Abnormal point identification method and device, equipment, storage medium and program product

Abstract

The embodiment of the invention provides an abnormal point identification method and device, equipment, a storage medium and a program product, and particularly relates to the technical field of positioning. The method comprises the following steps: determining a track point set corresponding to a current track point, wherein the track point set comprises a plurality of track points in original track data, and the time interval between the track points and the current track point does not exceed a preset threshold value; determining the motion state of the target object at the current track point based on the motion feature data of the track point set; according to the motion state, abnormal point screening conditions are adjusted; and judging whether the current track point is an abnormal point or not by adopting the adjusted screening condition. The method is used for achieving the effects of improving the precision and adaptability of GNSS data screening and providing reliable support for high-precision positioning and navigation.

Description

Technical Field

[0001] This application relates to the field of positioning technology, and in particular to an anomaly identification method, apparatus, device, storage medium, and program product. Background Technology

[0002] GNSS (Global Navigation Satellite System) data is widely used in intelligent transportation, autonomous driving, logistics, geographic information systems (GIS), and mobile device positioning. In these scenarios, the accuracy and continuity of GNSS data directly impact the system's decision-making capabilities and user experience. For example, in autonomous driving systems, vehicles need to receive high-precision positioning data in real time to plan routes, avoid obstacles, and maintain lanes; in logistics, the trajectory data of transport vehicles needs to accurately reflect their actual driving paths to optimize route planning and resource scheduling; and in GIS data acquisition and updating, high-quality GNSS trajectories are the foundation for building digital maps. However, GNSS data often faces multiple challenges in practical applications: signal obstruction in urban canyons leading to positioning drift, trajectory noise caused by differences in equipment accuracy, slow deviations when vehicles are stationary or at low speeds, and trajectory fragmentation in complex road scenarios (such as sharp turns and ramps). These factors result in a large number of outliers (i.e., "drift points") in the raw GNSS data, which, if used directly, can lead to serious consequences such as incorrect navigation paths, increased logistics costs, and distorted map data. Summary of the Invention

[0003] This application provides an anomaly identification method, apparatus, device, storage medium, and program product, which achieve the effect of using a dynamic threshold mechanism to enable the filtering logic to adapt to different scenarios and motion states (such as urban congestion, highway turns), avoiding misjudgments caused by fixed thresholds, and improving the accuracy and adaptability of GNSS data filtering.

[0004] In a first aspect, embodiments of this application provide an anomaly identification method, including:

[0005] Determine the set of trajectory points corresponding to the current trajectory point, wherein the set of trajectory points includes multiple trajectory points in the original trajectory data whose time interval with the current trajectory point does not exceed a preset threshold;

[0006] Based on the motion feature data of the trajectory point set, the motion state of the target object at the current trajectory point is determined;

[0007] Adjust the anomaly point screening criteria based on the described motion state;

[0008] The adjusted filtering criteria are used to determine whether the current trajectory point is an anomaly.

[0009] Secondly, embodiments of this application provide an anomaly identification device, comprising:

[0010] The first determining module is used to determine the set of trajectory points corresponding to the current trajectory point. The set of trajectory points includes multiple trajectory points in the original trajectory data whose time interval with the current trajectory point does not exceed a preset threshold.

[0011] The second determining module is used to determine the motion state of the target object at the current trajectory point based on the motion feature data of the trajectory point set;

[0012] An adjustment module is used to adjust the anomaly point filtering conditions according to the motion state;

[0013] The judgment module is used to determine whether the current trajectory point is an anomaly point using the adjusted filtering conditions.

[0014] Thirdly, embodiments of this application provide an anomaly identification device, including: a memory and a processor;

[0015] The memory stores computer-executed instructions;

[0016] The processor executes computer execution instructions stored in the memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.

[0017] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible implementations of the first aspect.

[0018] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.

[0019] The anomaly identification method, apparatus, device, storage medium, and program product provided in this application first determine a set of trajectory points corresponding to the current trajectory point. This set of trajectory points is filtered based on time intervals and includes multiple trajectory points in the original trajectory data whose time interval with the current trajectory point does not exceed a preset threshold. Next, based on the motion characteristic data of the trajectory point set, such as velocity, acceleration, and heading angle, the motion state of the target object at the current trajectory point is accurately determined. The anomaly filtering conditions are dynamically adjusted according to different motion states, rather than using a fixed threshold. This dynamic adjustment mechanism enables the filtering logic to adapt to different scenarios and motion states, effectively avoiding misjudgment problems caused by fixed thresholds, thereby significantly improving the accuracy and adaptability of GNSS data filtering and providing reliable support for high-precision positioning and navigation. Attached Figure Description

[0020] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0021] Figure 1 Flowchart of the anomaly identification method provided in this application Figure 1 ;

[0022] Figure 2 Flowchart of the anomaly identification method provided in this application Figure 2 ;

[0023] Figure 3 A schematic diagram illustrating the process of the anomaly identification method provided in this application;

[0024] Figure 4 A schematic diagram of the anomaly identification device provided in this application;

[0025] Figure 5 This is a schematic diagram of the anomaly identification device provided in this application.

