Unmanned aerial vehicle low-altitude operation positioning method and system based on multi-modal perception

Abstract

The invention relates to the technical field of unmanned aerial vehicle positioning, in particular to an unmanned aerial vehicle low-altitude operation positioning method and system based on multi-modal sensing. The method comprises the steps of firstly determining navigation system acceleration and unmanned aerial vehicle navigation system speed; respectively determining an initial speed conversion coefficient of the three-dimensional data by combining the navigation system acceleration, the unmanned aerial vehicle navigation system speed and the change of the three-dimensional data at the current moment; classifying the point cloud data to obtain a point cloud cluster; matching the volumes and types of the point cloud clusters collected at different moments to obtain autonomous movement features of each point cloud cluster; determining an environment correction coefficient at the current moment in combination with the autonomous movement characteristics and the ideal displacement of the unmanned aerial vehicle; according to the environment correction coefficient, adjusting the initial speed conversion coefficient of the three-dimensional data to obtain a correction speed conversion coefficient; and determining an estimated coordinate position at the next moment according to the correction speed conversion coefficient. According to the invention, the positioning precision of the unmanned aerial vehicle under the condition of poor signals is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle (UAV) positioning, and in particular to a method and system for positioning UAV low-altitude operations based on multimodal perception. Background Art

[0002] With the advancement of low-altitude airspace opening policies and the development of drone technology, drones (UAVs) are becoming increasingly popular in low-altitude operation scenarios such as agricultural inspections, power line inspections, security patrols, emergency rescue, and disaster assessments. However, low-altitude operation environments often experience satellite signal obstruction and environmental dynamic interference. A single sensor is unable to meet the positioning requirements during the operation. How to achieve high, real-time, and interference-resistant drone positioning has become a key link in ensuring operational safety and efficiency.

[0003] For example, in complex low-altitude operating environments such as cities, forests, and mountainous areas, drone positioning mainly relies on the GNSS module to obtain real-time location information. However, this signal is susceptible to occlusion, multipath effects, and environmental interference, resulting in reduced positioning accuracy or signal interruption, which seriously threatens the safety of drone operations. In order to ensure that drones can still achieve high-precision and continuous positioning when GNSS signals are abnormal, it is necessary to analyze the position information through multimodal data. For example, analysis can be performed by combining IMU and LiDAR data. However, in actual operations, due to the complex low-altitude environment, when adjusting the drone positioning information using LiDAR data to assist IMU data, it may be affected by moving obstacles such as flying birds, resulting in inaccurate adjustment information and affecting the positioning accuracy when satellite signals are poor. Summary of the Invention

[0004] In order to solve the technical problem that when adjusting the positioning information of a UAV, it is affected by moving obstacles, which in turn leads to poor positioning accuracy when the satellite signal is poor, the purpose of the present invention is to provide a method and system for positioning a UAV at low altitude operation based on multimodal perception. The technical solutions adopted are as follows:

[0005] In a first aspect, an embodiment of the present invention provides a method for positioning a UAV for low-altitude operation based on multimodal perception, the method comprising:

[0006] Obtaining positioning data of the UAV during low-altitude operations; analyzing periods of possible signal anomalies in the positioning data to determine a motion reference period;

[0007] Obtaining the navigation system acceleration and the UAV navigation system velocity during the motion reference period; and determining the initial velocity conversion coefficients of the three-dimensional data based on the navigation system acceleration, the UAV navigation system velocity, and the change in the three-dimensional data at the current moment;

[0008] Point cloud data in low-altitude operation is acquired, and the point cloud data is classified to obtain point cloud clusters; volumes and types of the point cloud clusters collected at different time instants are matched to obtain autonomous movement features of each point cloud cluster; in combination with the autonomous movement features and ideal displacement of the unmanned aerial vehicle, an environment correction coefficient of the point cloud data at the current time instant is determined;

[0009] According to the environment correction coefficient, an initial speed conversion coefficient of the three-dimensional data is adjusted to obtain a corrected speed conversion coefficient; according to the corrected speed conversion coefficient, an estimated coordinate position of the unmanned aerial vehicle at the next time instant is determined.

[0010] Further, the analysis of the period in which the positioning data may appear signal abnormalities, to determine the motion reference period, comprising:

[0011] The positioning data includes: PDOP value of the unmanned aerial vehicle in low-altitude operation, positioning distance between adjacent time instants, and unmanned aerial vehicle acceleration data;

[0012] In combination with the PDOP value, the positioning distance between adjacent time instants, and the unmanned aerial vehicle acceleration data, the satellite signal loss degree of the unmanned aerial vehicle is determined; when the satellite signal loss degree of the unmanned aerial vehicle is abnormal, the abnormal time instant is recorded; the period between the current time instant and the last abnormal time instant is taken as the motion reference period.

[0013] Further, the combination of the PDOP value, the positioning distance between adjacent time instants, and the unmanned aerial vehicle acceleration data to determine the satellite signal loss degree of the unmanned aerial vehicle, comprising:

[0014] The product of the PDOP value and the positioning distance between adjacent time instants is taken as the numerator, and the unmanned aerial vehicle acceleration data is taken as the denominator, and the ratio formed by the numerator and the denominator is taken as the satellite signal loss degree of the unmanned aerial vehicle.

[0015] Further, the method for obtaining the navigation system speed of the unmanned aerial vehicle is: a DCM matrix is constructed by using the point attitude angle at each time instant in the motion reference period, and the three-axis acceleration of the body system is converted to the navigation system to obtain the navigation system acceleration; the navigation system acceleration is combined with the integral method to obtain the navigation system speed of the unmanned aerial vehicle at the current time instant.

[0016] Further, the combination of the navigation system acceleration, the navigation system speed of the unmanned aerial vehicle, and the change of the three-dimensional data at the current time instant to determine the initial speed conversion coefficient of the three-dimensional data, comprising:

[0017] Any dimension data in the three-dimensional data is taken as target dimension data;

[0018] The change value of the target dimension data at the current time instant and the previous time instant is obtained as the dimension change value;

[0019] Determine the weights of the navigation system acceleration and the UAV navigation system speed according to the time length between the current moment and the previous moment, perform weighted summation on the navigation system acceleration and the UAV navigation system speed, and obtain the speed parameter value;

[0020] An initial speed conversion coefficient of the target dimensional data is determined according to the dimensional change value and the speed parameter value; wherein the dimensional change value is positively correlated with the initial speed conversion coefficient, and the speed parameter value is negatively correlated with the initial speed conversion coefficient.

