Rapid positioning system and method for unmanned aerial vehicle

Abstract

The invention provides a rapid positioning system and method for an unmanned aerial vehicle, and relates to the technical field of unmanned aerial vehicle navigation, and the system comprises a multi-source sensing data collection module which is used for collecting the position sensing data of the unmanned aerial vehicle; the dynamic pose rough calculation module is used for calculating initial pose data based on the instantaneous motion state high-frequency motion data; the environment feature matching module is used for extracting real-time environment feature data based on the environment image data and the environment three-dimensional point cloud data and determining relative pose reference data; and the positioning result correction module is used for determining final pose data based on the initial pose data and the relative pose reference data. According to the system and the method provided by the invention, by acquiring the multi-source position sensing data and the data measured based on the inertial measurement unit, the pose calculation is carried out, and the feature matching and the dynamic weight fusion are carried out through the environment image data and the environment three-dimensional point cloud data, so that the positioning precision is improved, and the precise positioning of the unmanned aerial vehicle is realized.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) navigation technology, and in particular to a rapid positioning system and method for UAVs. Background Technology

[0002] As the application scope of unmanned aerial vehicles (UAVs) expands, the requirements for real-time positioning, anti-interference capabilities, and environmental adaptability are becoming increasingly stringent. Existing positioning technologies have significant limitations: when positioning based on the Global Navigation Satellite System (GNSS), positioning is easily interrupted due to signal loss in obstructed environments; when positioning based on the Inertial Measurement Unit (IMU), although it can output pose at high frequency, there is a cumulative error, and the accuracy decreases significantly over long-term use; when positioning based on vision or lidar, it is easily affected by sudden changes in lighting and sparse environmental textures, and the stability of feature extraction and matching is insufficient.

[0003] Therefore, how to achieve precise positioning of unmanned aerial vehicles has become a technical problem that the industry urgently needs to solve. Summary of the Invention

[0004] This invention provides a rapid positioning system and method for unmanned aerial vehicles (UAVs) to address the shortcomings of low positioning accuracy of UAVs in complex environments in existing technologies, thereby achieving precise positioning of UAVs.

[0005] This invention provides a rapid positioning system for unmanned aerial vehicles, including a multi-source sensing data acquisition module, a dynamic pose coarse calculation module, an environmental feature matching module, and a positioning result correction module; The multi-source sensing data acquisition module is used to acquire the position sensing data of the unmanned aerial vehicle; the position sensing data includes at least one of global position coordinate data, instantaneous motion state data, environmental image data, and environmental three-dimensional point cloud data; The dynamic pose coarse calculation module is used to calculate preliminary pose data based on the instantaneous motion state data; The environmental feature matching module is used to extract real-time environmental feature data based on the environmental image data and the environmental 3D point cloud data, and match the real-time environmental feature data with the pre-stored environmental feature data to determine the relative pose reference data. The positioning result correction module is used to determine the final pose data based on the preliminary pose data and the relative pose reference data.

[0006] In some embodiments, the system further includes a data preprocessing and synchronization module and a positioning information output and feedback module; The data preprocessing and synchronization module is used to preprocess the location-aware data and align the location-aware data in the time dimension. The positioning information output and feedback module is used to transmit the final pose data to the flight control system and adjust the sampling parameters of the position perception data based on the motion state data of the unmanned aerial vehicle fed back by the flight control system.

[0007] In some embodiments, the dynamic pose coarse calculation module includes an instantaneous motion state data parsing unit and a pose recursion unit; The instantaneous motion state data parsing unit is used to solve the processed instantaneous motion state data and extract the angular velocity component and the linear acceleration component. The pose recursion unit is used to calculate the attitude angle based on the angular velocity component, calculate the position coordinate based on the linear acceleration component, and determine the preliminary pose data based on the position coordinate and the attitude angle.

[0008] In some embodiments, the environmental feature matching module includes a feature extraction unit and a feature matching evaluation unit; The feature extraction unit is used to extract corner features based on the environmental image data, extract planar features and edge features based on the environmental 3D point cloud data, and construct a real-time environmental feature set based on the corner features, the planar features and the edge features; The feature matching evaluation unit is used to match the real-time environmental feature data in the real-time environmental feature set with the pre-stored environmental feature data in the pre-stored environmental feature library to determine the matching similarity. If the matching similarity meets the preset conditions, relative pose reference data is determined based on the environmental image data and the environmental three-dimensional point cloud data.

[0009] In some embodiments, the positioning result correction module includes an error analysis unit and a weighted fusion unit; The error analysis unit is used to determine the error value of the preliminary pose data and the error value of the relative pose reference data. The weighted fusion unit is used to determine the weight coefficients of the preliminary pose data based on the error values ​​of the preliminary pose data and the error values ​​of the relative pose reference data, and to fuse the preliminary pose data and the relative pose reference data based on the weight coefficients to determine the final pose data.