[0026] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0027] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0028] In existing technologies, GNSS data filtering mainly relies on a fixed-threshold coordinate distance determination method. Specifically, this method calculates the straight-line distance between adjacent points by traversing continuous GPS track points and sets a fixed threshold (e.g., 10 meters). If the distance between two points exceeds this threshold, at least one point is considered an anomaly, and legitimate tracks are retained by segmenting the track and removing anomalies. However, this type of method has significant limitations:

[0029] Fixed thresholds have poor adaptability: A single threshold cannot adapt to different scenarios (such as highways and congested urban roads) and movement states (going straight, turning, stationary). For example, when a vehicle is traveling on a highway, the reasonable distance traveled is much greater than in a low-speed urban scenario. A fixed threshold can easily lead to normal points being misjudged as abnormal points, or abnormal points not being identified at all.

[0030] Lack of dynamic adjustment mechanism: The dynamic characteristics of the trajectory (such as speed changes and heading angle trends) are not considered, which makes it impossible for the threshold to be adaptively optimized according to the scene. For example, a sudden change in heading angle when the vehicle turns may be misjudged as abnormal, while slow drift in a stationary state may not be recognized.

[0031] Weak ability to handle consecutive anomalies: When signal loss or strong interference causes multiple consecutive anomalies, existing methods tend to divide the trajectory into multiple fragmented segments, making it difficult to recover the complete path in subsequent processing and affecting data continuity.

[0032] High risk of misjudgment: The slow drift of stationary equipment (such as vehicles in front of a red light) due to signal reflection may be misjudged as "movement", thus causing trajectory segmentation errors.

[0033] The anomaly identification method provided in this application introduces a dynamic threshold adjustment mechanism to automatically adjust the threshold to adapt to different scenarios; it corrects the position point by introducing a heading angle to avoid trajectory fragmentation caused by consecutive anomalies, and can automatically correct the trajectory to maintain integrity even if consecutive anomalies occur; it determines whether the vehicle is turning, driving on a straight road, or stationary through a state detection mechanism to avoid misjudging as "moving" due to GPS point distance exceeding the threshold. This method performs well in complex environments and dynamic scenarios, significantly improving the accuracy and reliability of GPS drift point processing.

[0034] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0035] Figure 1 Flowchart of the anomaly identification method provided in this application Figure 1 ,like Figure 1 As shown, the method includes:

[0036] S101. Determine the set of trajectory points corresponding to the current trajectory point. The set of trajectory points includes multiple trajectory points in the original trajectory data whose time interval with the current trajectory point does not exceed a preset threshold.

[0037] In this step, multiple trajectory points with time intervals not exceeding a preset threshold are selected from the original trajectory data to form a trajectory point set. These trajectory points can specifically be GNSS points. The time interval selection ensures that the selected trajectory points are temporally correlated with the current trajectory point, providing a data foundation for subsequent analysis.

[0038] S102. Based on the motion feature data of the trajectory point set, determine the motion state of the target object at the current trajectory point.

[0039] In this step, the motion characteristic data of the trajectory point set (such as velocity, acceleration, rate of change of heading angle, etc.) are analyzed to determine the motion state of the target object at the current trajectory point (such as stationary, turning, straight, etc.).

[0040] S103. Adjust the outlier filtering criteria according to the motion state.

[0041] In this step, the anomaly filtering conditions (such as distance filtering threshold, heading angle filtering threshold, etc.) are dynamically adjusted according to the motion state of the target object, so that the filtering conditions can adapt to different motion states and improve the accuracy and adaptability of anomaly identification.

[0042] S104. Use the adjusted filtering conditions to determine whether the current trajectory point is an anomaly.

[0043] Using the adjusted filtering criteria, determine whether the current trajectory point is an anomaly.