[0021] Furthermore, the volume and type of point cloud clusters collected at different times are matched to obtain the autonomous movement characteristics of each point cloud cluster, including:

[0022] Match the volume and type of different types of point cloud data collected at different times to obtain the matching object of each point cloud cluster;

[0023] Get the actual displacement distance and movement direction between the point cloud cluster and the matching object at the current moment;

[0024] Calculate the similarity between the motion direction and the acceleration direction of the drone at the current moment;

[0025] Based on the current speed of the drone and the ideal displacement distance of the drone at the next moment under the current acceleration conditions;

[0026] The actual displacement distance is compared with the ideal displacement distance to obtain the initial displacement feature; and each point cloud cluster corresponds to the autonomous movement feature of the object by combining the initial displacement feature and the direction similarity.

[0027] Furthermore, the step of determining the environmental correction coefficient of the point cloud data at the current moment by combining the autonomous movement characteristics and the ideal displacement of the UAV includes:

[0028] Determine the first correction parameter according to the autonomous movement characteristics of the point cloud cluster at each moment;

[0029] Determine a second correction parameter according to the actual displacement distance and the ideal displacement distance of the point cloud cluster;

[0030] The first correction parameter and the second correction parameter are combined to obtain an environmental correction parameter.

[0031] Furthermore, the initial velocity conversion coefficient of the three-dimensional data is adjusted according to the environmental correction coefficient to obtain a corrected velocity conversion coefficient, including:

[0032] Any dimension data in the three-dimensional data is used as the target dimension data; any moment in the motion reference period is used as the target moment;

[0033] For the target moment, the initial speed conversion coefficient of the target dimension data is weighted using the environmental correction coefficient as the weight to obtain the target conversion coefficient at the target moment;

[0034] Perform negative correlation mapping on the satellite signal loss degree at the target time to obtain the conversion adjustment coefficient at the target time;

[0035] The conversion adjustment coefficient is used as the weight of the target conversion coefficient, and the target conversion coefficient is weighted to obtain the moment conversion coefficient; the sum of the moment conversion coefficients of all moments in the motion reference period is used as the corrected speed conversion coefficient of the target dimension data.

[0036] Furthermore, determining the estimated coordinate position of the drone at the next moment based on the corrected speed conversion coefficient includes:

[0037] Any dimension data in the three-dimensional data is used as the target dimension data; any moment in the motion reference period is used as the target moment;

[0038] The normalized value of the correction speed conversion coefficient of the current target dimension data is weighted by the UAV navigation system speed corresponding to the target dimension data to obtain the navigation system correction speed;

[0039] Determine the coordinates of the target dimension data at the next moment based on the navigation system correction speed, the navigation system acceleration, and the time difference between the current moment and the next moment;

[0040] The estimated coordinate position of the drone at the next moment is constructed based on the coordinates of the three-dimensional data at the next moment.

[0041] Secondly, a multimodal sensing-based UAV low-altitude operation positioning system is provided, which includes the following modules:

[0042] The time period determination module is used to obtain the positioning data of the UAV when performing low-altitude operations; analyze the time period when the positioning data may have signal anomalies, and determine the movement reference time period;

[0043] An initial determination module is configured to obtain the navigation system acceleration and the UAV navigation system velocity during a motion reference period; and determine initial velocity conversion coefficients for the three-dimensional data based on the navigation system acceleration, the UAV navigation system velocity, and the change in the three-dimensional data at the current moment;

[0044] The environmental correction module is used to obtain point cloud data during low-altitude operations and classify the point cloud data to obtain point cloud clusters. The volume and type of point cloud clusters collected at different times are matched to obtain the autonomous movement characteristics of each point cloud cluster. The environmental correction coefficient of the point cloud data at the current moment is determined by combining the autonomous movement characteristics and the ideal displacement of the UAV.

[0045] The positioning determination module is used to adjust the initial speed conversion coefficient of the three-dimensional data according to the environmental correction coefficient to obtain a corrected speed conversion coefficient; and determine the estimated coordinate position of the drone at the next moment according to the corrected speed conversion coefficient.

[0046] In a third aspect, an embodiment of the present invention provides an electronic device, comprising a memory and a processor, wherein the memory stores executable code, and when the processor executes the executable code, it implements the various possible implementations of the first aspect.

[0047] In a fourth aspect, an embodiment of the present invention provides a computer program product, which includes: computer program code, which, when running on a computer, enables the computer to execute the method in the above-mentioned first aspect or any possible implementation of the first aspect.

[0048] In a fifth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed in a computer, the computer is caused to execute various possible implementations of the first aspect.

[0049] The embodiments of the present invention have at least the following beneficial effects:

[0050] The present invention proposes a multimodal sensing-based method and system for positioning low-altitude unmanned aerial vehicle (UAV) operations. The method first analyzes received satellite signals, assesses signal anomalies at the UAV's operating position within the current low-altitude operating area, and determines a motion reference period. The method then calculates the UAV's initial velocity conversion coefficient based on coordinate change characteristics within the motion reference period and combined with IMU module monitoring data. The method then constructs a point cloud of the current environment using LiDAR data, and calculates an environmental correction coefficient based on the change characteristics of the point cloud data at different moments. The initial velocity conversion coefficient is adjusted based on the environmental correction coefficient, and the coordinate movement relationship of the current position is analyzed to obtain a corrected velocity conversion coefficient. Based on the corrected velocity conversion coefficient, the estimated position coordinates at the next moment are calculated, achieving relatively accurate UAV positioning. This method reduces the interference of IMU unit cumulative errors and the impact of autonomous moving objects on the analysis process in low-altitude environments. It can obtain a relatively accurate UAV position estimate even when the GNSS module signal is weak, providing a foundation for subsequent UAV operations such as flight control, route planning, obstacle avoidance, operating point positioning, and target tracking. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, and the advantages thereof, a brief introduction will be given to the drawings required in the description of the embodiments or the prior art. Obviously, the drawings in the following description only show some embodiments of the present application, and for those skilled in the art, other drawings can be obtained from these drawings without any creative effort.

[0052] Figure 1 A method flowchart of a multi-modal perception-based unmanned aerial vehicle low-altitude operation positioning method provided by an embodiment of the present application;

[0053] Figure 2 An unmanned aerial vehicle attitude angle data schematic diagram provided by an embodiment of the present application;

[0054] Figure 3 A system block diagram of a multi-modal perception-based unmanned aerial vehicle low-altitude operation positioning system provided by an embodiment of the present application;

[0055] Figure 4 A structural schematic diagram of a computer device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0056] In order to further illustrate the technical means and effects adopted by the present application to achieve the predetermined purposes, the multi-modal perception-based unmanned aerial vehicle low-altitude operation positioning method and system according to the present application, its specific implementation, structure, features and effects are described in detail as follows in combination with the drawings and preferred embodiments.

