A method and system for analyzing UAV flight data

Abstract

This invention discloses a method and system for analyzing UAV flight data, relating to the field of UAV flight data processing and intelligent analysis technology. The method includes the following steps: collecting continuous positioning change trajectories in a high-reflectivity urban environment; simultaneously organizing signal arrival time fluctuation records to construct an original position sequence containing detour traces; and splitting the signal arrival time fluctuation records in the original position sequence to expand the segments with elongated propagation times into propagation offset change trajectories that maintain the position progression order. This invention identifies elongated propagation time segments and symmetrically flipped sections to determine mirror change segments and position offset change trends, thereby extracting the time range of reflection impact. It also implements time scale misalignment processing at the path adjustment entry point, slowing down the position update rhythm of abnormal segments and reducing the involvement of mirror trajectories in path planning, thus improving the stability of trajectory reconstruction and flight safety.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) flight data processing and intelligent analysis technology, specifically to a UAV flight data analysis method and system. Background Technology

[0002] Unmanned aerial vehicle (UAV) flight data analysis refers to the systematic organization, calculation, and interpretation of various operational data generated by UAVs during flight, combined with big data processing technologies for centralized analysis and in-depth mining. This data typically includes position coordinates, flight altitude, speed changes, attitude angles, motor speeds, battery voltage and current, control command response, and environmental perception information. By performing time synchronization processing, trend calculation, anomaly identification, trajectory reconstruction, or performance evaluation on this data, and relying on big data processing to uniformly aggregate, correlate, and continuously calculate multi-source flight data, it is possible to reconstruct the UAV's flight status at a specific mission stage, assess flight stability, control precision, and energy utilization, identify potential risks or malfunctions, and provide data support for flight optimization, maintenance decisions, or algorithm improvements.

[0003] The existing technology has the following shortcomings: In existing technologies, drones typically rely on satellite positioning data for trajectory reconstruction and path planning. When drones are in densely populated urban environments with high reflectivity, satellite signals are reflected multiple times between building facades, easily leading to extended propagation paths. This results in spatial offsets in the positioning results, and may even produce mirror trajectories symmetrically distributed with respect to the true location. Existing data analysis systems often directly incorporate this positioning result into the trajectory reconstruction process when processing flight logs, lacking a mechanism to identify reflection anomalies. This can easily lead to false trajectories being treated as true flight paths in subsequent planning calculations. During automatic path correction, the system regenerates a corrected flight path based on the offset position, which can easily result in spatial judgment errors, causing the drone to fly towards building facades, posing a serious safety hazard.

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for analyzing unmanned aerial vehicle (UAV) flight data to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for analyzing UAV flight data, comprising the following steps: Collect continuous location change trajectories in urban high-reflection environments, simultaneously organize signal arrival time fluctuation records, and construct an original location sequence containing detour traces; The signal arrival time fluctuation records in the original position sequence are split and processed to expand the segments with elongated propagation time to form a propagation offset change trajectory that maintains the position advancement order. The positions along the propagation offset change trajectory are compared in direction, and the segments where the position and direction are symmetrically flipped are extracted to form mirror change segments. The position and offset change trend are then calculated within the mirror change segments. Based on the position, the offset change trend is calculated, and the signal arrival time fluctuation record in the original position sequence is traced back. The time range of continuous lengthening propagation time is extracted to form the reflection influence time range, and the reflection influence time range is determined as the path adjustment entry point. Around the path adjustment entry point, time scale misalignment is implemented in the subsequent track reconstruction process. Within the reflection influence time range, the position update rhythm is slowed down, while outside the reflection influence time range, the position update order is maintained. By alternating the time rhythm before and after, the participation of the mirror change segment in path planning is weakened.

[0007] Preferably, the steps for constructing the original location sequence are as follows: The positioning results are recorded around the flight time axis to form a continuous positioning change trajectory. The changes in satellite signal arrival time are recorded synchronously to form a signal arrival time fluctuation record. Each positioning result is established to correspond to the signal arrival time value at the corresponding time point and arranged in a unified time scale order. The signal arrival time fluctuation records are sorted out segment by segment, the time segments in which the signal arrival time is lengthened are marked, the positioning results within the corresponding time segments are retained, and the propagation time change indicators are added to the original time scale. The continuous positioning change trajectory is expanded and processed around the time sequence. The positioning results corresponding to the time segments where the signal arrival time is lengthened are expanded separately to form continuous subsequences, while maintaining the original position advancement sequence and the time advancement sequence. By embedding continuous subsequences into the time scale system corresponding to the continuous positioning change trajectory, the original position sequence of the positioning result and signal arrival time change information corresponding to each time node is formed.