[0044] The anomaly identification method provided in this application first determines a set of trajectory points corresponding to the current trajectory point. This set of trajectory points is filtered based on time intervals and includes multiple trajectory points in the original trajectory data whose time interval with the current trajectory point does not exceed a preset threshold. Next, based on the motion feature data of the trajectory point set, such as velocity, acceleration, and heading angle, the motion state of the target object at the current trajectory point is accurately determined. The anomaly filtering conditions are dynamically adjusted according to different motion states, rather than using a fixed threshold. This dynamic adjustment mechanism allows the filtering logic to adapt to different scenarios and motion states, effectively avoiding misjudgment problems caused by fixed thresholds, thereby significantly improving the accuracy and adaptability of GNSS data filtering and providing reliable support for high-precision positioning and navigation.

[0045] Figure 2 Flowchart of the anomaly identification method provided in this application Figure 2 ,like Figure 2 As shown, in this embodiment... Figure 1 Based on the embodiments, the anomaly identification method is described in detail, which includes:

[0046] S201. Determine the set of trajectory points corresponding to the current trajectory point. The set of trajectory points includes multiple trajectory points in the original trajectory data whose time interval with the current trajectory point does not exceed a preset threshold.

[0047] The raw trajectory data can be a data file containing GPS trajectory point information, where each trajectory point specifically includes a timestamp, longitude, latitude, heading angle, and speed field.

[0048] Timestamp (time): Reflects the time when GPS track points were collected. It is used to sort track points in time sequence and calculate the time interval between adjacent points. It is one of the key bases for track segment segmentation and is represented by time.

[0049] Longitude (lon): The longitude coordinates of a GPS track point, together with latitude, determine the geographical location of the track point. It is the basic data for calculating the distance between two points, predicting the position of the track point, and judging the track deviation. It is represented by lon.

[0050] Latitude (lat): The latitude coordinates of a GPS track point. It is used in conjunction with longitude to determine the spatial location of the track point and participates in core logic such as distance calculation, location prediction, and judgment of track deviation. It is represented by lat.

[0051] Heading (had): An angular information reflecting the direction of motion, measured in radians. It is used to calculate changes in heading angle, determine turning status, and predict the direction of trajectory points. It is an important basis for dynamically adjusting the screening threshold and is represented by heading (had).

[0052] Speed ​​(m / s): Information reflecting the rate of motion, measured in meters per second. It is used to determine a stationary state (e.g., if the speed is below a threshold, it is considered stationary), dynamically adjust the time interval threshold, and affect the filtering logic of trajectory points. It is represented by speed (m / s).

[0053] In this step, the length of the time window is dynamically adjusted based on the target object's speed at the current trajectory point. For example, when the target object's speed is high, the time window length is extended to include more trajectory points, ensuring sufficient motion state analysis; when the speed is low, the time window length is shortened to avoid introducing irrelevant trajectory points and improve processing efficiency.

[0054] When determining the set of trajectory points corresponding to the current trajectory point, based on the dynamically adjusted time window length, multiple trajectory points whose timestamps and the time interval between the current trajectory point do not exceed a preset threshold (i.e., the time window length) are selected from the original trajectory data to form a trajectory point set. This trajectory point set is used for subsequent extraction of motion feature data (such as speed and rate of change of heading angle) and determination of motion state (such as straight ahead, turning, or stationary).

[0055] S202. Based on the motion feature data of the trajectory point set, determine the motion state of the target object at the current trajectory point.

[0056] In this step, the motion characteristic data of the set of trajectory points corresponding to the current trajectory point is analyzed to determine the specific motion state of the target object at the current trajectory point. Motion characteristic data includes, but is not limited to, velocity, acceleration, heading angle, and its rate of change. Using this data, it is possible to identify whether the target object is stationary, turning, or moving straight, among other states.

[0057] S203. If the motion state is stationary, then adjust the distance filtering threshold and heading angle filtering threshold according to the first strategy.

[0058] In this step, when the target object is stationary, the first strategy is applied to adjust the distance filtering threshold and the heading angle filtering threshold.

[0059] For example, when the average velocity of the trajectory point set is less than a preset static velocity threshold and the standard deviation of the velocity of the trajectory point set is less than a preset standard deviation threshold, the target object is determined to be in a static state.

[0060] The specific judgment formula can be expressed as follows:

[0061] Calculate the average velocity of all points within the trajectory point set. ;

[0062] Calculate the standard deviation of the velocities of all points within the trajectory point set. ;

[0063] The calculated average velocity Compared with the preset static velocity threshold Compare;

[0064] The calculated speed standard deviation Compared with the preset standard deviation threshold Compare;

[0065] If both of the following conditions are met:

[0066] .

