A wireless positioning review method for unmanned aerial vehicle to patrol water level monitoring points of water conservancy projects

Abstract

The application discloses a kind of unmanned aerial vehicle tour survey water conservancy project water level monitoring point's wireless positioning rechecking method, belong to radio ranging and positioning technical field, its method is: obtaining unmanned aerial vehicle real-time attitude data and forming with the corresponding attitude displacement sequence of synchronous ranging moment, according to antenna center point coordinates and water surface geometric face model constructs first reflection wave vector, attitude displacement is mapped as reference delay interval and compared with measured propagation delay, determine abnormal reflection path;With measured propagation delay as total amount constraint joint decomposition attitude displacement influence quantity and multipath propagation influence quantity, through water surface reflection physical boundary extraction remaining reflection propagation component, according to the space angle of its propagation vector and attitude motion vector constructs multipath error correction quantity, synchronous compensation original ranging signal and carry out consistency rechecking, obtain water level monitoring point position coordinate, the advantage of the application lies in: it can reduce the influence of attitude change and water surface multipath propagation aliasing, improve the accuracy of positioning rechecking result.

Description

Technical Field

[0001] This invention relates to the field of radio ranging and positioning technology, specifically to a wireless positioning verification method for unmanned aerial vehicle (UAV) surveying water level monitoring points in water conservancy projects. Background Technology

[0002] With the application of drones in water conservancy engineering inspections, wireless ranging equipment can accompany drones to measure distances and verify the positions of water level monitoring points. During flight, the drone will undergo attitude changes such as yaw, pitch, and roll, causing the antenna phase center to undergo instantaneous geometric displacement along with the motion of the body; at the same time, water surface reflection will also superimpose multipath propagation components into the original ranging signal.

[0003] Existing technologies typically utilize attitude information output by the inertial measurement unit to perform geometric correction on the position of the antenna or ranging end, and use time synchronization, state estimation, or interpolation to make the attitude data correspond to the wireless ranging observations. For abnormal ranging observations, ranging residuals, filtering, or multipath detection methods can also be used to identify abnormal components and compensate or eliminate the original ranging observations.

[0004] However, the geometric propagation changes of the antenna phase center caused by attitude changes and the additional propagation changes caused by water surface reflections both affect the same ranging moment and manifest as propagation delay changes. Without an interpretable propagation boundary jointly determined by attitude displacement and water surface reflection geometry, it is difficult to distinguish between the two types of changes. Furthermore, if attitude compensation and multipath compensation are treated independently, the two types of influencing quantities may absorb or contaminate each other, making it difficult to determine the correction direction. Moreover, the consistency between the compensated observation and the attitude geometry lacks verification constraints, thus affecting the verification of the water level monitoring point's position.

[0005] The information disclosed in this background section is intended only to enhance the understanding of the overall background of the invention and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0006] This invention provides a wireless positioning verification method for unmanned aerial vehicle (UAV) surveying water level monitoring points in water conservancy projects. This method addresses the problems in existing technologies where it is difficult to distinguish between the geometric propagation changes of the antenna phase center attitude and the additional propagation changes of water surface reflection at the same ranging time. This leads to mutual absorption or contamination between attitude compensation and multipath compensation, difficulty in determining the correction direction, and a lack of verification constraints on the consistency between the compensated observation and the attitude geometry.

[0007] A wireless positioning verification method for unmanned aerial vehicle (UAV) inspection of water level monitoring points in water conservancy projects includes: acquiring real-time attitude data from the UAV's inertial measurement unit (IMU); calculating the instantaneous geometric distance change of the antenna phase center relative to a reference point on the UAV body; smoothly interpolating the distance at the synchronous ranging time to obtain a second attitude displacement sequence; acquiring the original ranging signal at the synchronous ranging time and extracting the measured propagation delay change; constructing a first reflected wave vector based on the antenna center point coordinates and the water surface geometry model; obtaining a propagation path difference set by multiplying the second attitude displacement sequence with the first reflected wave vector; converting the upper and lower bounds of the propagation path difference set into a reference delay interval based on the propagation speed; comparing the reference delay interval with the measured propagation delay change at the same ranging time to determine abnormal reflection paths; if an abnormal reflection path exists... Using the measured propagation delay change as the total constraint, the geometric propagation change and additional propagation change that can be explained by attitude displacement are jointly decomposed to obtain the attitude displacement influence and multipath propagation influence. The remaining reflection propagation components are extracted according to the physical boundary of water surface reflection, and their propagation vectors and the attitude motion vectors corresponding to the attitude displacement influence are extracted at the same ranging time. The correction direction is determined by the spatial angle between the two and a multipath error correction is constructed. The multipath error correction is used to synchronously compensate the original ranging signal to obtain the target ranging observation value. The attitude-explainable propagation range is re-determined and its consistency is verified based on the target ranging observation value and the UAV attitude data. The water level monitoring point positioning verification is completed with the verified target ranging observation value to obtain the position coordinates.

[0008] Optionally, the calculation of the instantaneous geometric distance change of the antenna phase center relative to the aircraft reference point includes: extracting the yaw angle, pitch angle, and roll angle from the real-time attitude data; obtaining the relative position of the aircraft reference point to the antenna phase center; constructing geometric constraints based on the relative position and the yaw angle, pitch angle, and roll angle; obtaining the instantaneous distance quantities in the three coordinate axis directions through coordinate mapping; and correcting the initial displacement value at the synchronous ranging moment based on the instantaneous distance quantities to obtain the instantaneous geometric distance change.

[0009] Optionally, the step of smoothing interpolation based on the synchronous ranging time to obtain the second attitude displacement sequence includes: extracting the synchronous ranging time from the ranging observation data of the wireless receiver and setting it as a unified time reference axis; smoothing the instantaneous geometric distance change corresponding to the unified time reference axis to obtain a smoothed displacement component; obtaining adjacent attitude sample points corresponding to the smoothed displacement component; and when the timestamp of the adjacent attitude sample point falls within a preset time interval of the unified time reference axis, performing sequence interpolation based on the time relationship between the adjacent attitude sample point and the synchronous ranging time to obtain the second attitude displacement sequence.