[0008] Preferably, the steps for forming the propagation offset change trajectory while maintaining the position advancement order are as follows: The signal arrival time fluctuation record is continuously read along the continuous time scale corresponding to the original position sequence. Time segments in which the propagation time is continuously lengthened are marked, and the position advancement order in the original position sequence is maintained. The marked segments with extended propagation time are split and processed in chronological order. The positioning results corresponding to the segments with extended propagation time are extracted to form extended propagation time subsequences containing time markers, while keeping the time progression order consistent with the location progression order. The process is unfolded along the time progression direction of the elongated subsequence, forming a continuous trajectory representation while maintaining the correspondence between the position progression order and the time scale in the original position sequence. By interpolating the continuous trajectory representation with the time scale system in the original position sequence, a propagation offset change trajectory that maintains the position advancement order is formed.

[0009] Preferably, when marking the signal arrival time fluctuation record along the continuous time scale corresponding to the original position sequence, the propagation time is stretched into a continuous interval on the time axis. During the splitting process, the time progression order is kept consistent with the position progression order. During the unfolding process, the correspondence between the time marker and the spatial coordinate is preserved, forming a propagation offset change trajectory that combines time scale information and propagation time change characteristics.

[0010] Preferably, the steps for extracting the mirror-image change segment, determining its position, and calculating the offset change trend are as follows: The positioning results are read point by point along the time scale sequence of the propagation offset change trajectory. A continuous directional expression is constructed according to the spatial coordinate change relationship between adjacent time nodes to form a continuous directional change sequence, while maintaining the position advancement order in the propagation offset change trajectory. By comparing directions along the time progression sequence of continuous directional change, continuous segments in which the spatial displacement direction exhibits a symmetrical flipping state are identified, forming candidate flipping segments, and the time interval expression of the candidate flipping segments is preserved. By comparing the spatial progression order within the candidate flipped sections with the overall advancement direction of the propagation offset change trajectory, candidate flipped sections that exhibit a symmetrical distribution relationship are identified as mirror change sections, and the temporal and spatial correspondence of the mirror change sections in the propagation offset change trajectory is maintained. The spatial coordinate change status is read sequentially over time around the mirror change segment, and the evolution path of the position calculation offset generated within the mirror change segment is extracted to form the position calculation offset change trend.

[0011] Preferably, the spatial displacement direction inside the mirror change section is compared with the spatial advancement direction outside the mirror change section. The continuous time section in which the spatial displacement direction shows an opposite distribution relationship is defined as the mirror change section. The spatial coordinate change state is continuously extracted along the time advancement sequence of the mirror change section to clarify the time evolution relationship of the position calculation offset change trend.

[0012] Preferably, the steps for determining the time range of reflection influence and identifying the path adjustment entry point are as follows: Read the position offset change trend sequentially along the time scale corresponding to the mirror change segment, and map the time node in the mirror change segment to the corresponding time node in the original position sequence, so that the position offset change trend and the signal arrival time fluctuation record form a time correspondence. By back-reading the signal arrival time fluctuation records in the original position sequence around the time segment of the position offset change trend, continuous segments with continuously lengthening propagation time are marked to maintain the consistency of the time progression order; By combining the time ranges of the continuously lengthening propagation period and the mirror change period, a corresponding time range of reflection influence with continuous time interval expression is formed; The time range of reflection impact is marked in the original location sequence, and the starting time node of the reflection impact time range is determined as the path adjustment entry point.

[0013] Preferably, the continuous segments with continuously elongated propagation time and the mirror change segments form an overlapping interval on the time scale. The reflection influence time range covers the overlapping interval and maintains the boundary of the continuous interval along the time progression sequence. The path adjustment entry corresponds to the starting time node of the reflection influence time range.

[0014] Preferably, the steps for implementing time scale misalignment processing to reduce the involvement of mirrored change segments in path planning are as follows: The subsequent track reconstruction process is carried out along the time scale of the original position sequence around the entry point of the path adjustment, maintaining the consistency of the position advancement sequence, and segmenting the time range of reflection impact; Adjust the entry point along the path to the time segment corresponding to the time range of the reflection effect, implement time scale misalignment processing, redistribute the time interval, and slow down the position update pace while keeping the position advancement order unchanged; Maintain the original time scale spacing and position update order around the time segment outside the time range affected by reflection, so that the time scale forms an alternating state of original rhythm and slowed rhythm; By combining the adjusted time scale system with the subsequent track reconstruction process, the spatial location within the time range affected by reflections will participate in path planning at a slower pace, thus reducing the degree of participation of the mirror change segment in path planning.

[0015] A UAV flight data analysis system includes a raw sequence construction module, a propagation offset generation module, a mirror segment identification module, a reflection interval determination module, and a time modulation module. The original sequence construction module collects continuous location change trajectories in urban high-reflection environments, synchronously organizes signal arrival time fluctuation records, and constructs an original location sequence containing detour traces. The propagation offset generation module splits the signal arrival time fluctuation record in the original position sequence and expands the segments with elongated propagation time to form a propagation offset change trajectory that maintains the position advancement order. The mirror segment identification module compares the directions along the position progression sequence of the propagation offset change trajectory, extracts the segments where the position and direction are symmetrically flipped to form mirror change segments, and determines the position and calculates the offset change trend within the mirror change segments. The reflection interval determination module calculates the offset change trend based on the location and traces back the signal arrival time fluctuation record in the original location sequence. It extracts the time range of continuous propagation time to form the reflection influence time range and determines the reflection influence time range as the path adjustment entry point. The time modulation module implements time scale misalignment processing for the subsequent track reconstruction process around the path adjustment entry point. Within the reflection influence time range, it slows down the position update rhythm, and maintains the position update order outside the reflection influence time range. By alternating the time rhythm before and after, it weakens the participation of the mirror change segment in the path planning.