[0067] Then it is determined that the target object is stationary at the current trajectory point.

[0068] At this point, adjust the distance filtering threshold according to the first strategy. and heading angle screening threshold The specific adjustment method is as follows:

[0069] Distance threshold corresponding to the static state Set as the adjusted distance filtering threshold ;

[0070] Set the basic heading angle difference threshold Double the adjusted heading angle filtering threshold .

[0071] The adjusted distance filtering threshold and heading angle filtering threshold can be expressed by the following formulas:

[0072] .

[0073] S204. If the motion state is turning, then adjust the distance filtering threshold and heading angle filtering threshold according to the second strategy.

[0074] In this step, when the target object is turning, the second strategy is applied to adjust the distance filtering threshold and the heading angle filtering threshold.

[0075] For example, when the change in heading angle of several consecutive trajectory points in the trajectory point set is greater than the preset turning state heading angle change threshold, and the change in heading angle shows the same trend (i.e. the direction of the continuous heading angle change is consistent), the target object is determined to be in a turning state.

[0076] At this point, adjust the distance filtering threshold according to the second strategy. and heading angle screening threshold The specific adjustment formula is as follows:

[0077] ;

[0078] ;

[0079] Adjusted distance filtering threshold By using the base distance threshold Multiply by distance adjustment factor This can be calculated.

[0080] Adjusted heading angle filtering threshold By thresholding the basic heading angle difference value Multiply by the heading angle adjustment factor This can be calculated.

[0081] S205. If the motion state is straight, then adjust the distance filtering threshold and heading angle filtering threshold according to the third strategy.

[0082] In this step, when the target object is in a straight-line state, the third strategy is applied to adjust the distance filtering threshold and the heading angle filtering threshold.

[0083] When the motion feature data of the trajectory point set does not meet the conditions for determining a stationary state and does not meet the conditions for determining a turning state, it can be determined that the target object is in a straight-moving state.

[0084] At this point, adjust the distance filtering threshold according to the third strategy. and heading angle screening threshold The specific adjustment formula is as follows:

[0085] ;

[0086] ;

[0087] Base distance threshold Set as the adjusted distance filtering threshold ;

[0088] Set the basic heading angle difference threshold Set to the adjusted heading angle filtering threshold .

[0089] S206. Use the adjusted filtering conditions to determine whether the current trajectory point is an anomaly.

[0090] In one possible implementation, the adjusted filtering criteria are used to determine whether the current trajectory point is an outlier, which may specifically include the following steps:

[0091] If the motion state is stationary, calculate the relative distance between the current trajectory point and the previous trajectory point;

[0092] If the relative distance between the current trajectory point and the previous trajectory point is not less than the adjusted distance filtering threshold, then the current trajectory point is determined to be an anomaly.

[0093] In this implementation, if the target object is determined to be stationary, the relative distance between the current trajectory point and the previous trajectory point is calculated. This relative distance refers to the actual spatial distance between two consecutive trajectory points. The calculated relative distance is compared with a distance filtering threshold previously adjusted based on the stationary state. If the relative distance between the current trajectory point and the previous trajectory point is not less than this adjusted distance filtering threshold, the current trajectory point is determined to be an anomaly. This is because a large movement distance should not occur in a stationary state; if such a situation occurs, it may be due to GNSS signal errors or other reasons causing abnormal data points.

[0094] The formula for calculating the distance between adjacent trajectory points can be shown below:

[0095]

[0096] in:

[0097] , representing the difference in latitude between two points (converted to radians);

[0098] , representing the difference in longitude between two points (converted to radians);

[0099] , It is the latitude and longitude of the first point (after converting to radians);

[0100] , It is the latitude and longitude of the second point (after converting to radians);

[0101] It is the Earth's radius.

[0102] In one possible implementation, the adjusted filtering threshold is used to determine whether the current trajectory point is an outlier, which may specifically include the following steps:

[0103] If the motion state is turning or going straight, then trajectory prediction is performed based on the motion feature data of the trajectory point set to obtain the predicted trajectory point. The predicted trajectory point is the predicted position point of the target object at the current time, while the current trajectory point is the actual position point of the target object at the current time.

[0104] Calculate the deviation between the current trajectory point and the predicted trajectory point, and determine the difference in heading angle between the target object at the current trajectory point and the previous trajectory point;

[0105] Based on the degree of deviation and / or the difference in heading angle, an adjusted filtering threshold is used to determine whether the current trajectory point is an anomaly.