[0010] In some embodiments, the positioning information output and feedback module includes an information format conversion unit and a feedback adjustment unit; The information format conversion unit is used to convert the final pose data into a data format supported by the flight control system, and transmit the converted final pose data to the flight control system. The feedback adjustment unit is used to receive the motion state data of the unmanned aerial vehicle fed back by the flight control system, and adjust the sampling parameters of the position perception data based on the motion state data.

[0011] This invention provides a method for rapid localization of unmanned aerial vehicles, comprising: The unmanned aerial vehicle (UAV) acquires position perception data; the position perception data includes at least one of global position coordinate data, instantaneous motion state data, environmental image data, and environmental 3D point cloud data. Based on the instantaneous motion state data, preliminary pose data are calculated; Real-time environmental feature data is extracted based on the environmental image data and the environmental 3D point cloud data, and the real-time environmental feature data is matched with the pre-stored environmental feature data to determine the relative pose reference data. Based on the preliminary pose data and the relative pose reference data, the final pose data is determined.

[0012] The present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the rapid positioning method for unmanned aerial vehicles.

[0013] The present invention provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the aforementioned rapid positioning method for unmanned aerial vehicles.

[0014] This invention provides a computer program product, including a computer program that, when executed by a processor, implements the rapid positioning method for unmanned aerial vehicles.

[0015] The rapid positioning system and method for unmanned aerial vehicles (UAVs) provided by this invention solves the limitations of single sensors in environments such as obstruction and indoor environments by collecting multi-source position perception data, and can adapt to various scenarios such as open, complex, and enclosed spaces. Based on the data measured by the inertial measurement unit, pose estimation is performed, achieving millisecond-level pose updates to meet the response requirements of high-speed movement of the UAV. By performing feature matching and dynamic weighted fusion of environmental image data and environmental 3D point cloud data, errors are effectively reduced, positioning accuracy is improved, and precise positioning of the UAV is achieved. Attached Figure Description

[0016] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0017] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0018] Figure 1 This is one of the structural schematic diagrams of the rapid positioning system for unmanned aerial vehicles provided by the present invention.

[0019] Figure 2 This is a schematic diagram of the multi-source sensing data acquisition module of the rapid positioning system for unmanned aerial vehicles provided by the present invention.

[0020] Figure 3 This is the second structural schematic diagram of the rapid positioning system for unmanned aerial vehicles provided by the present invention.

[0021] Figure 4 This is a flowchart illustrating the rapid positioning method for unmanned aerial vehicles provided by the present invention.

[0022] Figure 5 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0023] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0024] It should be noted that the terms "first," "second," etc., used in this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that comprises a series of steps, units, or modules is not necessarily limited to those explicitly listed, but may include other steps, units, or modules not explicitly listed or inherent to such processes, methods, products, or devices.

[0025] Figure 1 This is one of the structural schematic diagrams of the rapid positioning system for unmanned aerial vehicles provided by the present invention, such as... Figure 1 As shown, the system includes a multi-source sensing data acquisition module 110, a dynamic pose coarse calculation module 120, an environmental feature matching module 130, and a positioning result correction module 140.

[0026] The multi-source sensing data acquisition module 110 is used to acquire the position sensing data of the unmanned aerial vehicle; the position sensing data includes at least one of global position coordinate data, instantaneous motion state data, environmental image data, and environmental three-dimensional point cloud data. The dynamic pose coarse calculation module 120 is used to calculate preliminary pose data based on the instantaneous motion state data; The environmental feature matching module 130 is used to extract real-time environmental feature data based on the environmental image data and the environmental three-dimensional point cloud data, and match the real-time environmental feature data with the pre-stored environmental feature data to determine the relative pose reference data. The positioning result correction module 140 is used to determine the final pose data based on the preliminary pose data and the relative pose reference data.

[0027] Specifically, the rapid positioning system for unmanned aerial vehicles (UAVs) provided in this embodiment of the invention is mainly used for locating UAVs. The rapid positioning system can run on a computer. Each module in the rapid positioning system can be a hardware device with corresponding functions installed, or a software program with corresponding functions.

[0028] Unmanned aerial vehicles (UAVs) are aircraft that achieve autonomous or remote-controlled flight through an onboard navigation system without the need for a human pilot. They can use their energy, propulsion, and various types of onboard sensors to complete perception, positioning, decision-making, and control, thereby performing flight missions such as takeoff, hovering, cruising, and landing in three-dimensional space, and can also carry out photography, measurement, transportation, or operations according to predetermined mission payloads.

[0029] The multi-source sensing data acquisition module is used to collect the position sensing data of the unmanned aerial vehicle, providing multi-dimensional raw data support for positioning; the position sensing data includes at least one of global position coordinate data, instantaneous motion state data, environmental image data, and environmental 3D point cloud data.