[0016] The technical effects and advantages provided by the present invention in the above technical solution are as follows: This invention synchronizes and organizes records of continuous positioning change trajectories and signal arrival time fluctuations, and constructs propagation offset change trajectories around segments with elongated propagation times. Furthermore, it combines the extraction of symmetrical flipped segments in position direction with the determination of position offset change trends to achieve directional identification of mirror trajectories in urban high-reflection environments. This enables the spatial offset source to form a corresponding relationship in the time and spatial dimensions, thereby separating the impact of reflection interference before track reconstruction, preventing false trajectories from directly participating in the path planning process, and improving the stability and continuity of flight path representation.

[0017] This invention determines the time range of reflection impact and uses it as the entry point for path adjustment. In the subsequent track reconstruction process, it implements time scale misalignment processing, slows down the position update rhythm within the reflection impact time range, and maintains the position update order outside the reflection impact time range. By alternating the time rhythm, it reduces the participation of the mirror change segment in the overall path planning, so that the track reconstruction process can maintain the continuity of the position advancement order while weakening the impact of reflection interference on the path generation direction, thereby improving the spatial rationality of the path planning results and flight safety. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0019] Figure 1 This is a flowchart of a method for analyzing UAV flight data according to the present invention.

[0020] Figure 2 This is a schematic diagram of a module of an unmanned aerial vehicle (UAV) flight data analysis system according to the present invention. Detailed Implementation

[0021] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0022] This invention provides, for example Figure 1 The method for analyzing UAV flight data shown includes the following steps: Collect continuous location change trajectories in urban high-reflection environments, simultaneously organize signal arrival time fluctuation records, and construct an original location sequence containing detour traces; To generate the original location sequence containing detour traces, the continuous location change trajectory and signal arrival time fluctuation records are synchronized and organized under complex urban high-reflection environment conditions. The specific implementation steps are as follows: During the UAV's flight mission, positioning results are continuously recorded around the flight timeline to form a continuous positioning change trajectory. Simultaneously, changes in satellite signal arrival time are recorded around the same timeline to form a signal arrival time fluctuation record. During the recording process, a one-to-one correspondence is established between each positioning result in the continuous positioning change trajectory and the corresponding signal arrival time value, and they are arranged according to a unified time scale order, so that the continuous positioning change trajectory and the signal arrival time fluctuation record are under the same time reference system. On this basis, the position advancement order in the continuous positioning change trajectory is maintained, and the spatial order is not rearranged, so that the time advancement direction is consistent with the spatial advancement direction. This provides a continuous time reference for subsequent processing based on the changes in propagation time. During the recording process, the spatial backtracking traces formed by the flight path circling between buildings are preserved, so that the continuous positioning change trajectory exhibits continuous change characteristics in both the time and spatial dimensions.

[0023] After synchronizing the continuous positioning trajectory with the signal arrival time fluctuation records, the changes in the signal arrival time fluctuation records on the time axis are organized segment by segment. Time segments in which the signal arrival time is lengthened are marked separately according to time order. At the same time, the positioning results within the corresponding time segments are retained during the marking process, so that the positioning results in the marked time segments and the positioning results in the unmarked time segments maintain the original position progression order in the same continuous positioning trajectory. During the organization process, the time interval relationship between the positioning results is not changed. Instead, propagation time change indicators are added to the original time scale, so that the continuous positioning trajectory maintains time continuity while superimposing propagation time change information. This forms an original data set containing propagation time change indicators, and provides a clear time positioning range for the next stage of splitting and processing around the propagation time lengthened segments.

[0024] Based on the original dataset containing propagation time variation markers, the continuous positioning change trajectory is expanded according to the time sequence. The positioning results corresponding to the time segments where the signal arrival time is lengthened are expanded separately to form continuous subsequences. During the expansion process, the original position progression order and time progression order are kept consistent, and the relationship between the positioning results is not changed. This makes the propagation time lengthened segments appear as a continuous distribution on the time axis, while retaining the path turning information reflected by the detour traces in the spatial dimension. Through the above expansion process, the propagation time lengthened segments form independently observable sections in the continuous positioning change trajectory. This provides a clear time interval expression for identifying the phenomenon of extended propagation paths caused by building facade reflections in urban high-reflection environments, and ensures that the expanded subsequences still adhere to the original time scale system.