[0106] In this embodiment, if the motion state is identified as turning or straight-line, trajectory prediction is performed based on the motion feature data of the trajectory point set. Trajectory prediction involves calculating the position the target object should reach at the current moment, i.e., the predicted trajectory point. The predicted trajectory point (the theoretically correct position the target object should reach) is compared with the current trajectory point (the actual position the target object has reached), and the deviation between the current trajectory point and the predicted trajectory point is calculated. Simultaneously, the difference in heading angle between the current trajectory point and the previous trajectory point is determined. This helps in understanding changes in the target object's direction of motion. Based on the calculated deviation and / or heading angle difference, an adjusted filtering threshold is used to determine whether the current trajectory point is an anomaly.

[0107] In one approach, prediction can be made based on the position, velocity, and heading angle information of the previous trajectory point. The position prediction formula involved is shown below:

[0108] This indicates the velocity based on the previous trajectory point. and heading angle Calculated latitude change , where R is the Earth's radius;

[0109] This indicates the velocity based on the previous trajectory point. and heading angle Calculated change in longitude ,in, It is the latitude of the previous trajectory point;

[0110] This indicates that the coordinates are based on the latitude of the previous trajectory point. Add latitude variation To calculate the latitude of the predicted trajectory points;

[0111] This indicates that the longitude of the previous trajectory point is used as a reference. Add longitude variation To calculate the longitude of the predicted trajectory points.

[0112] In another approach, prediction can be made based on the position, velocity, and heading angle information of each trajectory point in the trajectory point set. Specifically, trajectory prediction can be based on the average velocity and average heading angle change of each trajectory point in the trajectory point set, as shown in the following formula:

[0113] , represents the average velocity of all points in the trajectory point set, and n represents the number of trajectory points;

[0114] , represents the average change in heading angle between adjacent points in the calculation trajectory point set;

[0115] , representing the heading angle based on the last trajectory point in the trajectory point set. Add the average heading angle change To calculate the heading angle of the predicted trajectory point;

[0116] The calculated average velocity and predicted heading angle Substitute into the above position prediction formula (i.e., use) In the substitution formula ,use In the substitution formula This allows us to obtain the position of the smoothed predicted trajectory points.

[0117] In one possible implementation, calculating the deviation between the current trajectory point and the predicted trajectory point may specifically include the following steps:

[0118] Calculate the vertical distance from the current trajectory point to the line connecting the predicted trajectory point and the previous trajectory point, and use this distance as a quantification of the degree of deviation.

[0119] In this embodiment, a predicted trajectory point is calculated using the sliding window smoothing prediction method or other trajectory prediction techniques described above. This point is the predicted position that the target object should reach at the current moment, based on the target object's historical motion characteristic data (such as speed, heading angle, etc.). A line is determined connecting the previous trajectory point and the predicted trajectory point. The vertical distance from the current trajectory point to this reference line is calculated. The calculated vertical distance is used as a quantified value of the deviation.

[0120] This can be achieved by first converting the current trajectory point, the predicted trajectory point, and the previous trajectory point into a three-dimensional coordinate vector form. For any point... Its three-dimensional coordinate vector can be represented as:

[0121] ;

[0122] set up:

[0123] Indicates the previous trajectory point:

[0124] Indicates the current trajectory point;

[0125] This represents the predicted trajectory point.

[0126] The formula for calculating vertical distance is as follows:

[0127] ;

[0128] This formula calculates the perpendicular distance from point b to the line connecting points a and c. This can be used to determine whether the current trajectory point b deviates from the predicted trajectory ac.

[0129] In one possible implementation, determining the difference in heading angle between the target object at the current trajectory point and the previous trajectory point may specifically include the following steps:

[0130] Calculate the difference between the heading angle of the current trajectory point and the heading angle of the previous trajectory point, and use this as the quantified value of the heading angle difference.

[0131] In this embodiment, the calculated heading angle difference can also be constrained to... Within the specified range (to avoid misjudgments caused by the periodicity of angles), the specific formula is as follows:

[0132]

[0133] in, , It is the absolute difference between the heading angle of the current trajectory point and the heading angle of the previous trajectory point.

[0134] if If > π, then = This means mapping the difference to the range [0,π].

[0135] if If ≤π, then = That is, the difference is already within the range of [0,π], and no adjustment is needed.

[0136] In one possible implementation, based on the degree of deviation and / or the difference in heading angle, an adjusted screening threshold is used to determine whether the current trajectory point is an outlier. This may specifically include the following steps:

[0137] If the quantified value of the deviation is not less than the adjusted distance screening threshold, or the quantified value of the heading angle difference is not less than the adjusted heading angle screening threshold, then the current trajectory point is determined to be an anomaly.