[0030] Figure 2 This is a schematic diagram of the multi-source sensing data acquisition module of the rapid positioning system for unmanned aerial vehicles provided by the present invention, as shown below. Figure 2 As shown, in this embodiment of the invention, the multi-source sensing data acquisition module includes a GNSS receiving unit, an IMU unit, a visual acquisition unit, and a lidar unit.

[0031] The GNSS receiving unit is used to receive signals from the Global Navigation Satellite System, obtain the global position coordinates of the aircraft, including longitude, latitude and altitude, and provide a global positioning reference.

[0032] The IMU unit is used to acquire the angular velocity and linear acceleration of the unmanned aerial vehicle (UAV) at high frequency to obtain IMU data, i.e., instantaneous motion state data, which reflects the instantaneous motion state of the UAV. Angular velocity is a vector describing the rate and direction of rotation of a rigid body about a certain instantaneous axis; in this embodiment of the invention, angular velocity is the rotational velocity about the x, y, and z axes. Linear acceleration is a vector describing the rate and direction of change of velocity of a point mass or the center of mass of a rigid body; in this embodiment of the invention, linear acceleration is the acceleration along the x, y, and z axes.

[0033] The visual acquisition unit is used to capture images of the environment around the aircraft, obtain environmental image data, and capture environmental texture features, such as wall textures and landmarks.

[0034] The lidar unit is used to scan the three-dimensional point cloud of the surrounding environment, obtain the three-dimensional point cloud data of the environment, and acquire spatial structural features, such as planes and edges.

[0035] It should be noted that each unit is equipped with a timestamp generator to add a unique timestamp to each frame of collected data for subsequent data synchronization.

[0036] For example, the system first performs initialization and feature library construction. It then starts each module of the positioning system, completes self-tests, and checks whether the GNSS, IMU, visual camera, and LiDAR are working properly and whether the data transmission link is unobstructed. The multi-source sensing data acquisition module collects initial environmental data, the GNSS obtains the initial global position, the visual camera captures environmental images within a 100-meter radius, and the LiDAR scans the 3D point cloud of the area. The environmental feature matching module builds a pre-stored environmental feature library based on the initial environmental data, extracts image ORB features and point cloud structure features, associates them with the initial GNSS position information, and stores them in system memory. The IMU unit performs zero-bias calibration, i.e., the IMU is placed stationary for 30 seconds, and the average angular velocity measurement is taken as the baseline. , , The initial value is set. The positioning information output module establishes a communication connection with the flight control system to confirm data format compatibility; after initialization, the system enters the positioning standby state.

[0037] Then, real-time acquisition of multi-source sensing data is performed. The GNSS receiving unit continuously receives satellite signals and outputs global position data every second, which is then timestamped and transmitted to the data preprocessing module. The IMU unit acquires angular velocity and linear acceleration data at a set frequency, timestamping each frame of data and transmitting it to the data preprocessing module via the Controller Area Network (CAN) bus. The visual acquisition unit captures environmental images at a set frequency, and the LiDAR scans 3D point clouds at a set frequency; both types of data are timestamped and transmitted to the data preprocessing module. Furthermore, it is ensured that the data acquisition process of each sensor is uninterrupted, forming a continuous data stream.

[0038] The dynamic pose coarse calculation module is used to perform real-time pose recursion based on instantaneous motion state data, quickly output preliminary pose data, and ensure positioning response speed.

[0039] The environmental feature matching module is used to extract real-time environmental feature data based on environmental image data and environmental 3D point cloud data, and match the real-time environmental feature data with pre-stored environmental feature data to determine relative pose reference data; that is, by extracting and matching stable environmental features, relative pose reference is obtained to make up for the positioning gap when the global positioning signal is missing.

[0040] The positioning result correction module is used to determine the final positioning data based on the preliminary pose data and the relative pose reference data; it aims to integrate the coarsely calculated pose and the feature matching results, dynamically compensate for positioning errors, and generate accurate pose.

[0041] The rapid positioning system for unmanned aerial vehicles (UAVs) provided in this invention solves the limitations of single sensors in environments such as obstruction and indoor environments by collecting multi-source position perception data, and can adapt to various scenarios such as open, complex, and enclosed spaces. Based on the data measured by the inertial measurement unit, pose estimation is performed, achieving millisecond-level pose updates to meet the response requirements of high-speed UAV movement. By performing feature matching and dynamic weighted fusion of environmental image data and environmental 3D point cloud data, errors are effectively reduced, positioning accuracy is improved, and precise positioning of the UAV is achieved.

[0042] Figure 3 This is the second structural schematic diagram of the rapid positioning system for unmanned aerial vehicles provided by the present invention, as shown below. Figure 3 As shown, the system also includes a data preprocessing and synchronization module 150 and a positioning information output and feedback module 160.

[0043] In some embodiments, the system further includes a data preprocessing and synchronization module 150 and a positioning information output and feedback module 160; The data preprocessing and synchronization module 150 is used to preprocess the location-aware data and align the location-aware data in the time dimension. The positioning information output and feedback module 160 is used to transmit the final pose data to the flight control system 170, and adjust the sampling parameters of the position perception data based on the motion state data of the unmanned aerial vehicle fed back by the flight control system.