[0010] Optionally, obtaining the second attitude displacement sequence includes: arranging the interpolated displacement states according to the unified time reference axis to form discrete displacement state quantities; performing spatial coordinate system transformation on the discrete displacement state quantities to obtain the spatial displacement state of the antenna phase center; and matching the spatial displacement state of the antenna phase center with the wireless ranging observation data point by point to form the second attitude displacement sequence.

[0011] Optionally, constructing the first reflected wave vector and obtaining the wireless propagation path difference set includes: determining the water surface reflection point and the first reflected wave vector based on the coordinates of the antenna center point and the water surface geometry model; filtering the second attitude displacement sequence to obtain a displacement sequence characterizing attitude changes; and performing an inner product operation between the displacement sequence and the first reflected wave vector to form the wireless propagation path difference set using the resulting distance change.

[0012] Optionally, the method further includes converting the upper and lower bounds of the wireless propagation path difference set into a reference delay interval, including: extracting the maximum and minimum values ​​of the wireless propagation path difference set to form the upper and lower bounds of the path difference; converting the upper and lower bounds of the path difference into a propagation time difference interval according to the wireless signal propagation speed; when the propagation time difference interval meets the preset interval length condition, determining it as the reference delay interval, and determining the measured signal whose delay deviates from the reference delay interval as a multipath signal candidate.

[0013] Optionally, the step of comparing the consistency of the reference delay interval with the measured propagation delay change at the same ranging time includes: comparing the measured propagation delay at the same ranging time with the reference delay interval point by point to obtain the deviation; determining a threshold matching degree based on the deviation and the interval width of the reference delay interval; and extracting the reflection propagation residual that cannot be explained by the reference delay interval when the threshold matching degree does not meet the preset matching standard.

[0014] Optionally, determining the anomalous reflection path and extracting the remaining reflection propagation components includes: inputting the reflection propagation residual into an isolated forest model to obtain the residual abnormality score of the corresponding propagation path, and determining the corresponding propagation path as an anomalous reflection path when the residual abnormality score exceeds a preset score line; using the measured propagation delay change as a total constraint, separating the geometric propagation change and the additional propagation change that can be explained by the attitude displacement; performing joint state estimation on the geometric propagation change and the additional propagation change to obtain the attitude displacement influence and the multipath propagation influence; and filtering the multipath propagation influence based on the physical boundary of water surface reflection to obtain the remaining reflection propagation components.

[0015] Optionally, constructing the multipath error correction amount includes: using a random forest model to predict the multipath fading intensity based on the signal strength of the correction direction and the original ranging signal; determining the offset amplitude based on the multipath fading intensity; and combining the correction direction and the offset amplitude to form the multipath error correction amount.

[0016] Optionally, the synchronous compensation of the original ranging signal and the consistency verification include: superimposing the multipath error correction amount with the original ranging observation value at the corresponding synchronous ranging time to obtain the target ranging observation value; redetermining the attitude interpretable propagation range based on the target ranging observation value and the UAV attitude data, and performing a consistency verification on the target ranging observation value; when the verification result does not meet the preset consistency condition, updating the multipath error correction amount and recompensating until the preset consistency condition is met, and calculating the position coordinates based on the verified target ranging observation value.

[0017] The beneficial effects of this invention are as follows:

[0018] This invention unifies the attitude data and wireless ranging observation time reference at the synchronous ranging time, transforming the antenna phase center attitude displacement along the water surface reflection reference direction into an attitude-interpretable reference delay interval. It also identifies abnormal reflection paths using the measured propagation delay at the same time, distinguishing attitude geometric changes from additional propagation by water surface reflection under a unified physical boundary. After an anomaly is identified, the influence of attitude displacement and multipath propagation is jointly decomposed using the measured propagation delay change as the total constraint. The remaining reflection propagation components are filtered through the water surface reflection physical boundary, and the correction direction is determined by the spatial angle between the remaining reflection propagation vector and the attitude motion vector, forming a multipath error correction amount that directly applies to the original ranging signal. After compensation, the attitude-interpretable propagation range is redefined, and a consistency check is performed, thereby reducing the mutual contamination between attitude compensation and multipath compensation, as well as the impact of inconsistencies between the compensation results and attitude geometry on the positioning check. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the overall process of the wireless positioning verification method for water level monitoring points of water conservancy projects by unmanned aerial vehicles provided in one embodiment of the present invention;

[0020] Figure 2 This is a schematic diagram showing the alignment of the attitude displacement sequence with the synchronous ranging time provided in one embodiment of the present invention;

[0021] Figure 3 This is a schematic diagram of a water surface reflection geometric reference and reference delay interval determination provided in an embodiment of the present invention;

[0022] Figure 4This is a schematic diagram of the construction of joint state decomposition and multipath error correction amount for abnormal reflection paths provided in an embodiment of the present invention;

[0023] Figure 5 This is a schematic diagram of the original ranging synchronization compensation and consistency verification closed loop provided in one embodiment of the present invention. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only for explaining the invention and are not intended to limit the invention; that is, the described embodiments are merely some embodiments of the invention, and not all embodiments. The components of the embodiments of the invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0025] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0026] It should be noted that relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0027] As mentioned earlier, when a drone equipped with wireless ranging equipment conducts surveys of water level monitoring points in water conservancy projects, changes in the drone's attitude cause an instantaneous geometric displacement of the antenna phase center relative to the drone's reference point, altering the propagation path of the corresponding wireless ranging signal. Simultaneously, water surface reflection adds an additional propagation component to the original ranging signal at the same measurement moment. If the geometrical changes in attitude and the propagation changes from water surface reflection cannot be distinguished at the same time reference, attitude compensation and multipath compensation are prone to mutual absorption or contamination, thus affecting the verification of the water level monitoring point's location.