[0138] If the target object is stationary, the current trajectory point can be determined as an anomaly only if the relative distance between the current trajectory point and the previous trajectory point is not less than the adjusted distance filtering threshold. The expression can be as follows:

[0139] ;

[0140] in, This indicates the relative distance between the current trajectory point and the previous trajectory point.

[0141] If the target object is not stationary, both the deviation degree condition and the heading angle difference condition must be met simultaneously to be determined as a normal point. In other words, if either condition is not met, the current trajectory point can be determined as an anomaly point. The expression can be as follows:

[0142] or .

[0143] In one possible implementation, if the number of abnormal points in the trajectory point set corresponding to the current trajectory point reaches or exceeds a preset threshold, such as n-1 (where n is the number of trajectory points in the trajectory point set), then the current trajectory point is forcibly marked as a normal point to avoid over-filtering. The expression can be as follows:

[0144] .

[0145] The method described in this application can output a complete dataset segmented by trajectory. This dataset includes raw GPS trajectory point information (including timestamps, longitude, latitude, heading angle, and velocity), quality markers for trajectory points (distinguishing between normal and abnormal points), motion state markers (stationary, turning, or straight-ahead), and dynamic threshold parameters. All trajectory points marked as "normal" can be filtered from the dataset; these points can be used to construct continuous and reliable GPS trajectories. Detailed statistics for each trajectory segment are provided, including the number of points and the proportion of normal points (the percentage of normal points stationary, turning, and straight-ahead), to support a comprehensive evaluation of the trajectory data quality.

[0146] Reference Figure 3 The diagram illustrates the process of the anomaly identification method provided in this application. First, trajectory data is input. Information such as the position, velocity, heading angle, and timestamp of each trajectory point in the trajectory data is loaded. During the initialization phase, all points within the front window of the trajectory segment to be analyzed are preset as normal points. This step sets an initial benchmark for trajectory data processing, ensuring that subsequent analysis and judgment are based on a relatively reliable set of starting points. For the current trajectory point, historical trajectory point data of the corresponding window is extracted, including velocity and heading angle sequences. The motion characteristic data of the trajectory point set within the window is analyzed to determine the specific motion state of the target object at the current trajectory point. Based on the motion state of the target object, the anomaly filtering conditions are dynamically adjusted. If the target object is in a non-stationary state, trajectory prediction (smoothing prediction) is performed based on the motion characteristic data of the trajectory point set within the window to determine the position the target object should reach at the current moment, i.e., the predicted trajectory point. The predicted trajectory point (the theoretically the position the target object should reach) is compared with the current trajectory point (the actual position the target object reaches), and the vertical distance from the current trajectory point to the line connecting the predicted trajectory point and the previous trajectory point (i.e., the predicted trajectory) is calculated, and it is determined whether it is less than the adjusted distance filtering threshold. If not, the point is an anomaly, and a predicted trajectory point is used instead of the actual trajectory point. Additionally, the normalized difference between the heading angle of the current trajectory point and the heading angle of the previous trajectory point can be calculated, and it can be determined whether it is less than the adjusted heading angle filtering threshold. If both conditions are met, the point is a normal point; otherwise, the point is an anomaly. After correcting the trajectory data, the target trajectory can be output.

[0147] This application provides a method for GNSS data filtering based on heading angle, applicable to various scenarios and GNSS data of different precision. The method extracts key information such as timestamp, latitude and longitude, heading angle, and velocity corresponding to each trajectory point, sorts the trajectory points by time, processes each trajectory segment, and calculates the predicted trajectory point position using a sliding window. The predicted trajectory point position is calculated based on latitude and longitude, heading angle, and velocity. GNSS trajectory points meeting the criteria are filtered based on dynamic thresholds (adjusted according to stationary, turning, etc.). This method solves the trajectory fragmentation problem caused by consecutive anomalies.

[0148] This application breaks through the limitations of existing methods that rely solely on coordinates (distance) for point selection, and innovatively incorporates heading angle into the core judgment dimension. By introducing heading angle to smoothly predict trajectory position, this design solves the shortcomings of simple coordinate selection—for example, for abnormal points with similar coordinates but abrupt changes in heading (not in line with the continuity of actual motion), distance alone cannot identify them, but combining heading angle can accurately eliminate them, making the selection results more consistent with the physical laws of real motion (such as the heading continuity of vehicle movement).