[0044] Specifically, in this embodiment of the invention, the data preprocessing and synchronization module is used to preprocess the location-aware data to suppress data noise and to align the location-aware data in the time dimension to unify the data time dimension.

[0045] The data preprocessing and synchronization module includes a noise suppression unit and a time synchronization unit.

[0046] The noise suppression unit is used to remove random noise, such as vibration interference, from instantaneous motion data using an adaptive filtering algorithm, smooth pixel noise from environmental image data using Gaussian filtering, and remove clutter points, such as dust reflections in the air, from environmental 3D point cloud data using statistical filtering. The adaptive filtering algorithm is an algorithm that adjusts filtering parameters online. Its transfer function or weighting coefficients automatically change with the real-time estimation of the statistical characteristics of the input signal, ensuring rapid tracking of abrupt signals while maximizing the suppression of steady-state noise, achieving a dynamically optimal signal-to-noise ratio. Gaussian filtering is a linear smoothing filtering method that applies a Gaussian kernel weighted average to the signal / image, making the output value of each sampling point equal to the weighted average of the values ​​of its neighborhood points using a two-dimensional (or one-dimensional) Gaussian function, thereby suppressing high-frequency noise and maintaining boundary continuity. Statistical filtering is a type of noise removal method based on probability distribution assumptions. By estimating the statistics of neighboring samples (such as mean, variance, standard deviation, or histogram) and setting confidence intervals, outliers falling in low-probability regions are treated as noise and removed or downweighted, thus effectively suppressing impulsive and non-Gaussian noise.

[0047] The time synchronization unit is used to synchronize GNSS time with Coordinated Universal Time (UTC) based on the timestamp of the GNSS receiver unit, ensuring global uniformity. It uses a linear interpolation algorithm to calibrate the timestamps of instantaneous motion state data, environmental image data, and environmental 3D point cloud data, aligning data from different sources to the same time node and eliminating positioning deviations caused by "data timing misalignment".

[0048] For example, during data preprocessing and time synchronization, the noise suppression unit processes multi-source data: adaptive Kalman filtering is used to remove random noise from instantaneous motion state data; 5×5 Gaussian filtering is used to smooth pixel noise from environmental image data; and statistical filtering is used to remove clutter points beyond three standard deviations from the mean distance from environmental 3D point cloud data. When calibrating data timing, the time synchronization unit uses the timestamp of the GNSS data as a reference (e.g., t=10.0s) and compares the timestamps of instantaneous motion state data (t=9.99s, 10.00s, 10.01s), environmental image data (t=9.95s, 10.05s), and environmental 3D point cloud data (t=9.9s, 10.1s). Through linear interpolation, data from non-reference time points are calibrated to the reference time point; for example, visual data at t=9.95s is interpolated to t=10.0s, resulting in time-aligned preprocessed data, which is then transmitted to the dynamic pose coarse calculation module and the environmental feature matching module.

[0049] The positioning information output and feedback module is used to transmit the final pose data to the flight control system and adjust the sampling parameters of the position perception data based on the motion state data of the unmanned aerial vehicle fed back by the flight control system. In other words, it adjusts the positioning strategy according to the flight control feedback to form a closed-loop control.

[0050] The rapid positioning system for unmanned aerial vehicles provided in this embodiment of the invention ensures the validity and temporal consistency of multi-source data by preprocessing the position perception data; by outputting positioning information and adjusting parameters based on feedback from the flight control system, a closed-loop feedback adjustment mechanism is constructed, enabling the system to dynamically optimize parameters according to the flight status and avoid positioning failure caused by external interference.

[0051] In some embodiments, the dynamic pose coarse calculation module includes an instantaneous motion state data parsing unit and a pose recursion unit; The instantaneous motion state data parsing unit is used to solve the processed instantaneous motion state data and extract the angular velocity component and the linear acceleration component. The pose recursion unit is used to calculate the attitude angle based on the angular velocity component, calculate the position coordinate based on the linear acceleration component, and determine the preliminary pose data based on the position coordinate and the attitude angle.

[0052] Specifically, in this embodiment of the invention, the dynamic pose coarse calculation module includes an instantaneous motion state data parsing unit and a pose recursion unit.

[0053] The instantaneous motion state data parsing unit is used to solve the preprocessed instantaneous motion state data and extract the angular velocity components in the carrier coordinate system. With linear acceleration components The carrier coordinate system is an orthogonal right-handed coordinate system fixed to the unmanned aerial vehicle (UAV) body and rotating in real time with its attitude. It is usually defined with the origin at the center of gravity of the aircraft, the X-axis pointing in the direction of the nose forward, the Y-axis pointing in the right wing, and the Z-axis pointing vertically downward to form a right-handed system. All linear accelerations, angular velocities, and control surface moments are first measured and described in this system, and then converted into the navigation or Earth system for fusion calculation through coordinate transformation.