[0025] After unfolding the extended propagation time segments, the resulting subsequences are re-embedded into the time scale system corresponding to the original continuous positioning trajectory, constructing an original position sequence containing detour traces. In this original position sequence, each time node corresponds to the positioning result and signal arrival time change information, maintaining a strict continuous progression order on the time axis. At the same time, the path offset traces caused by building reflection environment are retained in the spatial dimension, so that the original position sequence reflects both the actual flight trajectory changes of the UAV in the urban high-reflection environment and fully presents the position offset characteristics caused by the change in signal propagation time. This forms a comprehensive data sequence that combines temporal continuity, spatial progression order, and propagation time fluctuation indicators, providing a continuous time reference and spatial change basis for subsequent identification and processing of propagation offset change trajectory, mirror change segment, and reflection influence time range.

[0026] The signal arrival time fluctuation records in the original position sequence are split and processed to expand the segments with elongated propagation time to form a propagation offset change trajectory that maintains the position advancement order. After constructing the original position sequence containing the detour traces, in order to further extract propagation offset features from the perspective of time dimension and propagation time variation, segmentation and sequence expansion processing can be carried out around the signal arrival time fluctuation record to form a propagation offset change trajectory that maintains the position advancement order. The specific implementation steps are as follows: The signal arrival time fluctuation records are continuously read along the continuous time scale corresponding to the original position sequence. The signal arrival time values ​​of each time node are arranged and organized based on the time progression direction, so that the signal arrival time fluctuation records present a continuous distribution on the time axis. In this process, each positioning result in the original position sequence is synchronously compared with the signal arrival time fluctuation records of the corresponding time node, so that the time scale, positioning results and signal arrival time fluctuation records maintain a one-to-one correspondence. The changes in the signal arrival time values ​​are observed point by point around the time axis. Time segments in which the propagation time is continuously lengthened are marked, while keeping the position progression order in the original position sequence unchanged, so that the marked propagation time lengthened segments form a continuous segment expression on the time axis.

[0027] After continuously marking the segments with extended propagation time, these marked segments are split from the original location sequence according to time order. During the splitting process, the time progression order and location progression order in the original location sequence are kept consistent. No compression or skipping of the time scale is performed. Instead, the positioning results corresponding to the segments with extended propagation time are extracted separately in time order to form subsequences. During the extraction process, the time identifier of each positioning result in the original time scale is retained. This ensures that the segments with extended propagation time are both separated from the overall expression of the original location sequence and still maintain their association with the original time scale system. This forms subsequences with extended propagation time containing time identifiers, and provides basic data for subsequent processing based on the spatial progression order.

[0028] After obtaining the elongated propagation time subsequence, the positioning results in this subsequence are expanded along the time progression direction. This makes the elongated propagation time segments, which were originally embedded in the original position sequence, appear as a continuous expansion in the time dimension. During the expansion process, the order between the positioning results is kept completely consistent with the position progression order in the original position sequence. No time node correspondence is changed, nor is the numerical expression of the positioning results in the spatial coordinate system altered. Instead, while maintaining the original position progression order, the positioning results in the elongated propagation time segments form an independent and continuous trajectory expression, thus constructing the initial form of the propagation offset change trajectory. In this process, the propagation offset change trajectory inherits the time scale information from the original position sequence and retains the propagation time change characteristics reflected in the signal arrival time fluctuation record, so that the elongated propagation time segments simultaneously possess a continuous expression form in both the time and spatial dimensions.

[0029] After completing the expansion of the extended propagation time subsequence, the resulting continuous trajectory expression is re-aligned with the time scale system in the original position sequence. This allows the propagation offset change trajectory to form an independently observable trajectory segment expression within the overall time axis, while maintaining complete consistency with the original position sequence in the order of position progression. During this process, other time segments in the original position sequence are not altered; only the extended propagation time segments are expanded. This ensures that the propagation offset change trajectory clearly demonstrates the spatial offset evolution process caused by the extended signal propagation path on the positioning results. Furthermore, based on the continuous advancement of the time scale, a propagation offset change trajectory that maintains the order of position progression is formed. This provides a continuous and complete temporal and spatial reference expression for subsequent comparisons of position directions and extraction of mirror change segments.

[0030] The positions along the propagation offset change trajectory are compared in direction, and the segments where the position and direction are symmetrically flipped are extracted to form mirror change segments. The position and offset change trend are then calculated within the mirror change segments. After a propagation offset change trajectory that maintains the order of positional advancement has been formed, in order to identify the spatial mirror features caused by the extension of the propagation path in a high-reflectivity urban environment, directional comparison and segment extraction can be carried out around the spatial advancement state of the propagation offset change trajectory. Based on this, the positional offset change trend can be determined. The specific implementation steps are as follows: The positioning results are read point by point along the time scale sequence of the propagation offset change trajectory, and a continuous directional expression is constructed according to the spatial coordinate change relationship between adjacent time nodes. This ensures that the advancement relationship of each position in the propagation offset change trajectory is expressed as a spatial displacement vector based on time advancement. During the construction process, the original position advancement order of the propagation offset change trajectory is kept unchanged, the continuity of the time scale is not changed, and the spatial coordinate numerical expression is not changed. By recording the spatial displacement direction between adjacent time nodes segment by segment, the entire propagation offset change trajectory forms a continuous directional change sequence in the time dimension, providing a complete directional expression basis for subsequent directional comparison.