[0028] To address this, the present invention establishes a correspondence between the antenna phase center attitude displacement and wireless ranging observations simultaneously, converts the attitude displacement along the water surface reflection reference direction into an attitude-interpretable reference delay interval, identifies abnormal reflection paths using the measured propagation delay at the same ranging time, and jointly decomposes the attitude displacement influence and multipath propagation influence under the constraint of the total measured propagation delay. After filtering by the physical boundary of water surface reflection, the remaining reflection propagation component is formed. The correction direction is determined based on the spatial angle between the remaining reflection propagation vector and the attitude motion vector, forming a multipath error correction amount and simultaneously compensating the original ranging signal. Finally, a consistency check is performed by redetermining the attitude-interpretable propagation range.

[0029] The present invention will now be described in detail with reference to the accompanying drawings.

[0030] Example 1

[0031] like Figure 1 The diagram shows the overall flow of a wireless positioning verification method for water level monitoring points in water conservancy projects using unmanned aerial vehicles (UAVs) according to an embodiment of the present invention. The execution entity of this method can be a data processing terminal that interacts with the UAV's inertial measurement unit and wireless receiver. The method mainly includes the following steps:

[0032] S100: Acquire the real-time attitude data of the UAV's inertial measurement unit, calculate the instantaneous geometric distance change of the antenna phase center relative to the reference point of the aircraft, and smoothly interpolate according to the synchronous ranging time to obtain the second attitude displacement sequence.

[0033] S200. Acquire the original ranging signal and extract the measured propagation delay change at the synchronous ranging moment;

[0034] S300. Construct a first reflected wave vector based on the coordinates of the antenna center point and the geometric model of the water surface. Involve the second attitude displacement sequence with the first reflected wave vector to obtain the propagation path difference set, and convert its upper and lower bounds into a reference delay interval according to the propagation speed. Compare the reference delay interval with the measured propagation delay change at the same ranging time to determine the abnormal reflection path.

[0035] S400. If an abnormal reflection path exists, take the measured propagation delay change as the total constraint, and jointly decompose the geometric propagation change and additional propagation change that can be explained by the attitude displacement to obtain the attitude displacement influence amount and the multipath propagation influence amount; extract the remaining reflection propagation components according to the physical boundary of water surface reflection, and extract its propagation vector and the attitude motion vector corresponding to the attitude displacement influence amount at the same ranging time. Use the spatial angle between the two to determine the correction direction and construct the multipath error correction amount.

[0036] S500: The target ranging observation value is obtained by synchronously compensating the original ranging signal with the multipath error correction amount. The interpretable propagation range of the attitude is re-determined and the consistency is verified based on the target ranging observation value and the UAV attitude data. The water level monitoring point positioning verification is completed with the verified target ranging observation value to obtain the position coordinates.

[0037] Based on the above steps, the wireless positioning verification method provided by this invention first forms the antenna phase center attitude displacement corresponding to each point of the original ranging observation at the synchronous ranging time. Then, it uses the geometric relationship of water surface reflection to convert the attitude displacement into an interpretable propagation delay range. The abnormal reflection path is determined by the measured propagation delay at the same time. The attitude influence and multipath influence are jointly decomposed using the measured propagation delay as a common total. Subsequently, the correction direction is determined according to the spatial angle, a multipath error correction amount is formed and applied in reverse to the original ranging observation. The compensation result also needs to be re-verified through the consistency of the attitude interpretable propagation range, thus constituting a continuous process of wireless ranging error identification, decomposition, correction and verification.

[0038] Example 2

[0039] To provide a more detailed explanation of the technical solutions provided in the above embodiments, the present invention also provides Embodiment Two. For example... Figure 2 As shown, this embodiment still uses the synchronous ranging time as a unified time reference and proceeds according to the processing sequence described in Embodiment 1.

[0040] In this second embodiment, step S100 may further include:

[0041] S100: Acquire the real-time attitude data of the UAV's inertial measurement unit, calculate the instantaneous geometric distance change of the antenna phase center relative to the reference point of the aircraft, and smoothly interpolate according to the synchronous ranging time to obtain the second attitude displacement sequence.

[0042] S110. Extract the yaw angle, pitch angle and roll angle from the real-time attitude data;

[0043] S120. Obtain the relative position from the airframe reference point to the antenna phase center, construct geometric constraints based on the relative position and the yaw angle, pitch angle and roll angle, and obtain the instantaneous distance in the three coordinate axis directions through coordinate mapping.

[0044] S130. Correct the initial displacement value at the synchronous ranging time according to the instantaneous distance measurement to obtain the instantaneous geometric distance change.

[0045] S140. Extract the synchronization ranging time from the ranging observation data of the wireless receiver and set it as a unified time reference axis;

[0046] S150. Perform state smoothing on the instantaneous geometric distance change corresponding to the unified time reference axis to obtain smoothed displacement components;

[0047] S160. Obtain the adjacent attitude sample points corresponding to the smooth displacement component. When the timestamp of the adjacent attitude sample point falls within the preset time interval of the unified time reference axis, perform sequence interpolation according to the time relationship between the adjacent attitude sample points and the synchronous ranging time to obtain the second attitude displacement sequence.

[0048] S170. Arrange the displacement states after sequence interpolation according to the unified time reference axis to form discrete displacement state quantities;

[0049] S180. Perform spatial coordinate system transformation on the discrete displacement state quantity to obtain the spatial displacement state of the antenna phase center.

[0050] S190. Match the spatial displacement state of the antenna phase center with the wireless ranging observation data point by point to form the second attitude displacement sequence.