[0057] In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0058] In the description of the embodiments of the present application, unless otherwise specified, " / " represents the meaning of or, for example, A / B can represent A or B: "and / or" in the text only describes the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B, which means that there are three cases of A alone, A and B together, and B alone. In addition, in the description of the embodiments of the present application, "multiple" means two or more than two.

[0059] Hereinafter, the terms "first" and "second" are only used for descriptive purposes, and cannot be understood as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Therefore, the features defined with "first" and "second" can explicitly or implicitly include one or more features.

[0060] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0061] The embodiments of the present invention are described below with reference to the accompanying drawings. Those skilled in the art will appreciate that, with the development of technology and the emergence of new scenarios, the technical solutions provided by the embodiments of the present invention are also applicable to similar technical problems.

[0062] The specific scheme of the low-altitude operation positioning method and system of UAV based on multimodal perception provided by the present invention is described in detail below with reference to the accompanying drawings.

[0063] Example 1:

[0064] See also Figure 1 , which shows a flowchart of a method for positioning a low-altitude UAV operation based on multimodal perception according to an embodiment of the present invention, the method comprises the following steps:

[0065] Step S100: obtaining positioning data of the UAV during low-altitude operation; analyzing the time period when the positioning data may have signal anomalies, and determining a motion reference time period.

[0066] In order to achieve flight positioning of drones during low-altitude operations, the use of a single sensor is prone to interference from specific environmental factors, resulting in failure or a sharp drop in accuracy, such as satellite signal obstruction, loss of visual texture, LiDAR encountering specific materials, etc.; therefore, it is necessary to install sensors on the drone to collect multimodal data when the drone is performing low-altitude operations.

[0067] Obtain the current low-altitude operation target location area and use the A* algorithm to perform path planning to generate the drone's initial operation route. It should be noted that the process of obtaining the drone's path using the A* algorithm is an existing step and will not be elaborated on here.

[0068] Install multiple sensors, including a GNSS (Global Navigation Satellite System) module (such as the U-blox F9P), an IMU (Inertial Measurement Unit) such as the ADIS16470, and a LiDAR (Livox Mid-360), to collect the drone's position data as it flies along the initial operational route.

[0069] Among them: the GNSS antenna is installed on the top of the drone to ensure that it is unobstructed; the IMU is installed in the core part of the flight control and fixed in the same direction; the LiDAR is installed in the front of the drone to ensure that the field of view is unobstructed.

[0070] A checkerboard calibration plate is used to determine the relative positions of the GNSS antenna phase center, IMU centroid, and LiDAR coordinate system, with the error controlled within ±1 cm.

[0071] The PTP protocol is used to achieve microsecond-level hardware synchronization, and the IMU (200Hz), LiDAR (20Hz), and GNSS (10Hz) are aligned through FPGA trigger signals.

[0072] After the drone is started, each sensor collects data in real time. The data collected by each sensor mainly includes:

[0073] GNSS: Receives satellite signals through the antenna, analyzes and outputs WGS-84 coordinates (including latitude, longitude, altitude, and PDOP);

[0074] IMU: obtain acceleration and angular velocity;

[0075] LiDAR: Scans the surrounding point cloud and outputs point cloud frames with synchronized timestamps.

[0076] The collected posture data is transmitted to the data processing module of the flight control main board, and the data is cleaned and processed. Then the angular velocity data is used to obtain the attitude angle data of the drone. The specific calculation process is the existing integration method, including the yaw angle (Yaw), roll angle (Roll) and pitch angle (Pitch). Figure 2 , which is a schematic diagram of the collected UAV attitude angle data.

[0077] The camera carried by the drone is used to collect images in real time and transmit them to the data processing module of the flight control main board, where non-local mean filtering is used to reduce the noise of the images.

[0078] When operating at low altitude, the GNSS module signal of the UAV is easily interfered with by the complex low-altitude environment, resulting in poor signal accuracy and affecting the judgment of the UAV's current posture state. However, the UAV's flight control, real-time route planning, obstacle avoidance, work point positioning, target tracking and other operations all rely on the real-time measurement of its own posture. Therefore, in the low-altitude operating environment, it is necessary to analyze the current posture in combination with the monitoring data of other sensing equipment.

[0079] During actual low-altitude drone operations, the GNSS module analyzes the position data, and then uses the IMU and LiDAR sensor data to assist in positioning the drone. This precise positioning requires a certain level of accuracy in the satellite positioning data.

[0080] In addition, the LiDAR or camera onboard a drone is usually designed to scan downward or forward, and the environmental data collected is mainly for the terrain below the drone or obstacles in front. However, in low-altitude operating environments, some environmental obstacles may also exist above the drone. When there are obstacles above the drone, the accuracy of the GNSS signal is significantly affected by occlusion, multipath, and attitude interference. Therefore, the data change characteristics of the GNSS module can be used to analyze the obstruction of environmental obstacles above the drone.

[0081] The Position Dilution of Precision (PDOP) value is a numerical indicator that measures the geometric distribution of satellites used for positioning. It is one of the output data of the satellite positioning process. It is only related to the position of the satellites relative to the receiver. A larger PDOP value indicates that the satellite distribution is more concentrated or closer to a straight line or plane, and the geometric configuration is poorer. Measurement errors will be significantly amplified, and the potential positioning accuracy will be lower.

[0082] In addition, GNSS signals are also affected by occlusion and multipath effects. In low-altitude environments, these effects are mainly manifested in a reduction in the number of satellite signals received by the GNSS module and jumpy changes in positioning data. However, during low-altitude operations, drones usually travel at non-uniform speeds, so the current positioning residual of the drone should also be analyzed in combination with the acceleration data measured by the IMU unit.

[0083] Obtain the PDOP value calculated by the GNSS module when the drone is operating at low altitude. Calculate the Euclidean distance between the current GNSS module's position data and the previous position data as the positioning distance between adjacent moments. Position data includes longitude, latitude, and altitude.

[0084] Get the drone acceleration data at the last moment. It should be noted that the drone acceleration data is a normalized value.

[0085] The satellite signal loss degree of the drone is determined by combining the PDOP value, the positioning distance between adjacent moments, and the drone acceleration data.

[0086] In this embodiment of the present invention, the product of the PDOP value and the positioning distance between adjacent moments is used as the numerator, and the UAV acceleration data is used as the denominator. The numerator and denominator form a ratio, which is used as the satellite signal loss degree of the UAV.

[0087] The calculation formula for the satellite signal loss degree is: Among them, Z is the satellite signal loss degree; P is the PDOP value; D is the positioning distance between adjacent moments; a is the UAV acceleration data; and tanh is the hyperbolic tangent function.