[0149] This application overcomes the limitations of traditional fixed thresholds by dynamically adjusting the filtering threshold based on the motion state (stationary, straight, turning): a larger distance threshold is used when stationary, a larger heading threshold is used when turning, and a basic threshold is used when straight. This dynamic threshold strategy solves the problem of poor adaptability of fixed thresholds in complex road conditions (such as turning and waiting at traffic lights), and improves the accuracy of point selection in different scenarios.

[0150] This application employs a sliding window approach to fuse historical data (average speed and heading trend) for trajectory point position prediction, rather than simply relying on the state of the previous moment. This multi-frame information fusion prediction method reduces the impact of single-point noise on trajectory continuity, especially in urban environments with large data fluctuations, generating smoother trajectories that better reflect actual motion patterns.

[0151] The anomaly identification method provided in this application first determines a set of trajectory points corresponding to the current trajectory point. This set of trajectory points is filtered based on time intervals and includes multiple trajectory points in the original trajectory data whose time interval with the current trajectory point does not exceed a preset threshold. Next, based on the motion feature data of the trajectory point set, such as velocity, acceleration, and heading angle, the motion state of the target object at the current trajectory point is accurately determined. The anomaly filtering conditions are dynamically adjusted according to different motion states, rather than using a fixed threshold. This dynamic adjustment mechanism allows the filtering logic to adapt to different scenarios and motion states, effectively avoiding misjudgment problems caused by fixed thresholds, thereby significantly improving the accuracy and adaptability of GNSS data filtering and providing reliable support for high-precision positioning and navigation.

[0152] Figure 4 A schematic diagram of the anomaly identification device provided in this application is shown below. Figure 4 As shown, the anomaly identification device 40 provided in this embodiment includes:

[0153] The first determining module 401 is used to determine the trajectory point set corresponding to the current trajectory point. The trajectory point set includes multiple trajectory points in the original trajectory data whose time interval with the current trajectory point does not exceed a preset threshold.

[0154] The second determining module 402 is used to determine the motion state of the target object at the current trajectory point based on the motion feature data of the trajectory point set.

[0155] Adjustment module 403 is used to adjust the abnormal point filtering conditions according to the motion state;

[0156] The judgment module 404 is used to determine whether the current trajectory point is an abnormal point using the adjusted filtering conditions.

[0157] In one possible implementation, the filtering criteria include a distance filtering threshold and a heading angle filtering threshold, and the adjustment module is specifically used for:

[0158] If the motion state is stationary, then the distance filtering threshold and heading angle filtering threshold are adjusted according to the first strategy;

[0159] If the motion state is turning, then adjust the distance filtering threshold and heading angle filtering threshold according to the second strategy;

[0160] If the motion state is straight, then the distance filtering threshold and heading angle filtering threshold are adjusted according to the third strategy.

[0161] In one possible implementation, the decision module is specifically used for:

[0162] If the motion state is stationary, calculate the relative distance between the current trajectory point and the previous trajectory point;

[0163] If the relative distance between the current trajectory point and the previous trajectory point is not less than the adjusted distance filtering threshold, then the current trajectory point is determined to be an anomaly.

[0164] In one possible implementation, the decision module is specifically used for:

[0165] If the motion state is turning or going straight, then trajectory prediction is performed based on the motion feature data of the trajectory point set to obtain the predicted trajectory point. The predicted trajectory point is the predicted position point of the target object at the current time, while the current trajectory point is the actual position point of the target object at the current time.

[0166] Calculate the deviation between the current trajectory point and the predicted trajectory point, and determine the difference in heading angle between the target object at the current trajectory point and the previous trajectory point;

[0167] Based on the degree of deviation and / or the difference in heading angle, an adjusted filtering threshold is used to determine whether the current trajectory point is an anomaly.

[0168] In one possible implementation, the decision module is specifically used for:

[0169] Calculate the vertical distance from the current trajectory point to the line connecting the predicted trajectory point and the previous trajectory point, and use this distance as a quantification of the degree of deviation.

[0170] In one possible implementation, the decision module is specifically used for:

[0171] Calculate the difference between the heading angle of the current trajectory point and the heading angle of the previous trajectory point, and use this as the quantified value of the heading angle difference.

[0172] In one possible implementation, the decision module is specifically used for:

[0173] If the quantified value of the deviation is not less than the adjusted distance screening threshold, or the quantified value of the heading angle difference is not less than the adjusted heading angle screening threshold, then the current trajectory point is determined to be an anomaly.