[0051] For example, steps S110 to S190 may further include the following processes:

[0052] First, the UAV's inertial measurement unit continuously outputs timestamped yaw, pitch, and roll angles. The data processing unit reads these three attitude angles according to a unified spatial coordinate system and performs Kalman state smoothing on the attitude sequence to ensure that the attitude states used in subsequent coordinate mapping correspond to the original timestamps. The fixed relative position from the airframe reference point to the antenna phase center is calibrated by the installation relationship (e.g., calibration value...). The three attitude angles are obtained in real time by the inertial measurement unit (e.g., the roll angle, pitch angle, and yaw angle at the reference time are all...). The first adjacent attitude times are respectively , and The second adjacent attitude times are respectively , and Each attitude angle is converted to radians before being substituted into the coordinate rotation relationship.

[0053] Based on the above inputs, rigid body geometric constraints are established according to the fixed relative position and the current attitude, and the relative position of the antenna phase center in the body coordinate system is mapped to a unified spatial coordinate system. The instantaneous geometric displacement of the antenna phase center relative to the reference attitude can be expressed as:

[0054] (1)

[0055] In the formula, The three-dimensional instantaneous geometric displacement of the antenna phase center relative to the reference attitude, in units of ; To the current roll angle Pitch angle and yaw angle The rigid body rotation matrix formed is dimensionless; , and Use radians in calculations; This is a fixed relative position vector from the reference point of the aircraft to the phase center of the antenna, in units of... ; , and The reference attitude angle is obtained based on the aforementioned example value at the first adjacent attitude time. Approximately The second adjacent attitude time obtained Approximately .

[0056] Subsequently, the synchronization ranging time is extracted from the ranging observation data of the wireless receiver and a unified time reference axis is formed. Whether adjacent attitude samples can be used for synchronization ranging time interpolation is determined by the nominal sampling period of the inertial measurement unit and the wireless receiver, timestamp synchronization jitter, and timestamp statistics during the initialization phase, which together determine the attitude-ranging time tolerance (e.g., taking...). Displacement state interpolation is only performed when the synchronous ranging moment can be sandwiched between two valid attitude samples, and the time difference between the samples on both sides and the synchronous ranging moment is within the time tolerance; otherwise, the current synchronous ranging moment is marked as invalid and will not proceed to the subsequent abnormal reflection judgment.

[0057] For synchronous ranging moments that meet the time tolerance, sequence interpolation is performed based on adjacent smooth displacement states. The interpolation relationship can be expressed as:

[0058] (2)

[0059] In the formula, and For adjacent valid attitude sample times, the unit is 1. ; The synchronous ranging time between the two is given in units of 1 / 2. ; and The adjacent attitude displacements formed by formula (1) are in units of ; This represents the interpolated displacement state corresponding to the synchronous ranging moment, in units of... For example, take , , The interpolation ratio is Substituting the two adjacent displacement states mentioned above, we obtain .

[0060] Then, the interpolated displacement states are arranged along a unified time reference axis to form discrete displacement state quantities, which are then transformed into a spatial coordinate system consistent with the water surface geometry model and wireless ranging observations. After coordinate unification, each displacement state is associated point-by-point with wireless ranging observations at the same synchronous ranging time, forming a second attitude displacement sequence. Thus, the sampling frequency of the aforementioned attitude data and the sampling frequency of the wireless ranging observations may differ, but each second attitude displacement sample entering subsequent processing corresponds one-to-one with a valid synchronous ranging time.

[0061] The three attitude angles are aligned and mapped according to a unified spatial coordinate system; when the relative position from the body reference point to the antenna phase center is not effectively calibrated, the attitude field is missing, or the coordinate transformation relationship is invalid, the second attitude displacement sample is not generated at the corresponding synchronous ranging moment.

[0062] After the second attitude displacement sequence is formed, the original ranging observations that enter the same synchronous ranging time are processed.

[0063] In this second embodiment, for example, step S200 may further include the following process:

[0064] The wireless receiver generates the original ranging signal at the synchronous ranging moment according to its existing ranging timing sequence, and retains the timestamp of the original ranging observation consistent with the unified time reference axis. The data processing end reads the propagation time information in the original ranging signal at the same moment, performs Kalman noise smoothing on the propagation time sequence, suppresses high-frequency measurement jitter that does not change the physical meaning of propagation time, and then generates the measured propagation delay change relative to adjacent effective ranging observations or the current ranging reference.

[0065] For example, following the aforementioned At the synchronous ranging moment, when a valid second attitude displacement sample has been formed at that moment, the method data processing end reads the data from the wireless receiver at the same time. The resulting original ranging observations and propagation time information; if the current propagation time increases relative to the ranging reference... Then As the measured change in propagation delay at that moment, and maintained The timestamp remains unchanged.

[0066] The second attitude displacement sequence and the original ranging observations are established in a one-to-one correspondence according to the synchronous ranging time. If the original ranging observations are missing, timestamps conflict, propagation time cannot be obtained, or the ranging observations do not correspond to the second attitude displacement samples, the current synchronous ranging time is directly set to invalid, and no new original ranging observations are generated through interpolation. The resulting measured propagation delay changes are used for comparison in subsequent reference delay intervals, and continue to be used as the total observation for joint state decomposition after the abnormal reflection path is confirmed.

[0067] It should be noted that, in this embodiment, the measured propagation delay and the total measured propagation delay referred to thereafter both refer to the measured propagation delay change formed relative to the adjacent effective ranging observation or the current ranging reference in the manner described above. The total amount indicates that the measured propagation delay change is used as the total observation of the joint state decomposition, without changing its physical quantity definition.

[0068] After obtaining the measured propagation delay change at the same moment, the process enters the construction of the geometric reference delay interval for water surface reflection.

