, is the Kalman gain parameter, is the updated covariance matrix.

[0146] Through steps 202-208 above, a combined positioning system integrating IMU, LiDAR, vision, and transponder signal perception is proposed, aiming to achieve accurate drone positioning in complex environments such as tunnels. This system utilizes the IMU for short-term, high-frequency attitude and position estimation, combines LiDAR with vision technology to perform SLAM to enhance environmental perception and relative positioning capabilities, and simultaneously uses transponders laid along the track to provide high-precision reference coordinates to correct accumulated errors. When the drone passes by the transponder, its antenna senses the signal and triggers an immediate position correction. The trajectory is optimized using an extended Kalman filter method, effectively reducing error accumulation during long-term flight while ensuring short-term, high-precision positioning, thereby improving the accuracy and stability of overall positioning.

[0147] By incorporating transponder signal correction into the extended Kalman filter (EKF) combined positioning framework, this solution effectively eliminates the drift error of the visual odometry and significantly improves the robustness and accuracy of positioning by leveraging multi-sensor information fusion (IMU, LiDAR, vision, and transponder). This approach is not only computationally intensive and suitable for real-time rail inspections, but also enables low-cost deployment, requiring only antennas attached to drones and leveraging existing rail transponders to achieve centimeter-level positioning accuracy. This enables inspection drones to fly autonomously in complex environments, significantly improving their application capabilities and efficiency in enclosed environments such as rail tunnels.

[0148] Corresponding to the above method embodiment, the present invention also provides an embodiment of a drone positioning device, Figure 4 FIG1 shows a schematic structural diagram of a UAV positioning device provided by an embodiment of the present invention. Figure 4 As shown, the device includes:

[0149] The positioning module 402 is configured to perform positioning based on the collected current motion state information and / or environmental information to determine the current reference position of the drone.

[0150] The determination module 404 is configured to determine the time point of passing the target transponder and the target position of the target transponder when the transponder receiving device receives the signal of the target transponder, wherein the target transponder is any one of the trackside transponders laid along the track in the rail transit system.

[0151] The correction module 406 is configured to correct the current reference position based on the time point of passing the target transponder and the target position of the target transponder to obtain the target position information of the UAV.

[0152] Optionally, the target position information includes the current corrected posture of the UAV; the correction module 406 is further configured to construct a current reference posture based on the time point of passing the target transponder, the current motion state information and the current reference position, wherein the current reference posture is used to describe the position coordinates and motion state of the UAV at the time point; based on the target position of the target transponder and the current reference posture, the current reference posture is corrected to obtain the current corrected posture.

[0153] Optionally, the correction module 406 is further configured to determine a historical time point adjacent to the time point based on the time point of passing the target transponder; obtain a historical posture corresponding to the historical time point; obtain state transition information, control input information and process noise information, wherein the state transition information is used to describe the posture change from the current time point to the subsequent time point, the control input information is the attitude control information input to the drone, and the process noise information is used to quantify the uncertainty of the posture change process; based on the current motion state information, state transition information, control input information and process noise information, the current reference posture is predicted.

[0154] Optionally, the correction module 406 is further configured to calculate the difference information between the target position of the target transponder and the current reference position; obtain a relative trust parameter, wherein the relative trust parameter is used to describe the degree of correction of the current reference posture; based on the relative trust parameter and the difference information, calculate and obtain first correction information, wherein the first correction information is used to describe the degree of correction of the current reference posture; correct the current reference posture based on the first correction information to obtain the current corrected posture.

[0155] Optionally, the correction module 406 is further configured to obtain covariance prediction information corresponding to a time point, wherein the covariance prediction information is used to describe prediction uncertainty; obtain observation information and observation noise parameters, wherein the observation information is used to convert posture into position information, and the observation noise parameters are used to describe the uncertainty of the observation information; based on the observation information, map the covariance prediction information to the observation space to obtain predicted observation covariance information, wherein the predicted observation covariance information is used to describe the result after converting the uncertainty in the state space to the observation space; based on the predicted observation covariance information and the observation noise parameters, adjust the weight calculation to obtain the relative trust parameter.