[0031] After obtaining the continuous directional change sequence, the directional change relationship between adjacent segments is compared and analyzed according to the time progression order. In the comparison process, the time scale is used as the main line, and the spatial displacement direction of the current time node is compared with the spatial displacement direction of the previous time node segment by segment to identify segments where the spatial displacement direction changes in reverse. In the identification process, the original time progression order and position progression order of the propagation offset change trajectory are kept consistent, and the spatial order is not rearranged. When the directional relationship between adjacent segments shows a symmetrical reversal state, the continuous segment is marked as a candidate reversal segment. In the marking process, the start time node and end time node of the segment are retained, so that the candidate reversal segment forms a clear interval expression on the time axis, laying the time positioning foundation for the formation of mirror change segments.

[0032] After marking the candidate flipped sections, further continuous observation is conducted around the spatial progression sequence within the candidate flipped sections. The spatial displacement directions of multiple consecutive time nodes within the candidate flipped sections are sequentially unfolded, so that the spatial progression trajectory within the flipped sections presents a complete unfolded state. Combined with the overall progression direction of the propagation offset change trajectory, a comprehensive comparison is made. When the spatial progression sequence within the candidate flipped section and the spatial progression sequence of the propagation offset change trajectory outside the flipped section show a symmetrical distribution relationship, the candidate flipped section is identified as a mirror change section. The temporal and spatial correspondence of the mirror change section in the propagation offset change trajectory remains unchanged, making the mirror change section an independent segment expression in the propagation offset change trajectory. At the same time, the position progression sequence is continuously expressed, thus forming a complete mirror change section identification result.

[0033] After identifying the mirror change segment, the spatial coordinate changes are continuously read according to the time progression sequence within the mirror change segment. The distribution of the direction and magnitude of spatial position changes at each time node within the mirror change segment on the time axis is observed. This is then compared with the spatial progression direction of the propagation offset trajectory outside the mirror change segment to extract the evolution path of the position calculation offset within the mirror change segment. This ensures that the spatial position changes within the mirror change segment form a continuous trend from the starting time node to the ending time node. During this process, the continuity of the time scale and the consistency of the position progression sequence are maintained, so that the position calculation offset change trend presents a continuous progression relationship in the time dimension and a trajectory pattern relatively distributed with respect to the original progression direction in the spatial dimension. This forms a mirror change segment with temporal positioning and spatial evolution characteristics in the propagation offset change trajectory, and clarifies the position calculation offset change trend within this mirror change segment, providing clear temporal and spatial basis for subsequent extraction of the time range of reflection influence.

[0034] Based on the position, the offset change trend is calculated, and the signal arrival time fluctuation record in the original position sequence is traced back. The time range of continuous lengthening propagation time is extracted to form the reflection influence time range, and the reflection influence time range is determined as the path adjustment entry point. After obtaining the mirror-image change segment and its corresponding positional offset change trend, in order to temporally correlate the spatial offset evolution process with signal propagation behavior, a backtracking process can be carried out around the positional offset change trend. This involves extracting the time range of continuously lengthening propagation time from the signal arrival time fluctuation records in the original position sequence, and based on this, forming the reflection influence time range and path adjustment entry point. The specific implementation steps are as follows: The position offset change trend is read sequentially along the time scale corresponding to the mirror change segment. The spatial offset evolution state of each time node within the mirror change segment is continuously organized according to the time progression direction, so that the position offset change trend forms a continuous distribution expression on the time axis, and the time start node and end node of the mirror change segment in the propagation offset change trajectory remain unchanged. During the organization process, the time nodes within the mirror change segment are mapped back to the corresponding time nodes in the original position sequence one by one, so that the position offset change trend and the signal arrival time fluctuation record in the original position sequence form a one-to-one time correspondence, and establish a clear time index foundation for subsequent backtracking and extraction of the signal arrival time fluctuation record.

[0035] After completing the time mapping, the signal arrival time fluctuation records in the original position sequence are read in reverse, following the evolution direction of the position offset change trend. Using the starting time node of the mirror change segment as the time reference, continuous readings are performed forward and backward along the time axis, so that the signal arrival time fluctuation records form a complete and continuous expression in the time dimension. During the reading process, the focus is on the signal arrival time change state corresponding to the time segment of the position offset change trend. When the signal arrival time shows a continuous lengthening of the propagation time within the continuous time nodes, the corresponding time nodes are continuously marked, and the time progression order is kept unchanged, so that the continuously lengthening propagation time segment forms a continuous interval expression in the original position sequence.

[0036] After obtaining the segment with continuously elongated propagation time, the time range of this segment is aligned with that of the mirror change segment. This allows the segments with continuously elongated propagation time and the trend of positional offset change to overlap on the time axis. The overlapping part is then continuously expanded along the time progression direction to form a complete interval on the time axis, while maintaining its time position in the original position sequence without shifting. During the organization process, the start and end time nodes of the continuously elongated propagation time are recorded as intervals, forming clear boundaries in the time dimension. These intervals are then used as the reflection influence time range, ensuring a correlation and correspondence between the reflection influence time range and the mirror change segment on the time axis.