[0054] The pose recursive unit is used to calculate the attitude angles and roll angles based on the Euler angle recursive model. Pitch angle Yaw angle The position coordinates are calculated based on kinematic formulas to obtain the coarse dynamic pose value; among which, the attitude angle recursive formula is: in, The current moment; This refers to the previous sampling time. The sampling interval of the IMU; , , They are respectively Time IMU measurement , , Axial angular velocity; , , These are the angular velocity zero offsets for each axis of the IMU, used to compensate for inherent sensor errors. Initial values ​​are obtained through static calibration during system initialization. , , They are respectively The roll angle, pitch angle, and yaw angle of the aircraft are measured at all times. , , They are respectively The attitude angle at any given moment.

[0055] During the coarse calculation of dynamic pose, the instantaneous motion state data parsing unit extracts the angular velocity at time t from the preprocessed data. With linear acceleration The pose recursive unit calculates the attitude angles based on the recursive formula, substituting... , , (coarse pose value at the previous moment) and , , Calculate time t , , Calculate the position coordinates and linear acceleration. The attitude angle is transformed to the geodetic coordinate system to eliminate the influence of the vehicle's attitude on the acceleration measurement. The transformed acceleration is integrated once to obtain the velocity, and then integrated twice to obtain the position coordinates. Integrating attitude angles and position coordinates, a coarse dynamic pose value is formed. The data is then transmitted to the positioning result correction module.

[0056] Among them, the geodetic coordinate system is a global three-dimensional orthogonal coordinate system based on the Earth's reference ellipsoid, which uniquely describes the spatial position of any point on the Earth's surface using longitude, latitude, and altitude; this system is the basic spatial framework output by global positioning systems such as GNSS, and can serve as an absolute reference for multi-source navigation fusion.

[0057] The rapid positioning system for unmanned aerial vehicles provided in this invention determines preliminary pose data based on instantaneous motion state data, realizes coarse calculation of dynamic pose, provides millisecond-level response for the flight control system, and meets the response requirements of the aircraft during high-speed movement.

[0058] In some embodiments, the environmental feature matching module includes a feature extraction unit and a feature matching evaluation unit; The feature extraction unit is used to extract corner features based on the environmental image data, extract planar features and edge features based on the environmental 3D point cloud data, and construct a real-time environmental feature set based on the corner features, the planar features and the edge features; The feature matching evaluation unit is used to match the real-time environmental feature data in the real-time environmental feature set with the pre-stored environmental feature data in the pre-stored environmental feature library to determine the matching similarity. If the matching similarity meets the preset conditions, relative pose reference data is determined based on the environmental image data and the environmental three-dimensional point cloud data.

[0059] Specifically, corner features refer to isolated pixels / voxels in two-dimensional or three-dimensional data where the curvature or image gradient changes significantly and abruptly. They are uniquely localizable in multiple directions within their spatial neighborhood, thus possessing high repeatability and high recognizability, and are often used for key feature matching in visual or laser point clouds.

[0060] Planar features refer to the set of coplanar points in a 3D point cloud that are larger than a given area threshold and have the same normal vector direction. By estimating the local surface normal vector and the residual from the point to the fitted plane, the extracted plane can be used for environmental structural constraints and geometric correction.

[0061] Edge features refer to a coherent sequence of pixels / points in a two-dimensional image or three-dimensional point cloud that exhibits abrupt changes in intensity or depth along a certain direction; their location corresponds to the contour of an object or the boundary of a depth discontinuity, providing significant geometric cues for tracking, pose constraints, and target recognition.

[0062] In this embodiment of the invention, the environmental feature matching module includes a feature extraction unit and a feature matching evaluation unit.

[0063] The feature extraction unit is used to extract corner features from visual images using the ORB algorithm. These features are rotation- and scale-invariant. The LiDAR point cloud is downsampled using a voxel grid filter to extract planar features, such as the ground and walls, and edge features, such as corners and railings, to form a real-time environmental feature set.

[0064] The feature matching evaluation unit is used to match real-time environmental feature data in the real-time feature set with pre-stored environmental feature data in the pre-stored environmental feature library, and calculate the matching similarity to determine the validity of the match; the similarity is calculated using a distance-weighted formula. in, For similarity matching, the value range is [0,1]. The closer to 1, the better the matching effect; The number of feature point pairs participating in the matching; For the first Weights for feature points (set based on feature point stability; for example, the higher the ORB feature response value, the greater the weight). For the first The Euclidean distance between feature points reflects feature similarity. The smaller the similarity, the higher the similarity. The attenuation coefficient is adaptively adjusted based on environmental complexity, especially in textured environments. Smaller values, sparse texture environment The value is relatively large; when ≥ When a match is deemed valid, the corresponding relative pose reference information is output, including the aircraft's position and attitude relative to the feature library. The similarity threshold is set through testing during system initialization.