[0088] The greater the satellite signal loss degree Z of the current position of the unmanned aerial vehicle, the stronger the shielding interference of the GNSS positioning signal, the lower the accuracy of the signal positioning, and the higher the environmental shielding degree above the unmanned aerial vehicle in the low-altitude operation environment and the more complex the current low-altitude operation environment.

[0089] When the satellite signal loss degree of the unmanned aerial vehicle is greater than the preset loss threshold , it is determined that the satellite signal loss degree of the unmanned aerial vehicle is abnormal at this time, and the moment is taken as an abnormal moment, indicating that the positioning signal received by the GNSS module at this time needs to analyze the data of the IMU and the LiDAR sensor to determine the positioning information of the unmanned aerial vehicle. In the embodiment of the application, the preset loss threshold is 0.45, and in other embodiments, the value can be adjusted by the implementer according to the actual situation.

[0090] The period between the current time point and the time point when the last satellite signal loss degree is greater than is taken as a motion reference period. That is, the period between the current time and the last abnormal moment is taken as a motion reference period. It should be noted that the motion reference period does not contain the current time.

[0091] In step S200, a DCM matrix is constructed using the point attitude angle at each time in the motion reference period, the three-axis acceleration of the aircraft body is converted to the navigation system, the navigation system acceleration and the navigation system speed of the unmanned aerial vehicle are obtained, and the initial speed conversion coefficient of the three-dimensional data is determined respectively by combining the navigation system acceleration, the navigation system speed of the unmanned aerial vehicle, and the change of the three-dimensional data at the current time.

[0092] During low-altitude operation, when the GNSS signal is lost, the position of the unmanned aerial vehicle needs to be analyzed by combining the IMU and LiDAR data to realize continuous positioning. The coordinate change characteristics of the period when the GNSS module signal is good are selected, and the initial speed conversion coefficient of the unmanned aerial vehicle is calculated by combining the monitoring data of the IMU module.

[0093] When the GNSS module signal is poor, the current positioning situation can be analyzed by the acceleration data and direction of the current unmanned aerial vehicle. First, the speed of the IMU sensor when the GNSS signal is relatively ideal needs to be evaluated: the relationship between the change of the positioning coordinates of the unmanned aerial vehicle at adjacent time and the flight speed is analyzed, which can provide a reference for the unmanned aerial vehicle navigation positioning process embodied by the IMU data sensor when the GNSS signal is poor.

[0094] When analyzing the correlation between drone coordinate changes and speed conditions, it is also necessary to combine the current drone's three-axis angular velocity for analysis: the speed and acceleration direction measured by the IMU are relative to the aircraft system, and the drone's attitude changes in real time, resulting in a constant change in the directional correspondence between the aircraft system and the geographic coordinate system. If the IMU velocity data is not converted to the geographic coordinate system using attitude angles (Roll, Pitch, Yaw) in real time, the actual relationship between the velocity vector and the GNSS coordinate changes cannot be correctly reflected.

[0095] During the motion reference period, a DCM matrix is ​​constructed based on the attitude angles (yaw, pitch, and roll) at each moment. This matrix is ​​then used to convert the three-axis acceleration of the aircraft system to the navigation system (ENU), yielding the navigation system acceleration. This conversion follows the existing method, specifically by multiplying the column vectors of the three-axis accelerations by the DCM matrix, corresponding to the accelerations in the east, north, and up directions, respectively. It should be noted that the east, north, and up directions correspond to longitude, latitude, and altitude in the following order: east corresponds to longitude, north corresponds to latitude, and up corresponds to altitude.

[0096] Subtract the gravitational acceleration constant G from the upward acceleration to obtain the navigation system acceleration after deducting the gravity component, which is recorded as A E ,A N ,A U ; Use the integration method combined with the navigation system acceleration to calculate the current UAV navigation system speed, which is recorded as V E ,V N ,V U In the embodiment of the present invention, the value of the gravitational acceleration constant G is 9.81m / s 2 .

[0097] Get the difference in longitude, latitude, and altitude between each time point and the previous time point, and record them as ΔE, ΔN, and ΔU respectively.

[0098] The initial velocity conversion coefficients for the 3D data are determined by combining the navigation system acceleration, the drone's navigation system velocity, and the current 3D data changes. It should be noted that the 3D data here refers to longitude, latitude, and altitude. The initial velocity conversion coefficients for the 3D data include: the initial velocity conversion coefficient for longitude, the initial velocity conversion coefficient for latitude, and the initial velocity conversion coefficient for altitude.

[0099] Using any dimension data in the three-dimensional data as the target dimension data;

[0100] The change value of the target dimension data between the current moment and the previous moment is obtained as the dimension change value. Specifically, the difference between the target dimension data between the current moment and the previous moment is used as the dimension change value.

[0101] Determine the weights of the navigation system acceleration and the UAV navigation system speed according to the time length between the current moment and the previous moment, perform weighted summation on the navigation system acceleration and the UAV navigation system speed, and obtain the speed parameter value;

[0102] An initial speed conversion coefficient of the target dimensional data is determined according to the dimensional change value and the speed parameter value; wherein the dimensional change value is positively correlated with the initial speed conversion coefficient, and the speed parameter value is negatively correlated with the initial speed conversion coefficient.

[0103] In the embodiment of the present invention, the ratio of the dimension change value to the speed parameter value may be used as the initial speed conversion coefficient.

[0104] In some embodiments of the present invention, the initial speed conversion coefficient I corresponding to the longitude data E The calculation formula is: Where ΔE is the latitude change value of the longitude data; t is the time length between the current moment and the previous moment; t 2 ×A E +t×V E is the speed parameter value of the longitude data; A E is the navigation system acceleration of the longitude data; V E The navigation speed of the UAV is the longitude data.

[0105] In some embodiments of the present invention, the initial speed conversion coefficient I corresponding to the latitude data N The calculation formula is: Among them, ΔN is the dimensional change value of the latitude data; t is the time length between the current moment and the previous moment; t 2 ×A N +t×V N is the speed parameter value of latitude data; A N V is the navigation system acceleration of the latitude data; N The navigation speed of the UAV is the latitude data.

[0106] In some embodiments of the present invention, the initial speed conversion coefficient I corresponding to the altitude data is N The calculation formula is: Among them, ΔU is the dimensional change value of the height data; t is the time length between the current moment and the previous moment; t 2 ×A U +t×V U is the speed parameter value of the altitude data; A U is the navigation system acceleration of altitude data; V U The speed of the drone navigation system is the altitude data.

