Interruption track association method and device, storage medium and electronic equipment

Abstract

The application relates to a method and device for interrupting track association, a storage medium and an electronic device, comprising: acquiring a pre-interruption track and a post-interruption track; extracting at least one target point information from the pre-interruption track and at least one target point information from the post-interruption track to obtain a point information sequence; determining point feature information corresponding to each target point information; determining target feature information corresponding to each target point information in the point information sequence and related to other target point information in the point information sequence according to the point feature information corresponding to each target point information; and associating the pre-interruption track and the post-interruption track according to the target feature information corresponding to each target point information. Thus, the pre-interruption track and the post-interruption track can be associated based on the association characteristics, without complex calculation of track data, and the environment can be self-adapted, so that the accuracy of the interrupting track association can be improved.

Description

Technical Field

[0001] This application relates to the field of computer technology, specifically to a method, apparatus, storage medium, and electronic device for linking interrupted flight paths. Background Technology

[0002] In the field of radar data processing, tasks such as target tracking, situational awareness, and information fusion rely heavily on track segment association. If a track is interrupted, it will have a severe impact on these tasks, causing significant losses. There are many reasons for track interruptions, such as high-speed target maneuvering, deliberate target stopping, low sensor detection probability, long sensor sampling intervals, clutter, and electromagnetic interference. Existing track association schemes for interrupted tracks mainly fall into two categories: the first predicts the track based on information from the track before the interruption and then calculates feature similarity between the predicted track and the newly acquired track information; the second directly calculates feature similarity between the features from the track before the interruption and the newly acquired track.

[0003] However, the first interrupted track association scheme heavily relies on the accuracy of track prediction. Due to target maneuverability, measurement tool errors, and other factors, it cannot accurately predict tracks; especially Kalman filter-based track prediction, which cannot predict tracks if new heading information is lacking. The second interrupted track association scheme mainly constructs target behavioral features and analyzes these features before and after track interruption to calculate track similarity. However, due to the diverse scenarios of targets and the difficulty in generalizing features, the similarity calculation results are not accurate enough, thus failing to accurately associate interrupted tracks.

[0004] Therefore, the accuracy of existing interrupted track association schemes needs to be improved. Summary of the Invention

[0005] The purpose of this application is to provide a method, apparatus, storage medium, and electronic device for interrupted track association, so as to improve the accuracy of interrupted track association.

[0006] This application provides a method for interrupting track association, which includes:

[0007] Obtain the track before and after the interruption;

[0008] Extract at least one target point location information from the track before the interruption and at least one target point location information from the track after the interruption to obtain a point location information sequence including multiple target point location information;

[0009] Determine the location feature information corresponding to each target location in the location information sequence;

[0010] Based on the location feature information corresponding to each target location information in the location information sequence, determine the target feature information related to other target location information in the location information sequence for each target location information in the location information sequence.

[0011] Based on the target feature information corresponding to each target point in the point information sequence, the pre-interruption trajectory and the post-interruption trajectory are correlated.

[0012] This application embodiment also provides a device for interrupting track association, the device for interrupting track association includes:

[0013] The acquisition module is used to acquire the track before and after the interruption;

[0014] The extraction module is used to extract at least one target point information from the pre-interruption track and at least one target point information from the post-interruption track, so as to obtain a point information sequence including multiple target point information.

[0015] The first determining module is used to determine the point feature information corresponding to each target point in the point information sequence;

[0016] The second determining module is used to determine the target feature information related to other target point information in the point information sequence based on the point feature information corresponding to each target point information in the point information sequence.

[0017] The association module is used to associate the pre-interruption trajectory and the post-interruption trajectory based on the target feature information corresponding to each target point in the point information sequence.

[0018] Specifically, the second determining module is used for:

[0019] The point feature information corresponding to each target point in the point information sequence is input into the trained attention mechanism model for processing, so as to obtain the target feature information related to other target points in the point information sequence for each target point in the point information sequence.

[0020] The associated modules include:

[0021] The input unit is used to input the target feature information corresponding to each target point in the point information sequence into the trained classification model to obtain the classification result. The classification result is used to indicate whether there is a correlation between the track before the interruption and the track after the interruption.

[0022] The association unit is used to associate the pre-interruption track and the post-interruption track when the classification result indicates that there is an association between them.

[0023] The first determining module includes:

[0024] The first processing unit is used to discretize each target point information in the point information sequence to obtain the corresponding discretized point information.