[0069] In this second embodiment, as Figure 3 As shown, step S300 may further include:

[0070] S300. Construct a first reflected wave vector based on the coordinates of the antenna center point and the geometric model of the water surface. Involve the second attitude displacement sequence with the first reflected wave vector to obtain the propagation path difference set, and convert its upper and lower bounds into a reference delay interval according to the propagation speed. Compare the reference delay interval with the measured propagation delay change at the same ranging time to determine the abnormal reflection path.

[0071] S310. Determine the water surface reflection point and the first reflected wave vector based on the coordinates of the antenna center point and the water surface geometry model.

[0072] S320. Filter the second attitude displacement sequence to obtain a displacement sequence representing the attitude change;

[0073] S330. Perform an inner product operation between the displacement sequence and the first reflected wave vector to form the wireless propagation path difference set using the resulting distance change.

[0074] S340. Extract the maximum and minimum values ​​of the wireless propagation path difference set to form the upper and lower bounds of the path difference;

[0075] S350. Convert the path difference upper and lower bounds into a propagation time difference interval according to the wireless signal propagation speed.

[0076] S360. When the propagation time difference interval meets the preset interval length condition, it is determined as the reference delay interval, and the measured signal whose delay deviates from the reference delay interval is determined as a multipath signal candidate.

[0077] S370. Compare the measured propagation delay at the same ranging time with the reference delay interval point by point to obtain the deviation.

[0078] S380. Determine the threshold matching degree based on the deviation amount and the width of the reference delay interval;

[0079] S390. When the threshold matching degree does not meet the preset matching standard, extract the reflection propagation residual that cannot be explained by the reference delay interval.

[0080] For example, steps S310 to S390 may further include the following processes:

[0081] In this embodiment, the antenna center point mentioned in steps S300 and S310 refers to the antenna phase center point used for wireless ranging observation, which is the same geometric object as the antenna phase center in step S100. First, the method data processing end obtains the coordinates of the antenna phase center point at the current synchronous ranging moment and a water surface geometry model that can characterize the current local water surface. The water surface geometry model is determined by the water surface geometry reference initialized on-site or from the previous valid verification cycle (for example, forming a first reflected wave unit vector pointing downwards towards the water surface in the current local region). (Dimensionless). The water surface reflection point serves as the geometric reference point for constructing the first reflected wave vector and is used to determine the propagation path change caused by the attitude displacement along the current water surface reflection reference direction.

[0082] Specifically, the antenna center point is projected onto the water surface geometry using the normal of the current local water surface geometry. The resulting projection point is used as a geometric reference reflection point, and the direction from the antenna center point to the geometric reference reflection point is normalized to obtain the first reflected wave unit vector. The geometric reference reflection point is only used to establish the attitude-interpretable water surface reflection geometric boundary and is not used as the sole reflection point of the actual multipath propagation path.

[0083] After applying Kalman state filtering to the second attitude displacement sequence, the synchronous ranging time marker for each displacement state is preserved. Then, the current displacement state is multiplied along the first reflected wave vector. The signed propagation path change corresponding to the synchronous ranging time can be represented as:

[0084] (3)

[0085] In the formula, For synchronous ranging time The signified propagation path change, in units of ; The second attitude displacement state obtained by formula (2) is expressed in units of . ; Let be the unit vector of the first reflected wave, defined by the current water surface geometry, and be dimensionless. Continuing from the previous... and ,get .

[0086] Subsequently, a set of wireless propagation path differences is formed within the valid processing sequence containing the current synchronous ranging time, and the maximum and minimum values ​​are extracted as the upper and lower bounds of the path differences. The wireless signal propagation speed adopts the propagation speed parameter used by the current ranging system under air propagation conditions, which is derived from the electromagnetic wave propagation physical constant and the propagation time conversion relationship of the ranging device (e.g., taking...). The path difference upper and lower bounds, converted to reference delay intervals, can be expressed as:

[0087] (4)

[0088] In the formula, and These are the minimum and maximum values ​​of the set of path differences within the effective processing sequence, respectively, in units of... ; The speed of wireless signal propagation, measured in units of 1000 m / s. ; and These are the lower and upper bounds of the reference delay interval, respectively, in units of... For example, obtaining in the current processing sequence , After substituting, the reference delay interval is approximately .

[0089] Whether the propagation time difference interval is considered a valid reference delay interval is determined by the ranging time resolution and attitude geometric uncertainty of the wireless receiver, based on a preset interval length condition (e.g., setting the effective interval width to...). ~ The width of the aforementioned example interval is... The interval falls within the specified range and is therefore determined as the reference delay interval for the current processing sequence. If the path difference samples are insufficient, the water surface geometry cannot form an effective reflection direction, or the propagation time difference interval does not meet the specified interval length condition, then subsequent effective synchronization samples will continue to be accumulated, and abnormal reflection paths will not be determined based on them.

[0090] Next, the measured propagation delay at the same synchronous ranging moment is compared point by point with the reference delay interval. This follows the procedure in step S200. The measured propagation delay change is higher than the upper bound of the reference delay interval. The deviation exceeding the upper bound is approximately Then, the threshold matching degree is determined by the ratio of the deviation to the width of the reference delay interval, resulting in approximately... The preset matching criterion is determined by the normalized super-boundary distribution of historically valid and anomaly-free synchronous ranging times of the same ranging link, or by the ranging uncertainty calibration of the wireless receiver (e.g., set to no greater than). Since the current threshold matching degree is approximately Since the preset matching criteria are not met, the portion exceeding the reference delay interval is extracted as the reflection propagation residual.

[0091] Then, the reflection propagation residuals are input into the isolated forest model. The fitted samples of the isolated forest model are taken from historical valid residuals of the same ranging link or continuous valid residuals during the task initialization phase. Its input consists only of reflection propagation residuals that have already been filtered through the reference delay interval, and its output is the residual anomaly score for the corresponding propagation path. A preset score line is calibrated based on the anomaly score distribution of the aforementioned valid residuals (e.g., taking...). (dimensionless). In the current example, if the residual abnormality score of the isolated forest output is... If the abnormal score exceeds the preset score line, the corresponding propagation path will be identified as an abnormal reflection path.