[0107] Step S300: Acquire point cloud data and classify the point cloud data; match the volume and type of different types of point cloud data collected at different times to obtain the autonomous movement characteristics of each type of point cloud data; perform environmental correction on the point cloud data at the current moment in combination with the autonomous movement characteristics and the ideal displacement of the drone to obtain an environmental correction coefficient; adjust the initial speed conversion coefficient of the three-dimensional data based on the environmental correction coefficient to obtain a corrected speed conversion coefficient; and determine the estimated coordinate position at the next moment based on the corrected speed conversion coefficient.

[0108] Due to the error accumulation problem in the IMU measurement process, the speed calculated by the IMU for a long time will gradually deviate from the true value. It is difficult to obtain an accurate position estimate only through the data of the IMU sensor. Therefore, it is also necessary to optimize the initial coordinate displacement coefficient through the change characteristics of the environmental point cloud constructed by the LiDAR data. The current environmental point cloud is constructed using the LiDAR data, and the environmental correction coefficient is calculated based on the change characteristics of the point cloud data at different times. Among them, the reflection intensity is first normalized in the processing of the LiDAR scan data, and the reflection intensity threshold is preset to 0.2. In other embodiments, the implementer can also adjust the threshold according to the actual situation. The preset reflection intensity threshold value range is 0.1 to 0.3; the number of echoes is set to be less than or equal to 2, usually set to 1 or 2.

[0109] However, in low-altitude working environments, LiDAR may scan point cloud data of moving objects such as flying birds. When adjusting the initial speed conversion coefficient based on the distance changes between the drone and the same part of the point cloud data at different time points, it is necessary to analyze the data change characteristics of different objects in the point cloud data to reduce the impact of object displacement on the drone's speed correction assessment.

[0110] First, establish a point cloud space within the motion reference period. For the scan data within the motion reference period: convert the raw distance data from the LiDAR scan into three-dimensional spatial coordinates (X, Y, Z), and record the reflection intensity and echo count of each point to generate raw point cloud data in LAS format.

[0111] Use point cloud processing tools to remove noise points. Point cloud processing tools such as LiDAR360 use a statistical outlier removal (SOR) algorithm to reduce data redundancy, improve point cloud density uniformity, and improve computational efficiency. In this embodiment of the present invention, the k value in the statistical outlier filtering algorithm ranges from 10 to 30. When the scene density is high, k is set to 30, and when the scene density is low, k is set to 10. The implementer determines the k value within this range based on actual conditions. The outlier determination criterion multiple ranges from 1.0 to 2.0, with a common default value of 1.0.

[0112] The DCM matrix derived from GNSS positioning and IMU data is used for coordinate transformation. With the LiDAR as the center, the point cloud data corresponding to the current moment is converted from the sensor coordinate system to the global coordinate system, obtaining the longitude, latitude, and altitude data for each point in the point cloud. In this embodiment of the present invention, the pose data accuracy during the coordinate transformation is less than 0.05m in translation error and less than 0.5° in attitude error.

[0113] The RangeNet++ algorithm performs semantic segmentation on the point cloud data corresponding to the current moment, outputting the semantic category label for each voxel point, and obtaining a point cloud with category labels. In this embodiment of the present invention, the semantic categories are defined as: ground, vehicle, pedestrian, vegetation (trees and shrubs), building, pole (telephone poles and signal poles), and others. For details, please refer to the SemanticKITTI standard. The RangeNet++ network structure is: encoder: Darknet-53, output dimension: 64, batch size: 8-16, learning rate: 0.001, input resolution: 64×1024 (vertical×horizontal).

[0114] Cluster voxels with the same semantic category label in the current point cloud data to obtain several clusters. Each cluster is recorded as a point cloud cluster, and each cluster is considered to be the same object. In this embodiment of the present invention, K-means clustering can be used. The cluster relationship is the Euclidean distance between voxels. The clustering parameter K is used to obtain several clusters through the elbow method.

[0115] According to the number of voxel points in the point cloud cluster with the same semantic category label at the current moment and the next moment, the volume similarity of the objects corresponding to the point cloud cluster with the same semantic category label is determined.

[0116] For an object corresponding to a certain point cloud cluster in the point cloud data corresponding to the current moment, in the point cloud cluster corresponding to the next moment, calculate the volume similarity V between the object corresponding to the current point cloud cluster and the object corresponding to the point cloud cluster with the same semantic category label kl , in, Indicates the number of voxel points in the k-th point cloud cluster at the t-th time; Represents the number of voxel points in the lth point cloud cluster with the same semantic category label as the kth object at the tth time.

[0117] And calculate the chamfer distance (Chamfer Distance) between two point cloud clusters with the same semantic category label, denoted as C kl .

[0118] The volume similarity and the chamfer distance are combined to determine the matching relationship between objects corresponding to two point cloud clusters with the same semantic category label at the current moment and the next moment.

[0119] In the embodiment of the present invention, the inverse of the product of the volume similarity and the chamfer distance is used as the matching relationship between the objects corresponding to two point cloud clusters with the same semantic category label.

[0120] According to the matching relationship, the matching objects corresponding to all point cloud clusters at the current moment are determined. Specifically:

[0121] The point cloud cluster at the t+1th moment with the largest matching relationship for the same semantic category label is selected as the matching object of the kth point cloud cluster at the tth moment, and the matching relationship is marked to avoid many-to-many matching.

[0122] After obtaining the matching objects, the autonomous movement characteristics of the objects corresponding to each point cloud cluster are evaluated by combining the posture and flight direction of the drone between the two moments.

[0123] First, obtain the actual displacement distance and movement direction between the point cloud cluster and the matching object at the current moment. Specifically: taking the kth point cloud cluster at the tth moment as an example, the actual displacement distance between the kth point cloud cluster at the tth moment and the matching object at the t+1th moment is the Euclidean distance between the center of mass of the two point cloud clusters; and the direction from the center of mass of the kth point cloud cluster at the tth moment to the center of mass of the matching object at the t+1th moment is taken as the movement direction.

[0124] Calculate the directional similarity between the motion direction and the acceleration direction of the drone at the current moment; and use the cosine similarity between the motion direction and the acceleration direction of the drone at the current moment as the directional similarity.

[0125] The ideal displacement distance at the next moment is determined based on the speed of the drone at the current moment and the current acceleration conditions of the drone. It should be noted that determining the position of an object at the next moment based on the position of the object at the current moment, the speed of the object, and the acceleration of the object is an existing technology and will not be repeated here.

[0126] The actual displacement distance is compared with the ideal displacement distance to obtain the initial displacement feature; and each point cloud cluster corresponds to the autonomous movement feature of the object by combining the initial displacement feature and the direction similarity.

[0127] In some embodiments, the initial displacement feature is used as the numerator, the direction similarity is used as the denominator, and the normalized value of the corresponding ratio is used as the autonomous movement feature of the object corresponding to the point cloud cluster.