[0156] Optionally, the correction module 406 is further configured to determine a historical time point adjacent to the time point based on the time point of passing the target transponder; obtain state transition information, process noise parameters and historical covariance information corresponding to the historical time point, wherein the process noise parameters are used to describe the uncertainty of the state transition information; and make predictions based on the historical reference covariance information, the state transition information and the process noise parameters to obtain covariance prediction information corresponding to the time point.

[0157] Optionally, the correction module 406 is further configured to obtain observation information and determine second correction information based on the observation information and the relative trust parameter, wherein the second correction information is used to describe the degree of correction of the covariance prediction information; and correct the predicted observation covariance information based on the second correction information to obtain the reference covariance information corresponding to the time point.

[0158] Optionally, the UAV positioning device also includes a trajectory correction module, which is configured to predict the target motion trajectory of the UAV based on the target position information and motion state information; and correct the reference motion trajectory based on the target motion trajectory, wherein the reference motion trajectory is the motion trajectory of the UAV predicted based on the reference position information.

[0159] This embodiment provides a drone positioning device that uses the drone's motion and environmental information to determine its initial position. Upon receiving a trackside transponder signal, the device corrects the drone's position based on the transponder's precise location and timing. This approach not only improves positioning accuracy but also reduces positioning errors in GPS-denied environments, such as in rail transit tunnels, enhancing the safety and efficiency of inspection missions. This real-time correction ensures the drone's accurate flight along its intended trajectory, improving overall operational reliability.

[0160] The above is a schematic diagram of a drone positioning device according to this embodiment. It should be noted that the technical solution of this drone positioning device and the technical solution of the aforementioned drone positioning method are based on the same concept. For details not described in detail in the technical solution of the drone positioning device, please refer to the description of the technical solution of the aforementioned drone positioning method.

[0161] Figure 5 1 shows a block diagram of a computing device 500 according to an embodiment of the present invention. Components of the computing device 500 include, but are not limited to, a memory 510 and a processor 520. The processor 520 is connected to the memory 510 via a bus 530, and a database 550 is used to store data.

[0162] The computing device 500 also includes an access device 540 that enables the computing device 500 to communicate via one or more networks 560. Examples of such networks include a public switched telephone network (PSTN), a local area network (LAN), a wide area network (WAN), a personal area network (PAN), or a combination of communication networks such as the Internet. The access device 540 may include one or more of any type of network interface (e.g., a network interface card (NIC)) whether wired or wireless, such as an IEEE 802.11 wireless local area network (WLAN) wireless interface, a Worldwide Interoperability for Microwave Access (Wi-MAX) interface, an Ethernet interface, a universal serial bus (USB) interface, a cellular network interface, a Bluetooth interface, or a near field communication (NFC) interface.

[0163] In one embodiment of the present invention, the above components of the computing device 500 and Figure 5 Other components not shown in the figure may also be connected to each other, for example, via a bus. Figure 5 The computing device structure block diagram shown is for illustrative purposes only and is not intended to limit the scope of the present invention. Those skilled in the art may add or replace other components as needed.

[0164] Computing device 500 can be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., a tablet computer, personal digital assistant, laptop computer, notebook computer, netbook computer, etc.), a mobile phone (e.g., a smartphone), a wearable computing device (e.g., a smartwatch, smart glasses, etc.), or other types of mobile devices, or a stationary computing device such as a desktop computer or personal computer (PC). Computing device 500 can also be a mobile or stationary server.

[0165] The processor 520 is configured to execute the following computer program / instructions, which, when executed by the processor, implement the steps of the above-mentioned drone positioning method.