[0037] After establishing the reflection impact time range, this time range is fixedly marked within the original position sequence, making it an independent time segment within the original position sequence. The starting time node of the reflection impact time range is determined as the path adjustment entry point, allowing subsequent track reconstruction to implement time scale adjustments starting from this entry point during time progression. In determining the path adjustment entry point, the continuity of the time scale in the original position sequence is maintained, and other time segments are not altered. This ensures that the reflection impact time range has a clear boundary on the time axis, and the path adjustment entry point serves as the starting point for subsequent track reconstruction, achieving a closed-loop expression of the temporal correlation between the position calculation offset change trend and the signal propagation time change.

[0038] Around the path adjustment entry point, time scale misalignment is implemented in the subsequent track reconstruction process. Within the reflection influence time range, the position update rhythm is slowed down, while outside the reflection influence time range, the position update order is maintained. By alternating the time rhythm before and after, the participation of the mirror change segment in the path planning is weakened. After establishing the timeframe for reflection effects and identifying the entry point for path adjustment, to reduce the involvement of mirror image change segments in subsequent track reconstruction, the time scale can be reconstructed around the path adjustment entry point. This allows the time progression rhythm to exhibit differentiated expressions in different time segments, thereby achieving time scale misalignment handling and position update rhythm adjustment. The specific implementation steps are as follows: Starting with the time node corresponding to the path adjustment entry, the time progression expression of the subsequent track reconstruction process is unfolded along the time scale direction of the original position sequence. During the unfolding process, the position progression order in the original position sequence is kept unchanged, so that the spatial coordinates corresponding to each time node are still arranged in the original order. At the same time, the reflection influence time range is segmented on the time axis, so that an independent time segment expression is formed between the path adjustment entry and the end node of the reflection influence time range, and this time segment becomes the effective range of time scale adjustment in the subsequent track reconstruction process.

[0039] A time scale misalignment process is implemented between the path adjustment entry point and the end node of the reflection impact time range. The time intervals within this time segment are redistributed in the time scale expression, so that the time advancement spacing between adjacent time nodes appears extended in the expression level. This slows down the position update rhythm without changing the position advancement order, and makes the spatial coordinate updates within the reflection impact time range appear as a sparser advancement expression on the time axis. During the implementation of the time scale misalignment process, each spatial coordinate still corresponds to its original time node. Only the time scale spacing is adjusted in the time expression system of subsequent track reconstruction, so that the trajectory expression within the reflection impact time range occupies a more dispersed time position in the overall time axis, thereby reducing the degree of clustering of mirror change segments in continuous time advancement.

[0040] After completing the time scale misalignment processing within the reflection influence time range, the original time scale spacing and position update order remain unchanged for time segments outside the reflection influence time range. This ensures that the time segments before the path adjustment entry point and after the reflection influence time range termination node continue to advance according to the original time scale, forming an alternating distribution state on the overall time axis with the original rhythm in the first segment, the slowed rhythm in the middle segment, and the original rhythm in the last segment. By alternating the time rhythm of maintaining the original update order in the first and second segments and the slowed position update rhythm in the middle segment, the time scale presents differentiated expressions in different time segments, while maintaining a completely consistent spatial coordinate arrangement order. This weakens the concentrated influence of mirror change segments in the continuous path expression in the time dimension.

[0041] After the time scale misalignment is resolved, the adjusted time scale system is integrated with the subsequent track reconstruction process. This allows the subsequent track reconstruction to read spatial coordinates according to the adjusted time intervals as time progresses. This slows down the pace of spatial position updates within the reflection influence time range, while maintaining the original pace of spatial position updates outside the reflection influence time range. By alternating the time pace, the participation of mirror change segments is weakened, allowing these segments to exist in a dispersed form during the overall track planning process. Thus, while maintaining the position progression order and spatial coordinate continuity, the time scale misalignment process dynamically adjusts the subsequent track reconstruction process.

[0042] This invention synchronizes and organizes records of continuous positioning change trajectories and signal arrival time fluctuations, and constructs propagation offset change trajectories around segments with elongated propagation times. Furthermore, it combines the extraction of symmetrical flipped segments in position direction with the determination of position offset change trends to achieve directional identification of mirror trajectories in urban high-reflection environments. This enables the spatial offset source to form a corresponding relationship in the time and spatial dimensions, thereby separating the impact of reflection interference before track reconstruction, preventing false trajectories from directly participating in the path planning process, and improving the stability and continuity of flight path representation.

[0043] This invention determines the time range of reflection impact and uses it as the entry point for path adjustment. In the subsequent track reconstruction process, it implements time scale misalignment processing, slows down the position update rhythm within the reflection impact time range, and maintains the position update order outside the reflection impact time range. By alternating the time rhythm, it reduces the participation of the mirror change segment in the overall path planning, so that the track reconstruction process can maintain the continuity of the position advancement order while weakening the impact of reflection interference on the path generation direction, thereby improving the spatial rationality of the path planning results and flight safety.