[0128] The calculation formula of the autonomous movement feature M is: Among them, norm is the normalization function; D is the actual displacement distance; is the ideal displacement distance; α is the direction similarity; is the initial displacement characteristic.

[0129] The larger the initial displacement feature, the greater the probability that the kth point cloud cluster at time t has autonomous mobility. When selecting points in the point cloud to correct the drone's velocity, the calculation weight of objects with large autonomous mobility features should be reduced to improve correction accuracy and reduce the interference of LiDAR data uncertainty of autonomous moving objects on the correction.

[0130] The first correction parameter is determined based on the autonomous motion characteristics of the point cloud cluster at each moment. Specifically, for any point cloud cluster at the current moment, the result of negative correlation mapping of the autonomous motion characteristics of the point cloud cluster is used as the first correction parameter. This negative correlation mapping is used to increase the calculation weight of objects without autonomous motion characteristics.

[0131] Determine a second correction parameter according to the actual displacement distance and the ideal displacement distance of the point cloud cluster; specifically, use the ratio of the actual displacement distance and the ideal displacement distance of the point cloud cluster as the second correction parameter;

[0132] The second correction parameter represents the difference between the actual displacement distance and the ideal displacement distance. It serves as the correction coefficient of the radar data to the IMU velocity data. The larger the difference, the farther the second correction parameter is from 1, and the stronger the subsequent correction of the initial velocity conversion coefficient.

[0133] The first and second correction parameters are combined to obtain the environmental correction parameter. For any point cloud cluster at the current moment, the product of the first and second correction parameters of the point cloud cluster is used as the single correction factor of the point cloud cluster. The sum of the single correction factors of all point cloud clusters at the current moment is used as the environmental correction coefficient at the current moment.

[0134] In some embodiments, the calculation formula of the environmental correction coefficient E is: Among them, m is the number of point cloud clusters at the current moment; M j is the autonomous movement feature of the j-th point cloud cluster; D j is the actual displacement distance of the j-th point cloud cluster; is the ideal displacement distance; 1-M j is the first correction parameter; is the second correction parameter; is the single correction factor of the j-th point cloud cluster.

[0135] Step S400 , adjusting the initial speed conversion coefficient of the three-dimensional data according to the environmental correction coefficient to obtain a corrected speed conversion coefficient; and determining an estimated coordinate position at the next moment according to the corrected speed conversion coefficient.

[0136] After obtaining the environmental correction coefficient, the initial speed conversion coefficient is adjusted according to the environmental correction coefficient, and the coordinate movement relationship of the current position is analyzed to obtain the corrected speed conversion coefficient.

[0137] After obtaining the environmental correction coefficient of the point cloud of each frame, the initial speed conversion coefficient is corrected by the environmental correction coefficient to evaluate the actual relationship between the IMU data and the GNSS coordinate changes in the current low-altitude working environment.

[0138] In the motion reference period, the initial speed conversion coefficient is adjusted according to the environmental correction coefficient E corresponding to each moment to obtain the corrected speed conversion coefficient.

[0139] Any dimension data in the three-dimensional data is used as the target dimension data; any moment in the motion reference period is used as the target moment;

[0140] For the target moment, the initial speed conversion coefficient of the target dimension data is weighted using the environmental correction coefficient as the weight to obtain the target conversion coefficient at the target moment;

[0141] The satellite signal loss degree at the target time is negatively mapped to obtain the conversion adjustment coefficient at the target time. The satellite signal loss degree after negative correlation mapping is used to amplify the calculation result of the GNSS coordinate with higher credibility when the satellite signal loss degree is small, and the actual relationship between the IMU data and the GNSS coordinate changes in the current operating environment during the past motion reference period is calculated.

[0142] The conversion adjustment coefficient is used as the weight of the target conversion coefficient to obtain the weighted moment conversion coefficient. The sum of the moment conversion coefficients for all moments within the motion reference period is used as the corrected velocity conversion coefficient for the target dimensional data. The greater the distance between the corrected velocity conversion coefficient and 1, the greater the degree of cumulative error in the IMU velocity of the drone's three-dimensional data at the current moment. In other words, the greater the degree of cumulative error in the IMU velocity in the longitude, latitude, and altitude directions, the greater the degree of adjustment required when subsequently calculating the drone's position at the next moment through the IMU.

[0143] Taking the i-th moment as the current moment as an example, the specific adjustment process is:

[0144]

[0145] Among them, I · E Z is the correction speed conversion coefficient of the longitude data at the current moment;i is the satellite signal loss degree at the current moment; E i is the environmental correction coefficient at the current moment; I Ei is the initial velocity conversion coefficient of the longitude data at the current moment; n is the number of moments in the motion reference period; I · N I is the correction speed conversion coefficient of the current latitude data; Ni is the initial velocity conversion coefficient of the latitude data at the current moment; I · U I is the corrected speed conversion coefficient of the altitude data at the current moment; Ui The initial velocity conversion coefficient of the altitude data at the current moment; 1-Z i is the conversion adjustment coefficient at the current moment; E i ×I Ei is the target conversion coefficient of the longitude data at the current moment; (1-Z i )×(E i ×I Ei ) is the time conversion coefficient of the longitude data at the current time; E i ×I Ni is the target conversion coefficient of the latitude data at the current moment; (1-Z i )×(E i ×I Ni ) is the moment conversion coefficient of the latitude data at the current moment; E i ×I Ui is the target conversion coefficient of the height data at the current moment; (1-Z i )×(E i ×I Ui ) is the moment conversion coefficient of the altitude data at the current moment.

[0146] Finally, the estimated coordinate position at the next moment is determined according to the modified speed conversion coefficient.

[0147] Use the UAV navigation system speed and acceleration obtained by the IMU at the current position to estimate the longitude, latitude, and altitude coordinates at the next moment. Take the longitude coordinate as an example: Where X represents the longitude coordinate of the next moment, X0 is the longitude coordinate of the current moment, is the normalized value of the correction speed conversion coefficient of the current longitude speed, V and A are the navigation system speed and navigation system acceleration of the UAV at the current latitude data, respectively, and t is the time difference between the current moment and the next moment.

[0148] Then the estimated coordinates of each dimension at the next moment are obtained, that is, the coordinates in each dimension are obtained, and then the estimated coordinate position of the drone at the next moment is obtained. For example, the longitude coordinate at the next moment is c1, the latitude coordinate at the next moment is c2, and the altitude coordinate at the next moment is c3. Then the estimated coordinate position of the drone at the next moment (c1, c2, c3) is obtained from c1, c2 and c3, realizing the positioning of the drone.