[0065] The pre-stored environment feature library is built during system initialization.

[0066] During environmental feature matching and relative pose acquisition, the feature extraction unit processes the preprocessed environmental image data and 3D point cloud data. It extracts ORB feature points (e.g., 500) from the image and extracts planar / edge features (e.g., 20) from the point cloud after downsampling, forming a real-time environmental feature set. The feature matching evaluation unit matches the real-time feature set with a pre-stored environmental feature library, using the Fast Library for Approximate Nearest Neighbors (FLANN) matching algorithm to find feature correspondences, obtaining n pairs of feature points. Then, the matching similarity is calculated. Substituting into the similarity formula of the distance-weighted formula above, calculate... Value; if ≥ (like =0.6), indicating a valid match. The relative pose is calculated using the Perspective-n-Point (PnP) algorithm for visual features and the Iterative Closest Point (ICP) algorithm for point cloud features to calculate the aircraft's relative pose to the pre-stored feature library. The data is transmitted to the positioning result correction module; if... < The system outputs a matching failure signal, and the positioning result correction module does not use feature matching data for the time being.

[0067] The rapid positioning system for unmanned aerial vehicles provided in this invention determines relative pose reference data by matching real-time environmental feature data with pre-stored environmental feature data, thereby compensating for the positioning gap when global positioning signals are missing, reducing errors, and improving positioning accuracy.

[0068] In some embodiments, the positioning result correction module includes an error analysis unit and a weighted fusion unit; The error analysis unit is used to determine the error value of the preliminary pose data and the error value of the relative pose reference data. The weighted fusion unit is used to determine the weight coefficients of the preliminary pose data based on the error values ​​of the preliminary pose data and the error values ​​of the relative pose reference data, and to fuse the preliminary pose data and the relative pose reference data based on the weight coefficients to determine the final pose data.

[0069] Specifically, in this embodiment of the invention, the positioning result correction module includes an error analysis unit and a weighted fusion unit.

[0070] The error analysis unit is used to analyze the error sources of the dynamic pose coarse value, i.e., the preliminary pose data, such as the cumulative error caused by IMU bias and GNSS signal fluctuation error, and to estimate the error magnitude by combining the IMU sampling interval and flight time. Simultaneously, it analyzes the error sources of the feature matching relative pose, i.e., the relative pose reference data, such as the matching deviation caused by feature occlusion, and combines the matching similarity. Estimate the magnitude of the error ( The larger the value, the smaller the error.

[0071] The weighted fusion unit is used to determine the weight coefficients of the two types of localization results based on the error magnitude, and to perform weighted fusion of the coarse dynamic pose value and the relative pose reference information, using the following formula: in, This is the corrected final pose; This is a coarse value for the dynamic pose; The relative pose obtained by feature matching; These are weighting coefficients; the smaller the error, the greater the weight of the corresponding data source. For example, when the GNSS signal is good... Take the larger value when GNSS fails and feature matching is effective. Take the smaller value; if feature matching is invalid, i.e. < If the dynamic pose coarse value is temporarily used as the temporary final pose, a feature re-extraction instruction is triggered to increase the feature extraction frequency in the next cycle.

[0072] When correcting the positioning results, the error analysis unit estimates the error based on the IMU zero bias. , , With sampling interval Estimate The cumulative error; for example, after 10 seconds of flight, the error is approximately 0.1 meters. Based on matching similarity... Estimate Error; for example, When the value is 0.8, the error is approximately 0.05m. Then, the weighting coefficients are determined. If the feature matching is valid, the weight is set according to the principle of "the smaller the error, the greater the weight". ,;For example, Error 0.1m When the error is 0.05m, =0.4; if feature matching is invalid, =1.0. Weighted fusion calculation of the final pose, substituted into... Calculations yielded If feature matching is invalid, = This triggers a feature re-extraction instruction, for example, increasing the visual acquisition frequency to 15Hz in the next cycle.

[0073] The rapid positioning system for unmanned aerial vehicles provided in this invention fuses preliminary pose data and relative pose reference data to determine the final pose data. It utilizes the high update rate of the inertial measurement unit to ensure positioning continuity and real-time performance, uses the stability of environmental feature matching to suppress accumulated errors, and dynamically adjusts the weight ratio of the two through real-time error analysis. This ensures that the final output has both high-frequency response and long-term accuracy at any time, thus solving the fundamental contradiction that a single sensor cannot simultaneously meet the requirements of response speed and positioning accuracy.

[0074] In some embodiments, the positioning information output and feedback module includes an information format conversion unit and a feedback adjustment unit; The information format conversion unit is used to convert the final pose data into a data format supported by the flight control system, and transmit the converted final pose data to the flight control system. The feedback adjustment unit is used to receive the motion state data of the unmanned aerial vehicle fed back by the flight control system, and adjust the sampling parameters of the position perception data based on the motion state data.