[0092] After confirming the abnormal reflection path, the measured propagation delay change is still used as the total input for subsequent joint state decomposition, and the reflection propagation residual is not directly equated to the final multipath error correction amount.

[0093] In this second embodiment, as Figure 4 As shown, step S400 may further include:

[0094] S400. If an abnormal reflection path exists, take the measured propagation delay change as the total constraint, and jointly decompose the geometric propagation change and additional propagation change that can be explained by the attitude displacement to obtain the attitude displacement influence amount and the multipath propagation influence amount; extract the remaining reflection propagation components according to the physical boundary of water surface reflection, and extract its propagation vector and the attitude motion vector corresponding to the attitude displacement influence amount at the same ranging time. Use the spatial angle between the two to determine the correction direction and construct the multipath error correction amount.

[0095] S410. Input the reflection propagation residual into the isolated forest model to obtain the residual abnormality score of the corresponding propagation path, and determine the corresponding propagation path as an abnormal reflection path when the residual abnormality score exceeds the preset score line.

[0096] S420. Using the measured propagation delay change as the total constraint, separate the geometric propagation change and the additional propagation change that can be explained by the attitude displacement;

[0097] S430. Perform joint state estimation on the geometric propagation change and the additional propagation change to obtain the attitude displacement influence and the multipath propagation influence.

[0098] S440. Based on the physical boundary of water surface reflection, the multipath propagation influence quantity is screened to obtain the remaining reflection propagation component.

[0099] S450. Using a random forest model, predict the multipath fading intensity based on the signal strength of the corrected direction and the original ranging signal, and determine the offset amplitude based on the multipath fading intensity.

[0100] S460, The correction direction and the offset magnitude are combined to form the multipath error correction amount.

[0101] For example, steps S410 to S460 may further include the following processes:

[0102] Once the anomalous reflection path is established, the measured propagation delay change at the same synchronous ranging moment is taken as the total observation. The attitude-interpretable propagation change formed by the second attitude displacement sequence along the first reflected wave vector is taken as the attitude state prior. Joint state estimation is then performed on the attitude displacement-interpretable geometric propagation change and the additional propagation change. The joint state estimation can be achieved using a Kalman-type joint estimation, which updates the attitude displacement influence and multipath propagation influence simultaneously at each synchronous ranging moment, ensuring that both maintain a closed-loop relationship with the measured total propagation delay.

[0103] During the joint state estimation process, the attitude state is constrained by the attitude geometry prediction formed by the attitude displacement along the first reflected wave vector, the multipath state is constrained by the current water surface reflection physical boundary, and the state combination that satisfies the total closure relationship is determined by combining the multipath state continuity of adjacent effective synchronous ranging times; state combinations that do not satisfy the total closure relationship or the water surface reflection physical boundary constraint are not considered as effective joint state results.

[0104] This total relationship can be expressed as:

[0105] (5)

[0106] In the formula, For synchronous ranging time The measured total propagation delay, in units of ; The attitude displacement influence obtained from the joint state estimation, in units of ; This represents the multipath propagation impact obtained from the joint state estimation, in units of... ; For closed residuals that cannot be explained by the two types of states, the unit is The allowable range of the closure residual is jointly determined by the propagation delay observation noise and the attitude geometry prediction uncertainty (e.g., taking no more than...). For example, the aforementioned The measured total propagation delay is decomposed into Influence of attitude displacement Multipath propagation impact and If the closed residual is within the allowable range, the current joint state result can continue to be used for water surface reflection physical boundary screening.

[0107] Based on this, the multipath propagation impact is screened according to the relationship between the abnormal reflection path and the reflection direction formed by the current water surface geometry. The screening process uses the compatibility of the current water surface reflection direction, the temporal correspondence of the abnormal reflection path, the interpretable range of the measured total propagation delay, and the continuity of the multipath state at adjacent effective synchronous ranging times as common conditions. The portion of the multipath propagation impact that simultaneously meets these conditions is retained as the remaining reflection propagation component; the portion that does not meet the above physical relationships is excluded as an unusable residual, and no other compensation path is established.

[0108] Following the previous example, after performing the above water surface reflection physical boundary screening on the 0.130 ns multipath propagation influence, if the retained compatible part is 0.100 ns, then this 0.100 ns is determined as the remaining reflection propagation component.

[0109] Subsequently, at the same synchronous ranging moment, a propagation vector is formed based on the water surface reflection direction of the remaining reflection propagation component, and an attitude motion vector is formed by the spatial displacement state of the phase centers of adjacent effective antennas that maintain the same synchronous ranging moment as the attitude displacement influence. In formula (5) The attitude displacement influence in the time domain is not directly used as a spatial vector; the attitude motion vector is the direction of motion characterized by the difference in spatial displacement states of the phase centers of adjacent effective antennas. Following the examples in steps S100 and S300, the first reflected wave unit vector is... The direction of adjacent attitude displacement changes is formed by the difference between the two aforementioned effective attitude states; the spatial angle calculated from the two is approximately... If the angle between the two vectors is acute, it is identified as positive multipath interference, and the positive direction is determined as the current multipath error correction direction. If the angle between the two vectors is obtuse, it is identified as negative multipath interference. If neither vector can form a valid direction, no new correction direction is generated at the current synchronous ranging moment.