[0044] This invention provides, for example Figure 2 The UAV flight data analysis system shown includes a raw sequence construction module, a propagation offset generation module, a mirror segment identification module, a reflection interval determination module, and a time modulation module. The original sequence construction module collects continuous location change trajectories in urban high-reflection environments, synchronously organizes signal arrival time fluctuation records, and constructs an original location sequence containing detour traces. The propagation offset generation module splits the signal arrival time fluctuation record in the original position sequence and expands the segments with elongated propagation time to form a propagation offset change trajectory that maintains the position advancement order. The mirror segment identification module compares the directions along the position progression sequence of the propagation offset change trajectory, extracts the segments where the position and direction are symmetrically flipped to form mirror change segments, and determines the position and calculates the offset change trend within the mirror change segments. The reflection interval determination module calculates the offset change trend based on the location and traces back the signal arrival time fluctuation record in the original location sequence. It extracts the time range of continuous propagation time to form the reflection influence time range and determines the reflection influence time range as the path adjustment entry point. The time modulation module implements time scale misalignment processing for the subsequent track reconstruction process around the path adjustment entry point. Within the reflection influence time range, it slows down the position update rhythm, and maintains the position update order outside the reflection influence time range. By alternating the time rhythm before and after, it weakens the participation of the mirror change segment in the path planning.

[0045] The present invention provides a method for analyzing UAV flight data, which is implemented by the aforementioned UAV flight data analysis system. For details of the specific method and process of the UAV flight data analysis system, please refer to the embodiment of the aforementioned UAV flight data analysis method, which will not be repeated here.

[0046] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A method for analyzing UAV flight data, characterized in that, Includes the following steps: Collect continuous location change trajectories in urban high-reflection environments, simultaneously organize signal arrival time fluctuation records, and construct an original location sequence containing detour traces; The signal arrival time fluctuation records in the original position sequence are split and processed to expand the segments with elongated propagation time to form a propagation offset change trajectory that maintains the position advancement order. The positions along the propagation offset change trajectory are compared in direction, and the segments where the position and direction are symmetrically flipped are extracted to form mirror change segments. The position and offset change trend are then calculated within the mirror change segments. Based on the position, the offset change trend is calculated, and the signal arrival time fluctuation record in the original position sequence is traced back. The time range of continuous lengthening propagation time is extracted to form the reflection influence time range, and the reflection influence time range is determined as the path adjustment entry point. Around the path adjustment entry point, time scale misalignment is implemented in the subsequent track reconstruction process. Within the reflection influence time range, the position update rhythm is slowed down, while outside the reflection influence time range, the position update order is maintained. By alternating the time rhythm before and after, the participation of the mirror change segment in path planning is weakened.

2. The method for analyzing UAV flight data according to claim 1, characterized in that, The steps for constructing the original location sequence are as follows: The positioning results are recorded around the flight time axis to form a continuous positioning change trajectory. The changes in satellite signal arrival time are recorded synchronously to form a signal arrival time fluctuation record. Each positioning result is established to correspond to the signal arrival time value at the corresponding time point and arranged in a unified time scale order. The signal arrival time fluctuation records are sorted out segment by segment, the time segments in which the signal arrival time is lengthened are marked, the positioning results within the corresponding time segments are retained, and the propagation time change indicators are added to the original time scale. The continuous positioning change trajectory is expanded and processed around the time sequence. The positioning results corresponding to the time segments where the signal arrival time is lengthened are expanded separately to form continuous subsequences, while maintaining the original position advancement sequence and the time advancement sequence. By embedding continuous subsequences into the time scale system corresponding to the continuous positioning change trajectory, the original position sequence of the positioning result and signal arrival time change information corresponding to each time node is formed.

3. The method for analyzing UAV flight data according to claim 2, characterized in that, The steps for forming the propagation offset change trajectory while maintaining the positional advancement order are as follows: The signal arrival time fluctuation record is continuously read along the continuous time scale corresponding to the original position sequence. Time segments in which the propagation time is continuously lengthened are marked, and the position advancement order in the original position sequence is maintained. The marked segments with extended propagation time are split and processed in chronological order. The positioning results corresponding to the segments with extended propagation time are extracted to form extended propagation time subsequences containing time markers, while keeping the time progression order consistent with the location progression order. The process is unfolded along the time progression direction of the elongated subsequence, forming a continuous trajectory representation while maintaining the correspondence between the position progression order and the time scale in the original position sequence. By interpolating the continuous trajectory representation with the time scale system in the original position sequence, a propagation offset change trajectory that maintains the position advancement order is formed.

4. The method for analyzing UAV flight data according to claim 3, characterized in that, When the signal arrival time fluctuation record is marked along the continuous time scale corresponding to the original position sequence, the propagation time is stretched into a continuous interval on the time axis. During the splitting process, the time progression order is kept consistent with the position progression order. During the unfolding process, the correspondence between the time marker and the spatial coordinate is preserved, forming a propagation offset change trajectory that combines time scale information and propagation time change characteristics.