[0075] Specifically, in this embodiment of the invention, the positioning information output and feedback module includes an information format conversion unit and a feedback adjustment unit.

[0076] The information format conversion unit is used to convert the corrected final pose, i.e., position coordinates and attitude angles, into a data format supported by the flight control system, such as the Micro Air Vehicle Link (MavLink) protocol or a custom serial port protocol, and transmit it to the flight control system in real time via Ethernet or serial port; providing positioning basis for aircraft attitude control and path planning.

[0077] The feedback adjustment unit receives feedback from the flight control system on the aircraft's motion status, such as flight speed, acceleration, and hover / cruise mode, and dynamically adjusts the operating parameters of the positioning module. For example, when the aircraft is cruising at high speed, the IMU sampling frequency is increased to enhance the real-time performance of attitude recursion; when the aircraft is hovering, the feature extraction frequency of the vision and lidar is increased to optimize positioning accuracy; when the aircraft enters a region with weak GNSS signals, the computational resource allocation of the feature matching module is increased in advance.

[0078] For example, when adjusting the output and feedback of positioning information, the information format conversion unit will... The information is converted to a format supported by the flight controller, such as MavLink messages, and transmitted to the flight control system via Ethernet. Based on this positioning information, the flight controller adjusts motor speeds and controls the aircraft to maintain its course or hover. The feedback adjustment unit receives motion status feedback from the flight controller: if the aircraft's speed exceeds a set threshold, such as 5 m / s in cruise mode, the IMU sampling frequency is increased to 150 Hz; if the aircraft is in hover mode, the visual acquisition frequency is increased to 15 Hz and the lidar frequency to 8 Hz. This process of repeatedly adjusting the UAV's positioning, output, and feedback enables continuous and rapid positioning until the aircraft completes its mission or the system shuts down.

[0079] The rapid positioning system for unmanned aerial vehicles provided in this invention transmits the positioning results to the flight control system and adjusts the positioning strategy based on the feedback from the flight control system to form a closed-loop control, thereby achieving adaptive optimization of the positioning system. This enables the positioning system to perceive mission requirements, intelligently allocate computing resources, and automatically switch to the optimal working mode at different flight stages. This not only improves positioning accuracy and robustness but also reduces system power consumption, realizing a shift from passive output to active adaptation.

[0080] The intelligent formulation method for UAV countermeasure strategy provided by the present invention is described below. The intelligent formulation method for UAV countermeasure strategy described below and the intelligent formulation system for UAV countermeasure strategy described above can be referred to and correspond to each other.

[0081] Figure 4 This is a flowchart illustrating the rapid positioning method for unmanned aerial vehicles provided by the present invention, as shown below. Figure 4 As shown, the method includes steps 410, 420, 430 and 440.

[0082] Step 410: Collect the position perception data of the unmanned aerial vehicle; the position perception data includes at least one of global position coordinate data, instantaneous motion state data, environmental image data, and environmental three-dimensional point cloud data; Step 420: Based on the instantaneous motion state data, calculate the preliminary pose data; Step 430: Extract real-time environmental feature data based on the environmental image data and the environmental 3D point cloud data, and match the real-time environmental feature data with the pre-stored environmental feature data to determine the relative pose reference data; Step 440: Determine the final pose data based on the preliminary pose data and the relative pose reference data.

[0083] The rapid positioning method for unmanned aerial vehicles (UAVs) provided in this invention solves the limitations of single sensors in environments such as obstruction and indoor environments by collecting multi-source position perception data, and can adapt to various scenarios such as open, complex, and enclosed spaces. Based on the data measured by the inertial measurement unit, pose estimation is performed, achieving millisecond-level pose updates to meet the response requirements of the UAV during high-speed movement. By performing feature matching and dynamic weighted fusion of environmental image data and environmental 3D point cloud data, errors are effectively reduced, positioning accuracy is improved, and precise positioning of the UAV is achieved.

[0084] Figure 5 This is a schematic diagram of the structure of the electronic device provided by the present invention, such as... Figure 5 As shown, the electronic device may include: a processor 510, a communications interface 520, a memory 530, and a communications bus 540, wherein the processor 510, the communications interface 520, and the memory 530 communicate with each other via the communications bus 540. The processor 510 can call logical commands stored in the memory 530 to execute the methods described in the above embodiments, for example: The system collects position awareness data of the unmanned aerial vehicle (UAV); the position awareness data includes at least one of global position coordinate data, instantaneous motion state data, environmental image data, and environmental 3D point cloud data; based on the instantaneous motion state data, preliminary pose data is calculated; based on the environmental image data and the environmental 3D point cloud data, real-time environmental feature data is extracted, and the real-time environmental feature data is matched with pre-stored environmental feature data to determine relative pose reference data; based on the preliminary pose data and the relative pose reference data, the final pose data is determined.