[0110] After the correction direction is determined, the random forest model is then used to determine the offset magnitude. The training samples for the random forest model are derived from valid synchronous ranging samples that have passed consistency verification in the history of the same ranging link. The input is fixed as the correction direction determined by the spatial angle and the signal strength of the original ranging signal at the same synchronous ranging moment. The output is the multipath fading intensity represented by a normalized amplitude coefficient. The input features of the training samples are the correction direction and signal strength corresponding to the valid synchronous ranging moment. The supervision target is obtained by normalizing the residual reflection propagation influence at that moment with the final valid correction magnitude that has passed consistency verification, and is used to form the normalized amplitude coefficient. For example, the signal strength measured by the wireless receiver at the current moment is... Under the condition of positive correction, the normalized amplitude coefficient of the random forest output characterizing the intensity of multipath fading is... Dimensionless; combined The remaining reflection propagation component and The distance domain offset amplitude is approximately .

[0111] Finally, the positive and negative correction directions determined by the spatial angle are combined with the offset amplitude to form the multipath error correction amount, while retaining the current synchronous ranging time identifier. The random forest model only generates the amplitude and does not change the correction direction already determined by the spatial angle; the isolated forest model only confirms anomalous reflection paths and does not participate in the total decomposition of attitude displacement influence and multipath propagation influence. When the joint state estimation does not satisfy the closure relation, the physical boundary of the water surface reflection cannot form compatible components, the spatial angle cannot determine the direction, or the random forest input is invalid, no new multipath error correction amount is generated at the current synchronous ranging time.

[0112] After generating the multipath error correction amount corresponding to each point at the synchronous ranging time, the synchronous compensation and post-compensation verification of the original ranging observations are processed.

[0113] In this second embodiment, as Figure 5 As shown, step S500 may further include:

[0114] S500: The target ranging observation value is obtained by synchronously compensating the original ranging signal with the multipath error correction amount. The interpretable propagation range of the attitude is re-determined and the consistency is verified based on the target ranging observation value and the UAV attitude data. The water level monitoring point positioning verification is completed with the verified target ranging observation value to obtain the position coordinates.

[0115] S510. The multipath error correction amount is superimposed on the original ranging observation value at the corresponding synchronous ranging time to obtain the target ranging observation value.

[0116] S520. Based on the target ranging observation value and the UAV attitude data, redetermine the attitude interpretable propagation range and perform a consistency check on the target ranging observation value.

[0117] S530. When the verification result does not meet the preset consistency condition, update the multipath error correction amount and re-compensate until the preset consistency condition is met, and calculate the position coordinates based on the verified target ranging observation value.

[0118] For example, steps S510 to S530 may further include the following processes:

[0119] First, the multipath error correction is superimposed in reverse direction with the original ranging observation at the same synchronous ranging moment, while keeping the timestamp unchanged. Continuing with the previous example, if the original ranging observation at the current synchronous ranging moment is... The multipath error correction is positive. Then, the target ranging observation value is obtained by inverse superposition. .

[0120] Then, the target ranging observation is used to replace the original ranging observation at the current moment. The same UAV attitude data and the current water surface geometry are used to redetermine the attitude interpretable propagation range. The propagation change corresponding to the target ranging observation is then checked for consistency with the redetermined attitude interpretable propagation range. The preset consistency conditions are jointly determined by the uncertainty of the compensated target ranging observation, the ranging time resolution of the wireless receiver, and the uncertainty of the re-established attitude geometric boundary (for example, the compensated over-boundary deviation is required to be no greater than...). If the current target ranging observation corresponds to the out-of-bounds deviation as follows: If so, then the consistency condition is satisfied.

[0121] When the verification result does not meet the preset consistency conditions, the multipath error correction amount is updated according to the current verification deviation. During the update process, the correction direction determined by the spatial angle at the synchronous ranging time remains unchanged. The current verification deviation is used as the state update input for the offset amplitude. Only the offset amplitude is updated and recombined to form the multipath error correction amount at the current time. Subsequently, the reverse superposition, attitude interpretable propagation range, and consistency verification are re-executed from the original ranging observations at the same synchronous ranging time. The upper limit of the number of recompensation is determined by the processing cycle of a single survey task and the allowable computation delay (e.g., set to 3 times). If the consistency conditions are still not met after reaching the upper limit of the number of times, the current synchronous ranging time is marked as invalid, and the target ranging observation input position that has not passed the verification is not calculated.

[0122] For target ranging observations that meet the consistency criteria, they are used as valid wireless ranging inputs to the existing 3D positioning solution process. Based on the spatial geometric relationships between valid wireless ranging observations within the same survey mission, the position of the water level monitoring point is verified, and the position coordinates are output. Therefore, the final position coordinates use target ranging observations that have undergone attitude-interpretable propagation range verification, rather than the unverified raw multipath observations.

[0123] Accordingly, this invention enables the propagation changes that can be explained by attitude displacement and the additional propagation changes generated by water surface reflection to be continuously carried out in the same wireless ranging chain through synchronous attitude displacement formation, propagation path difference projection, reference delay interval construction, simultaneous abnormal reflection confirmation, joint decomposition of measured propagation delay total constraint, water surface reflection physical boundary screening, spatial angle orientation, random forest amplitude determination, original ranging reverse compensation, and post-compensation re-boundary verification, so that the propagation changes that can be explained by attitude displacement and the additional propagation changes generated by water surface reflection are continuously carried out in the same wireless ranging chain, forming a complete implementation process from the original attitude and ranging observation to the final position coordinates.