5. The method for analyzing UAV flight data according to claim 3, characterized in that, The steps for extracting the mirror-image change segment, determining its location, and calculating the offset change trend are as follows: The positioning results are read point by point along the time scale sequence of the propagation offset change trajectory. A continuous directional expression is constructed according to the spatial coordinate change relationship between adjacent time nodes to form a continuous directional change sequence, while maintaining the position advancement order in the propagation offset change trajectory. By comparing directions along the time progression sequence of continuous directional change, continuous segments in which the spatial displacement direction exhibits a symmetrical flipping state are identified, forming candidate flipping segments, and the time interval expression of the candidate flipping segments is preserved. By comparing the spatial progression order within the candidate flipped sections with the overall advancement direction of the propagation offset change trajectory, candidate flipped sections that exhibit a symmetrical distribution relationship are identified as mirror change sections, and the temporal and spatial correspondence of the mirror change sections in the propagation offset change trajectory is maintained. The spatial coordinate change status is read sequentially over time around the mirror change segment, and the evolution path of the position calculation offset generated within the mirror change segment is extracted to form the position calculation offset change trend.

6. The method for analyzing UAV flight data according to claim 5, characterized in that, By comparing the spatial displacement direction inside the mirror change section with the spatial advancement direction outside the mirror change section, the continuous time segment in which the spatial displacement direction shows an inverse distribution relationship is defined as the mirror change section. The spatial coordinate change state is continuously extracted along the time advancement sequence of the mirror change section to clarify the temporal evolution relationship of the positional offset change trend.

7. The method for analyzing UAV flight data according to claim 5, characterized in that, The steps for determining the time range of reflection effects and identifying the path adjustment entry are as follows: Read the position offset change trend sequentially along the time scale corresponding to the mirror change segment, and map the time node in the mirror change segment to the corresponding time node in the original position sequence, so that the position offset change trend and the signal arrival time fluctuation record form a time correspondence. By back-reading the signal arrival time fluctuation records in the original position sequence around the time segment of the position offset change trend, continuous segments with continuously lengthening propagation time are marked to maintain the consistency of the time progression order; By combining the time ranges of the continuously lengthening propagation period and the mirror change period, a corresponding time range of reflection influence with continuous time interval expression is formed; The time range of reflection impact is marked in the original location sequence, and the starting time node of the reflection impact time range is determined as the path adjustment entry point.

8. The method for analyzing UAV flight data according to claim 7, characterized in that, The continuous segments with extended propagation time and the mirror change segments form overlapping intervals on the time scale. The reflection effect time range covers this overlapping interval and maintains the boundary of the continuous interval along the time progression sequence. The path adjustment entry corresponds to the starting time node of the reflection effect time range.

9. A method for analyzing UAV flight data according to claim 7, characterized in that, The steps to implement time scale misalignment processing to reduce the involvement of mirrored change segments in path planning are as follows: The subsequent track reconstruction process is carried out along the time scale of the original position sequence around the entry point of the path adjustment, maintaining the consistency of the position advancement sequence, and segmenting the time range of reflection impact; Adjust the entry point along the path to the time segment corresponding to the time range of the reflection effect, implement time scale misalignment processing, redistribute the time interval, and slow down the position update pace while keeping the position advancement order unchanged; Maintain the original time scale spacing and position update order around the time segment outside the time range affected by reflection, so that the time scale forms an alternating state of original rhythm and slowed rhythm; By combining the adjusted time scale system with the subsequent track reconstruction process, the spatial location within the time range affected by reflections will participate in path planning at a slower pace, thus reducing the degree of participation of the mirror change segment in path planning.

10. A UAV flight data analysis system, used to implement the UAV flight data analysis method according to any one of claims 1-9, characterized in that, It includes a raw sequence construction module, a propagation offset generation module, a mirror segment identification module, a reflection interval determination module, and a time modulation module: The original sequence construction module collects continuous location change trajectories in urban high-reflection environments, synchronously organizes signal arrival time fluctuation records, and constructs an original location sequence containing detour traces. The propagation offset generation module splits the signal arrival time fluctuation record in the original position sequence and expands the segments with elongated propagation time to form a propagation offset change trajectory that maintains the position advancement order. The mirror segment identification module compares the directions along the position progression sequence of the propagation offset change trajectory, extracts the segments where the position and direction are symmetrically flipped to form mirror change segments, and determines the position and calculates the offset change trend within the mirror change segments. The reflection interval determination module calculates the offset change trend based on the location and traces back the signal arrival time fluctuation record in the original location sequence. It extracts the time range of continuous propagation time to form the reflection influence time range and determines the reflection influence time range as the path adjustment entry point. The time modulation module implements time scale misalignment processing for the subsequent track reconstruction process around the path adjustment entry point. Within the reflection influence time range, it slows down the position update rhythm, and maintains the position update order outside the reflection influence time range. By alternating the time rhythm before and after, it weakens the participation of the mirror change segment in the path planning.

Images