[0085] Furthermore, when the logical commands in the aforementioned memory can be implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several commands to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0086] The processor in the electronic device provided in this embodiment of the invention can call logical instructions in the memory to implement the above method. Its specific implementation method is the same as the aforementioned method implementation method and can achieve the same beneficial effects, which will not be repeated here.

[0087] This invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform the methods provided in the above embodiments.

[0088] The specific implementation method is the same as the aforementioned method implementation method and can achieve the same beneficial effects, so it will not be repeated here.

[0089] This invention provides a computer program product, including a computer program that, when executed by a processor, implements the method described above.

[0090] The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0091] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0092] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A rapid positioning system for unmanned aerial vehicles, characterized in that, It includes a multi-source sensing data acquisition module, a dynamic pose coarse calculation module, an environmental feature matching module, and a positioning result correction module; The multi-source sensing data acquisition module is used to acquire the position sensing data of the unmanned aerial vehicle; the position sensing data includes at least one of global position coordinate data, instantaneous motion state data, environmental image data, and environmental three-dimensional point cloud data; The dynamic pose coarse calculation module is used to calculate preliminary pose data based on the instantaneous motion state data; The environmental feature matching module is used to extract real-time environmental feature data based on the environmental image data and the environmental 3D point cloud data, and match the real-time environmental feature data with the pre-stored environmental feature data to determine the relative pose reference data. The positioning result correction module is used to determine the final pose data based on the preliminary pose data and the relative pose reference data.

2. The rapid positioning system for unmanned aerial vehicles according to claim 1, characterized in that, The system also includes a data preprocessing and synchronization module and a positioning information output and feedback module; The data preprocessing and synchronization module is used to preprocess the location-aware data and align the location-aware data in the time dimension. The positioning information output and feedback module is used to transmit the final pose data to the flight control system and adjust the sampling parameters of the position perception data based on the motion state data of the unmanned aerial vehicle fed back by the flight control system.

3. The rapid positioning system for unmanned aerial vehicles according to claim 1, characterized in that, The dynamic pose coarse calculation module includes an instantaneous motion state data parsing unit and a pose recursion unit. The instantaneous motion state data parsing unit is used to solve the processed instantaneous motion state data and extract the angular velocity component and the linear acceleration component. The pose recursion unit is used to calculate the attitude angle based on the angular velocity component, calculate the position coordinate based on the linear acceleration component, and determine the preliminary pose data based on the position coordinate and the attitude angle.

4. The rapid positioning system for unmanned aerial vehicles according to claim 1, characterized in that, The environmental feature matching module includes a feature extraction unit and a feature matching evaluation unit; The feature extraction unit is used to extract corner features based on the environmental image data, extract planar features and edge features based on the environmental 3D point cloud data, and construct a real-time environmental feature set based on the corner features, the planar features and the edge features; The feature matching evaluation unit is used to match the real-time environmental feature data in the real-time environmental feature set with the pre-stored environmental feature data in the pre-stored environmental feature library to determine the matching similarity. If the matching similarity meets the preset conditions, relative pose reference data is determined based on the environmental image data and the environmental three-dimensional point cloud data.

5. The rapid positioning system for unmanned aerial vehicles according to claim 1, characterized in that, The positioning result correction module includes an error analysis unit and a weighted fusion unit; The error analysis unit is used to determine the error value of the preliminary pose data and the error value of the relative pose reference data. The weighted fusion unit is used to determine the weight coefficients of the preliminary pose data based on the error values ​​of the preliminary pose data and the error values ​​of the relative pose reference data, and to fuse the preliminary pose data and the relative pose reference data based on the weight coefficients to determine the final pose data.

6. The rapid positioning system for unmanned aerial vehicles according to claim 1, characterized in that, The positioning information output and feedback module includes an information format conversion unit and a feedback adjustment unit; The information format conversion unit is used to convert the final pose data into a data format supported by the flight control system, and transmit the converted final pose data to the flight control system. The feedback adjustment unit is used to receive the motion state data of the unmanned aerial vehicle fed back by the flight control system, and adjust the sampling parameters of the position perception data based on the motion state data.

7. A method for rapid positioning of an unmanned aerial vehicle (UAV), applied to the rapid positioning system for an UAV as described in any one of claims 1 to 6, characterized in that, The method includes: The unmanned aerial vehicle (UAV) acquires position perception data; the position perception data includes at least one of global position coordinate data, instantaneous motion state data, environmental image data, and environmental 3D point cloud data. Based on the instantaneous motion state data, preliminary pose data are calculated; Real-time environmental feature data is extracted based on the environmental image data and the environmental 3D point cloud data, and the real-time environmental feature data is matched with the pre-stored environmental feature data to determine the relative pose reference data. Based on the preliminary pose data and the relative pose reference data, the final pose data is determined.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the rapid positioning method for unmanned aerial vehicles as described in claim 7.

9. A non-transitory 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 rapid positioning method for unmanned aerial vehicles as described in claim 7.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the rapid positioning method for unmanned aerial vehicles as described in claim 7.

Images