[0124] The above description is merely a preferred embodiment of the technical solution of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A wireless positioning verification method for unmanned aerial vehicle (UAV) inspection of water level monitoring points in water conservancy projects, characterized in that, include: The real-time attitude data of the UAV's inertial measurement unit is acquired, the instantaneous geometric distance change of the antenna phase center relative to the reference point of the aircraft is calculated, and the second attitude displacement sequence is obtained by smooth interpolation according to the synchronous ranging time. At the synchronous ranging moment, the original ranging signal is acquired and the measured propagation delay change is extracted; A first reflected wave vector is constructed based on the coordinates of the antenna center point and the geometric model of the water surface. The propagation path difference set is obtained by multiplying the second attitude displacement sequence with the first reflected wave vector, and its upper and lower bounds are converted into a reference delay interval according to the propagation speed. The reference delay interval is compared with the measured propagation delay change at the same ranging time to determine the abnormal reflection path. If an abnormal reflection path exists, the measured propagation delay change is used as the total constraint. The geometric propagation change and additional propagation change that can be explained by the attitude displacement are jointly decomposed to obtain the attitude displacement influence amount and the multipath propagation influence amount. The remaining reflection propagation components are extracted according to the physical boundary of water surface reflection. At the same ranging time, the propagation vector and the attitude motion vector corresponding to the attitude displacement influence amount are extracted. The correction direction is determined by the spatial angle between the two and the multipath error correction amount is constructed. The target ranging observation value is obtained by synchronously compensating the original ranging signal with the multipath error correction amount. The interpretable propagation range of the attitude is re-determined and the consistency is verified based on the target ranging observation value and the UAV attitude data. The water level monitoring point positioning verification is completed with the verified target ranging observation value to obtain the position coordinates.

2. The wireless positioning verification method according to claim 1, characterized in that, The calculation of the instantaneous geometric distance change of the antenna phase center relative to the body reference point includes: Extract yaw angle, pitch angle and roll angle from the real-time attitude data; The relative position from the aircraft reference point to the antenna phase center is obtained. Geometric constraints are constructed based on the relative position and the yaw angle, pitch angle and roll angle. Instantaneous distances in the three coordinate axis directions are obtained through coordinate mapping. The instantaneous geometric distance change is obtained by correcting the initial displacement value at the time of synchronous ranging based on the instantaneous distance measurement.

3. The wireless positioning verification method according to claim 1, characterized in that, The smooth interpolation based on the synchronous ranging time to obtain the second attitude displacement sequence includes: Extract the synchronization ranging time from the ranging observation data of the wireless receiver and set it as a unified time reference axis; The instantaneous geometric distance change corresponding to the unified time reference axis is smoothed to obtain smoothed displacement components; The adjacent attitude sample points corresponding to the smooth displacement component are obtained. When the timestamp of the adjacent attitude sample point falls within the preset time interval of the unified time reference axis, sequence interpolation is performed according to the time relationship between the adjacent attitude sample points and the synchronous ranging time to obtain the second attitude displacement sequence.

4. The wireless positioning verification method according to claim 3, characterized in that, Obtaining the second attitude displacement sequence includes: The displacement states after sequence interpolation are arranged according to the unified time reference axis to form discrete displacement state quantities; The discrete displacement state quantity is transformed into a spatial coordinate system to obtain the spatial displacement state of the antenna phase center; The spatial displacement state of the antenna phase center is matched point by point with the wireless ranging observation data to form the second attitude displacement sequence.

5. The wireless positioning verification method according to claim 1, characterized in that, The process of constructing the first reflected wave vector and obtaining the set of wireless propagation path differences includes: The water surface reflection point and the first reflected wave vector are determined based on the coordinates of the antenna center point and the water surface geometry model. The second attitude displacement sequence is filtered to obtain a displacement sequence representing the attitude change; The displacement sequence is multiplied by the first reflected wave vector to form the wireless propagation path difference set.

6. The wireless positioning verification method according to claim 5, characterized in that, It also includes converting the upper and lower bounds of the set of wireless propagation path differences into reference delay intervals, including: Extract the maximum and minimum values ​​of the wireless propagation path difference set to form the upper and lower bounds of the path difference; The path difference upper and lower bounds are converted into propagation time difference intervals according to the wireless signal propagation speed. When the propagation time difference interval meets the preset interval length condition, it is determined as the reference delay interval, and the measured signal whose delay deviates from the reference delay interval is determined as a multipath signal candidate.

7. The wireless positioning verification method according to claim 1, characterized in that, The step of comparing the consistency of the reference delay interval with the measured propagation delay change at the same ranging time includes: The deviation is obtained by comparing the measured propagation delay at the same ranging time with the reference delay interval point by point. The threshold matching degree is determined based on the deviation amount and the width of the reference delay interval; When the threshold matching degree does not meet the preset matching standard, the reflection propagation residual that cannot be explained by the reference delay interval is extracted.

8. The wireless positioning verification method according to claim 7, characterized in that, The process of determining abnormal reflection paths and extracting remaining reflection propagation components includes: The reflection propagation residuals are input into the isolated forest model to obtain the residual abnormality scores of the corresponding propagation paths. When the residual abnormality scores exceed a preset score line, the corresponding propagation path is identified as an abnormal reflection path. Using the measured propagation delay change as the total constraint, the geometric propagation change and the additional propagation change that can be explained by attitude displacement are separated; By performing joint state estimation on the geometric propagation change and the additional propagation change, the influence of attitude displacement and the influence of multipath propagation are obtained. The remaining reflection propagation components are obtained by filtering the multipath propagation influence based on the physical boundary of water surface reflection.

9. The wireless positioning verification method according to claim 1, characterized in that, The constructed multipath error correction amount includes: A random forest model is used to predict the multipath fading intensity based on the signal strength of the corrected direction and the original ranging signal, and the offset amplitude is determined based on the multipath fading intensity. The correction direction and the offset magnitude are combined to form the multipath error correction amount.

10. The wireless positioning verification method according to claim 1, characterized in that, The synchronous compensation of the original ranging signal and the consistency verification include: The multipath error correction is superimposed on the original ranging observation value at the corresponding synchronous ranging time to obtain the target ranging observation value; The attitude interpretable propagation range is re-determined based on the target ranging observations and UAV attitude data, and the consistency of the target ranging observations is verified. When the verification result does not meet the preset consistency condition, the multipath error correction amount is updated and recompensated until the preset consistency condition is met, and the position coordinates are calculated based on the verified target ranging observation value.

Images