[0149] This method for estimating coordinate position reduces the interference of IMU unit cumulative errors and the impact of autonomous moving objects on the analysis process in low-altitude environments. It can obtain a relatively accurate drone position estimate even when the GNSS module signal is weak. This provides the basis for subsequent drone operations such as flight control, route planning, obstacle avoidance, work point positioning, and target tracking. Accurately knowing a drone's own position and attitude is the prerequisite and foundation for safe flight, stable hovering, path planning, and precise operations. Without accurate localization, a drone cannot reliably navigate or interact with its environment or targets.

[0150] Example 2:

[0151] See also Figure 3 , which shows a system block diagram of a UAV low-altitude operation positioning system based on multimodal perception provided by an embodiment of the present invention. The system includes the following modules:

[0152] The time period determination module is used to obtain the positioning data of the UAV when performing low-altitude operations; analyze the time period when the positioning data may have signal anomalies, and determine the movement reference time period;

[0153] An initial determination module is configured to obtain the navigation system acceleration and the UAV navigation system velocity during a motion reference period; and determine initial velocity conversion coefficients for the three-dimensional data based on the navigation system acceleration, the UAV navigation system velocity, and the change in the three-dimensional data at the current moment;

[0154] The environmental correction module is used to obtain point cloud data during low-altitude operations and classify the point cloud data to obtain point cloud clusters. The volume and type of point cloud clusters collected at different times are matched to obtain the autonomous movement characteristics of each point cloud cluster. The environmental correction coefficient of the point cloud data at the current moment is determined by combining the autonomous movement characteristics and the ideal displacement of the UAV.

[0155] The positioning determination module is used to adjust the initial speed conversion coefficient of the three-dimensional data according to the environmental correction coefficient to obtain a corrected speed conversion coefficient; and determine the estimated coordinate position of the drone at the next moment according to the corrected speed conversion coefficient.

[0156] Optionally, the transmission medium can be a wired link, such as but not limited to coaxial cable, optical fiber and digital subscriber line, or a wireless link, such as but not limited to Wireless Fidelity (WIFI), Bluetooth and mobile device network.

[0157] It should be noted that the device provided in the above embodiment is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above.

[0158] Figure 4 FIG. 1 is a schematic diagram of the structure of a computer device provided by an embodiment of the present invention. For example, Figure 4 As shown, the computer device 500 includes: a memory 510, a processor 520, and a computer program 530 stored in the memory 510 and running on the processor 520, wherein when the processor 520 executes the computer program 530, the computer device can execute any of the multimodal perception-based low-altitude operation positioning methods for drones introduced above.

[0159] In addition, an embodiment of the present invention also protects a device, which may include a memory and a processor, wherein the memory stores executable program code, and the processor is used to call and execute the executable program code to execute the low-altitude operation positioning method of a drone based on multimodal perception provided by an embodiment of the present invention.

[0160] In embodiments of the present invention, the device may be divided into functional modules based on the above-described method examples. For example, these modules may correspond to individual functional modules, or two or more functions may be integrated into a single processing module. The integrated modules may be implemented in hardware. It should be noted that the module division in this embodiment is illustrative and represents only a logical functional division. In actual implementation, other division methods may be employed.

[0161] In the case of dividing each module into modules corresponding to each function, the device may further include a signal uploading module, a determination module, an adjustment module, etc. It should be noted that all relevant contents of each step involved in the above method embodiment can be referred to the functional description of the corresponding functional module and will not be repeated here.

[0162] It should be understood that the device provided in the embodiment of the present invention is used to execute the above-mentioned UAV low-altitude operation positioning method based on multimodal perception, and therefore can achieve the same effect as the above-mentioned implementation method.

[0163] In the case of an integrated unit, the device may include a processing module and a storage module. When the device is applied to a device, the processing module may be used to control and manage the operation of the device. The storage module may be used to support the device in executing mutual program codes, etc. The processing module may be a processor or a controller that may implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the present disclosure. The processor may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of digital signal processing (DSP) and a microprocessor, etc. The storage module may be a memory.

[0164] In addition, the device provided in an embodiment of the present invention may specifically be a chip, component or module, and the chip may include a connected processor and memory; wherein the memory is used to store instructions, and when the processor calls and executes the instructions, the chip can execute the UAV low-altitude operation positioning method based on multimodal perception provided in the above embodiment.

[0165] An embodiment of the present invention also provides a computer-readable storage medium, which stores computer program code. When the computer program code is run on a computer, the computer executes the above-mentioned related method steps to implement the low-altitude operation positioning method of the unmanned aerial vehicle based on multimodal perception provided by the above embodiment.

[0166] An embodiment of the present invention also provides a computer program product. When the computer program product is run on a computer, it enables the computer to execute the above-mentioned related steps to implement the low-altitude operation positioning method of the unmanned aerial vehicle based on multimodal perception provided by the above embodiment.

[0167] Among them, the device, computer-readable storage medium, computer program product or chip provided in the embodiments of the present invention are all used to execute the corresponding methods provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding methods provided above, and will not be repeated here. Through the description of the above implementation methods, technical personnel in the relevant field can understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In the embodiments provided by the present invention, it should be understood that the disclosed device and method can be implemented in other ways.

[0168] The apparatus embodiments described above are merely exemplary, for example, the division of modules or units is merely a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another apparatus, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed each other can be indirect coupling or communication connection through some interfaces, apparatuses or units, and can be electrical, mechanical or other forms.

[0169] It should also be noted that the terms "comprising", "containing", or any other variant thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal device including a list of elements does not only include those elements, but also includes other elements not explicitly listed, or further includes elements inherent in such a process, method, article, or terminal device. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of additional identical elements in the process, method, article, or terminal device including the element.

[0170] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or can be advantageous.

[0171] Each embodiment in the specification is described in a progressive manner, and the same or similar parts between each embodiment can be referred to each other, and each embodiment mainly describes the difference from other embodiments.

[0172] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto, and any skilled person in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be covered within the protection scope of the present application.

Claims

1. A method for positioning low-altitude UAV operations based on multimodal perception, characterized in that: The method comprises the following steps: Obtaining positioning data of the UAV during low-altitude operations; analyzing periods of possible signal anomalies in the positioning data to determine a motion reference period; Obtaining the navigation system acceleration and the UAV navigation system velocity during the motion reference period; and determining the initial velocity conversion coefficients of the three-dimensional data based on the navigation system acceleration, the UAV navigation system velocity, and the change in the three-dimensional data at the current moment; Acquire point cloud data from low-altitude operations and classify it to generate point cloud clusters. Match the volume and type of point cloud clusters collected at different times to obtain the autonomous movement characteristics of each point cloud cluster. Combine the autonomous movement characteristics with the ideal displacement of the drone to determine the environmental correction coefficient of the point cloud data at the current moment. According to the environmental correction coefficient, the initial speed conversion coefficient of the three-dimensional data is adjusted to obtain a corrected speed conversion coefficient; according to the corrected speed conversion coefficient, the estimated coordinate position of the UAV at the next moment is determined.