[0166] The various embodiments of the present invention are described in a progressive manner. Similar portions between the various embodiments can be referenced to each other. Each embodiment focuses on the differences between the other embodiments. In particular, the computing device embodiment is generally similar to the drone positioning method embodiment, so its description is relatively simple. For relevant portions, refer to the description of the drone positioning method embodiment.

[0167] An embodiment of the present invention further provides a computer-readable storage medium storing a computer program / instruction, which implements the steps of the above-mentioned drone positioning method when executed by a processor.

[0168] The various embodiments of the present invention are described in a progressive manner. Similar portions between the various embodiments can be referenced to each other. Each embodiment focuses on the differences between the other embodiments. In particular, the computer-readable storage medium embodiment is generally similar to the drone positioning method embodiment, so its description is relatively brief. For relevant portions, refer to the description of the drone positioning method embodiment.

[0169] An embodiment of the present invention further provides a computer program product, comprising a computer program / instruction, which implements the steps of the above-mentioned drone positioning method when executed by a processor.

[0170] The above is a schematic diagram of a computer program product according to this embodiment. It should be noted that the technical solution of this computer program product is based on the same concept as the technical solution of the aforementioned drone positioning method. For details not described in detail in the technical solution of the computer program product, please refer to the description of the technical solution of the aforementioned drone positioning method.

[0171] The foregoing description describes specific embodiments of the present invention. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that described in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order shown or the sequential order to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0172] The computer instructions include computer program code, which may be in source code form, object code form, executable file, or some intermediate form. The computer-readable medium may include any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal, and software distribution medium. It should be noted that the content of the computer-readable medium may be appropriately increased or decreased based on the requirements of patent practice. For example, in some regions, according to patent practice, computer-readable media does not include electric carrier signals and telecommunication signals.

[0173] It should be noted that for the aforementioned method embodiments, for ease of description, they are all expressed as a series of action combinations. However, those skilled in the art should be aware that the embodiments of the present invention are not limited by the order of the actions described, because according to the embodiments of the present invention, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in this specification are all preferred embodiments, and the actions and modules involved are not necessarily required for the embodiments of the present invention.

[0174] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0175] The preferred embodiments of the present invention disclosed above are intended only to help illustrate the present invention. The alternative embodiments do not describe all details in detail, nor do they limit the invention to the specific embodiments described. Obviously, many modifications and variations can be made based on the content of the embodiments of the present invention. The present invention selects and describes these embodiments in detail to better explain the principles and practical applications of the embodiments of the present invention, thereby enabling those skilled in the art to better understand and utilize the present invention. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for positioning a drone, characterized in that: Applied to a drone, the drone is equipped with a transponder receiving device corresponding to a trackside transponder, and the method comprises: Performing positioning based on the collected current motion state information and / or environmental information to determine the current reference position of the UAV; determining, when the transponder receiving device receives a signal from a target transponder, a time point of passing the target transponder and a target position of the target transponder, wherein the target transponder is any one of the trackside transponders laid along the track in the rail transit system; Based on the time point of passing the target transponder and the target position of the target transponder, the current reference position is corrected to obtain the target position information of the UAV.

2. The method according to claim 1, characterized in that The target position information includes the current corrected position of the UAV; the correcting the current reference position based on the time point of passing the target transponder and the target position of the target transponder to obtain the target position information of the UAV includes: Constructing a current reference pose based on the time point of passing the target transponder, the current motion state information, and the current reference position, wherein the current reference pose is used to describe the position coordinates and motion state of the UAV at the time point; Based on the target position of the target transponder and the current reference pose, the current reference pose is corrected to obtain the current corrected pose.

3. The method according to claim 2, characterized in that The constructing a current reference pose based on the time point of passing the target transponder, the current motion state information, and the current reference position includes: determining, based on a time point at which the target transponder passes, a historical time point adjacent to the time point; Obtaining the historical posture corresponding to the historical time point; Acquire state transition information, control input information, and process noise information, wherein the state transition information is used to describe the posture change from a current time point to a subsequent time point, the control input information is attitude control information input to the UAV, and the process noise information is used to quantify the uncertainty of the posture change process; A current reference pose is predicted based on the current motion state information, the state transition information, the control input information, and the process noise information.