[0174] The anomaly identification device provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0175] Figure 5 This is a schematic diagram of the anomaly identification device provided in this application. Figure 5 As shown, the anomaly identification device 50 provided in this embodiment includes at least one processor 501 and a memory 502. Optionally, the device 50 further includes a communication component 503. The processor 501, memory 502, and communication component 503 are connected via a bus.

[0176] In a specific implementation, at least one processor 501 executes computer execution instructions stored in memory 502, causing at least one processor 501 to perform the above-described method.

[0177] The specific implementation process of processor 501 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0178] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0179] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0180] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0181] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0182] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0183] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0184] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0185] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0186] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0187] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0188] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0189] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0190] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. An anomaly identification method, characterized in that, include: Determine the set of trajectory points corresponding to the current trajectory point, wherein the set of trajectory points includes multiple trajectory points in the original trajectory data whose time interval with the current trajectory point does not exceed a preset threshold; Based on the motion feature data of the trajectory point set, the motion state of the target object at the current trajectory point is determined; Adjust the anomaly point screening criteria based on the described motion state; The adjusted filtering criteria are used to determine whether the current trajectory point is an anomaly.

2. The method according to claim 1, characterized in that, The filtering criteria include a distance filtering threshold and a heading angle filtering threshold. Adjusting the anomaly filtering criteria based on the motion state includes: If the motion state is a stationary state, then the distance filtering threshold and the heading angle filtering threshold are adjusted according to the first strategy; If the motion state is a turning state, then the distance filtering threshold and the heading angle filtering threshold are adjusted according to the second strategy; If the motion state is a straight-line state, then the distance filtering threshold and the heading angle filtering threshold are adjusted according to the third strategy.

3. The method according to claim 1 or 2, characterized in that, The step of determining whether the current trajectory point is an anomaly using the adjusted filtering criteria includes: If the motion state is a stationary state, then calculate the relative distance between the current trajectory point and the previous trajectory point; If the relative distance between the current trajectory point and the previous trajectory point is not less than the adjusted distance filtering threshold, then the current trajectory point is determined to be an anomaly.

4. The method according to claim 1 or 2, characterized in that, The step of determining whether the current trajectory point is an anomaly using the adjusted filtering threshold includes: If the motion state is a turning state or a straight state, then trajectory prediction is performed based on the motion feature data of the trajectory point set to obtain the predicted trajectory point. The predicted trajectory point is the predicted position point of the target object at the current time, while the current trajectory point is the actual position point of the target object at the current time. Calculate the degree of deviation between the current trajectory point and the predicted trajectory point, and determine the difference in heading angle between the target object at the current trajectory point and the previous trajectory point; Based on the degree of deviation and / or the difference in heading angle, an adjusted filtering threshold is used to determine whether the current trajectory point is an anomaly.

5. The method according to claim 4, characterized in that, The calculation of the deviation between the current trajectory point and the predicted trajectory point includes: Calculate the vertical distance from the current trajectory point to the line connecting the predicted trajectory point and the previous trajectory point, and use this distance as a quantification value of the degree of deviation.

6. The method according to claim 4, characterized in that, Determining the difference in heading angle between the target object at the current trajectory point and the previous trajectory point includes: Calculate the difference between the heading angle of the current trajectory point and the heading angle of the previous trajectory point, and use this as the quantified value of the heading angle difference.

7. The method according to claim 4, characterized in that, The step of determining whether the current trajectory point is an anomaly point based on the degree of deviation and / or the difference in heading angle using an adjusted filtering threshold includes: If the quantified value of the deviation is not less than the adjusted distance screening threshold, or the quantified value of the heading angle difference is not less than the adjusted heading angle screening threshold, then the current trajectory point is determined to be an anomaly.

8. An anomaly identification device, characterized in that, include: The first determining module is used to determine the set of trajectory points corresponding to the current trajectory point. The set of trajectory points includes multiple trajectory points in the original trajectory data whose time interval with the current trajectory point does not exceed a preset threshold. The second determining module is used to determine the motion state of the target object at the current trajectory point based on the motion feature data of the trajectory point set; An adjustment module is used to adjust the anomaly point filtering conditions according to the motion state; The judgment module is used to determine whether the current trajectory point is an anomaly point using the adjusted filtering conditions.

9. An anomaly identification device, characterized in that, include: Memory, processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1-7.

10. A computer-readable storage medium or computer program product, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-7; And / or, The computer program product includes a computer program that, when executed by a processor, implements the method described in any one of claims 1-7.

Images