2. The method for positioning a UAV at low altitude operation based on multimodal perception according to claim 1, characterized in that: The analyzing the time period during which signal anomalies may occur in the positioning data and determining the motion reference time period includes: The positioning data includes: the PDOP value when the UAV is performing low-altitude operations, the positioning distance between adjacent moments, and the UAV acceleration data; The satellite signal loss degree of the drone is determined by combining the PDOP value, the positioning distance between adjacent moments, and the drone acceleration data. When the satellite signal loss degree of the drone is abnormal, the abnormal time is recorded. The period between the current moment and the previous abnormal moment is used as the motion reference period.

3. The method for positioning a UAV at low altitude operation based on multimodal perception according to claim 2, characterized in that: Determining the satellite signal loss degree of the drone by combining the PDOP value, the positioning distance between adjacent moments, and the drone acceleration data includes: The product of the PDOP value and the positioning distance between adjacent moments is used as the numerator, and the UAV acceleration data is used as the denominator. The ratio formed by the numerator and denominator is used as the satellite signal loss degree of the UAV.

4. The method for positioning a UAV at low altitude operation based on multimodal perception according to claim 1, characterized in that: The method for obtaining the speed of the UAV navigation system is as follows: constructing a DCM matrix using the attitude angle of each point at each moment in the motion reference period, converting the three-axis acceleration of the aircraft system into the navigation system, and obtaining the navigation system acceleration; The navigation system velocity of the UAV at the current moment is obtained by combining the navigation system acceleration with the integration method.

5. The method for positioning a UAV at low altitude operation based on multimodal perception according to claim 1, characterized in that: The determining of the initial velocity conversion coefficient of the three-dimensional data by combining the navigation system acceleration, the UAV navigation system velocity, and the change of the three-dimensional data at the current moment includes: Using any dimension data in the three-dimensional data as the target dimension data; Get the change value of the target dimension data between the current moment and the previous moment as the dimension change value; Determine the weights of the navigation system acceleration and the UAV navigation system speed according to the time length between the current moment and the previous moment, perform weighted summation on the navigation system acceleration and the UAV navigation system speed, and obtain the speed parameter value; An initial speed conversion coefficient of the target dimensional data is determined according to the dimensional change value and the speed parameter value; wherein the dimensional change value is positively correlated with the initial speed conversion coefficient, and the speed parameter value is negatively correlated with the initial speed conversion coefficient.

6. The method for positioning a UAV at low altitude operation based on multimodal perception according to claim 1, characterized in that: The volume and type of point cloud clusters collected at different times are matched to obtain the autonomous movement characteristics of each point cloud cluster, including: Match the volume and type of different types of point cloud data collected at different times to obtain the matching object of each point cloud cluster; Get the actual displacement distance and movement direction between the point cloud cluster and the matching object at the current moment; Calculate the similarity between the motion direction and the acceleration direction of the drone at the current moment; Based on the current speed of the drone and the ideal displacement distance of the drone at the next moment under the current acceleration conditions; The actual displacement distance is compared with the ideal displacement distance to obtain the initial displacement feature; and each point cloud cluster corresponds to the autonomous movement feature of the object by combining the initial displacement feature and the direction similarity.

7. The method for positioning a UAV at low altitude operation based on multimodal perception according to claim 6, characterized in that: The step of combining the autonomous movement characteristics and the ideal displacement of the UAV to determine the environmental correction coefficient of the point cloud data at the current moment includes: Determine the first correction parameter according to the autonomous movement characteristics of the point cloud cluster at each moment; Determine a second correction parameter according to the actual displacement distance and the ideal displacement distance of the point cloud cluster; The first correction parameter and the second correction parameter are combined to obtain an environmental correction parameter.

8. The method for positioning a UAV at low altitude operation based on multimodal perception according to claim 2, characterized in that: The adjusting the initial velocity conversion coefficient of the three-dimensional data according to the environmental correction coefficient to obtain the corrected velocity conversion coefficient includes: Any dimension data in the three-dimensional data is used as the target dimension data; any moment in the motion reference period is used as the target moment; For the target moment, the initial speed conversion coefficient of the target dimension data is weighted using the environmental correction coefficient as the weight to obtain the target conversion coefficient at the target moment; Perform negative correlation mapping on the satellite signal loss degree at the target time to obtain the conversion adjustment coefficient at the target time; The conversion adjustment coefficient is used as the weight of the target conversion coefficient, and the target conversion coefficient is weighted to obtain the moment conversion coefficient; the sum of the moment conversion coefficients of all moments in the motion reference period is used as the corrected speed conversion coefficient of the target dimension data.

9. The method for positioning a UAV at low altitude operation based on multimodal perception according to claim 1, characterized in that: Determining the estimated coordinate position of the drone at the next moment based on the corrected speed conversion coefficient includes: Any dimension data in the three-dimensional data is used as the target dimension data; any moment in the motion reference period is used as the target moment; The normalized value of the correction speed conversion coefficient of the current target dimension data is weighted by the UAV navigation system speed corresponding to the target dimension data to obtain the navigation system correction speed; Determine the coordinates of the target dimension data at the next moment based on the navigation system correction speed, the navigation system acceleration, and the time difference between the current moment and the next moment; The estimated coordinate position of the drone at the next moment is constructed based on the coordinates of the three-dimensional data at the next moment.

10. A low-altitude UAV positioning system based on multimodal perception, characterized in that: The system includes the following modules: The time period determination module is used to obtain the positioning data of the UAV when performing low-altitude operations; Analyzing a period during which signal anomalies may occur in the positioning data to determine a motion reference period; An initial determination module is used to obtain the navigation system acceleration and the UAV navigation system velocity during the motion reference period; Determine the initial velocity conversion coefficient of the three-dimensional data by combining the navigation system acceleration, the UAV navigation system velocity, and the change of the three-dimensional data at the current moment; The environmental correction module is used to obtain point cloud data during low-altitude operations and classify the point cloud data to obtain point cloud clusters. The volume and type of point cloud clusters collected at different times are matched to obtain the autonomous movement characteristics of each point cloud cluster. Determine the environmental correction coefficient of the point cloud data at the current moment by combining the autonomous movement characteristics and the ideal displacement of the UAV; The positioning determination module is used to adjust the initial speed conversion coefficient of the three-dimensional data according to the environmental correction coefficient to obtain a corrected speed conversion coefficient; and determine the estimated coordinate position of the drone at the next moment according to the corrected speed conversion coefficient.

Images