4. The method according to claim 2, characterized in that The correcting the current reference posture based on the target position of the target transponder and the current reference position to obtain the current corrected posture includes: Calculating difference information between a target position of the target transponder and the current reference position; Obtaining a relative trust parameter, wherein the relative trust parameter is used to describe the degree of correction of the current reference pose; Calculating first correction information based on the relative trust parameter and the difference information, wherein the first correction information is used to describe a degree of correction of the current reference posture; The current reference posture is corrected based on the first correction information to obtain the current corrected posture.

5. The method according to claim 4, characterized in that The obtaining of the relative trust parameter includes: Obtaining covariance prediction information corresponding to the time point, wherein the covariance prediction information is used to describe prediction uncertainty; Obtaining observation information and observation noise parameters, wherein the observation information is used to convert the pose into position information, and the observation noise parameters are used to describe the uncertainty of the observation information; Based on the observation information, mapping the covariance prediction information to the observation space to obtain predicted observation covariance information, wherein the predicted observation covariance information is used to describe the result after converting the uncertainty in the state space to the observation space; Based on the predicted observation covariance information and the observation noise parameter, an adjustment weight calculation is performed to obtain a relative trust parameter.

6. The method according to claim 5, characterized in that The obtaining of the covariance prediction information corresponding to the time point includes: determining, based on a time point at which the target transponder passes, a historical time point adjacent to the time point; Acquiring state transition information, a process noise parameter, and historical covariance information corresponding to the historical time point, wherein the process noise parameter is used to describe the uncertainty of the state transition information; Prediction is performed based on the historical reference covariance information, the state transition information, and the process noise parameter to obtain covariance prediction information corresponding to the time point.

7. The method according to claim 5, characterized in that After performing the weight adjustment calculation based on the predicted observation covariance information and the observation noise parameter to obtain the relative trust parameter, the method further includes: Obtaining observation information, and determining second correction information based on the observation information and the relative trust parameter, wherein the second correction information is used to describe a degree of correction of the covariance prediction information; The predicted observation covariance information is corrected based on the second correction information to obtain reference covariance information corresponding to the time point.

8. The method according to claim 1, characterized in that After correcting the current reference position based on the time point of passing the target transponder and the target position of the target transponder to obtain the target position information of the UAV, the method further includes: Predicting a target motion trajectory of the UAV based on the target position information and the motion state information; The reference motion trajectory is corrected based on the target motion trajectory, wherein the reference motion trajectory is the motion trajectory of the UAV predicted based on the reference position information.

9. A UAV positioning device, characterized in that: Applied to UAVs, the UAV is equipped with a transponder receiving device corresponding to a trackside transponder, including: a positioning module, configured to perform positioning based on the collected current motion state information and / or environmental information to determine a current reference position of the UAV; a determination module configured to determine, when the transponder receiving device receives a signal from a target transponder, a time point of passing the target transponder and a target position of the target transponder, wherein the target transponder is any one of the trackside transponders laid along the track in the rail transit system; The correction module is configured to correct the current reference position based on the time point of passing the target transponder and the target position of the target transponder to obtain the target position information of the UAV.

10. A computing device, characterized in that include: memory and processor; The memory is used to store computer programs / instructions, and the processor is used to execute the computer programs / instructions. When the computer program / instructions are executed by the processor, the steps of the drone positioning method according to any one of claims 1 to 8 are implemented.

11. A computer-readable storage medium, characterized in that It stores a computer program / instruction, which, when executed by a processor, implements the steps of the drone positioning method described in any one of claims 1 to 8.

12. A computer program product, characterized in that The method comprises a computer program / instruction, which, when executed by a processor, implements the steps of the drone positioning method according to any one of claims 1 to 8.

Images