[0025] The determining unit is used to determine the point feature information corresponding to each target point in the point information sequence based on the discretized point information.

[0026] The target location information includes continuous variable data of longitude and continuous variable data of latitude; the first processing unit is specifically used for:

[0027] For each target point in the point information sequence, the continuous variable data of longitude and latitude are converted into degrees, minutes and seconds to obtain the corresponding integer part data, fractional part data and second part data of longitude and latitude.

[0028] The second portion of the data is rounded down to obtain the rounded second portion of the data, resulting in discretized point information that includes the integer portion of the data, the fractional portion of the data, and the rounded second portion of the data.

[0029] The target point information includes continuous variable data of point speed and continuous variable data of point heading; the first processing unit is specifically used for:

[0030] The continuous variable data of point speed and continuous variable data of point heading included in each target point information sequence are rounded to obtain the corresponding rounded point speed data and rounded point heading data, thus obtaining discretized point information including the rounded point speed data and rounded point heading data.

[0031] Specifically, the determining unit is used for:

[0032] The discretized point information corresponding to each target point in the point information sequence is encoded to obtain the corresponding point feature information.

[0033] This application also provides a computer-readable storage medium storing a computer program adapted for loading by a processor to perform the steps in the interrupted track association method described above.

[0034] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the steps in the interrupted track association method described above by calling the computer program stored in the memory.

[0035] The interrupted flight track association method, apparatus, storage medium, and electronic device provided in this application acquire the pre-interruption flight track and the post-interruption flight track, then extract at least one target point information from the pre-interruption flight track and at least one target point information from the post-interruption flight track to obtain a point information sequence including multiple target point information. Next, the point feature information corresponding to each target point information in the point information sequence is determined, and based on the point feature information corresponding to each target point information in the point information sequence, target feature information related to other target point information in the point information sequence is determined. Then, based on the target feature information corresponding to each target point information in the point information sequence, the pre-interruption flight track and the post-interruption flight track are associated. Therefore, when associating interrupted flight tracks, the association features of the pre-interruption and post-interruption flight tracks can be extracted, and the interrupted flight track association is performed based on the extracted association features. This eliminates the need for complex calculations on the flight track data and allows for environmental adaptation, thus improving the accuracy of associated interrupted flight tracks. Attached Figure Description

[0036] The technical solution and other beneficial effects of this application will become apparent from the following detailed description of specific embodiments in conjunction with the accompanying drawings.

[0037] Figure 1 This is a schematic diagram of a scenario for the interrupted track association system provided in an embodiment of this application;

[0038] Figure 2 This is a flowchart illustrating the interrupted track association method provided in an embodiment of this application;

[0039] Figure 3 This is a schematic diagram illustrating the effect of associating the pre-interruption track and the post-interruption track according to an embodiment of this application;

[0040] Figure 4 This is another flowchart illustrating the interrupted track association method provided in this application embodiment;

[0041] Figure 5 This is a schematic diagram of the structure of the classification model provided in the embodiments of this application;

[0042] Figure 6 This is a schematic diagram of the structure of the interrupted track association device provided in the embodiments of this application;

[0043] Figure 7 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0044] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0045] In the field of radar data processing, tasks such as target tracking, situational awareness, and information fusion rely heavily on track segment association. If a track is interrupted, it will have a severe impact on these tasks, causing significant losses. There are many reasons for track interruptions, such as high-speed target maneuvering, deliberate stopping; low sensor detection probability and long sampling intervals; clutter; and electromagnetic interference. To stitch together interrupted tracks and lay a solid foundation for downstream tasks, various interrupted track reconnection algorithms have been proposed. These algorithms can be divided into two categories: prediction methods and similarity measurement methods.

[0046] An interrupted track can be viewed as two parts: the track before the interruption and the track after the interruption. The prediction method involves predicting the track before the interruption forward and the track after the interruption backward, thereby repairing the interrupted interval and establishing a connection. The difference between various prediction methods lies in the different track prediction techniques used. One track prediction method uses a discrete optimization method based on Kalman filtering to predict and connect the old and new tracks. However, Kalman filtering cannot accurately model the probability density function of the target motion. It only achieves good tracking performance when the nonlinear characteristics and non-Gaussian noise of the target motion remain within a reasonable range. When the nonlinear characteristics and non-Gaussian noise of the target motion become severe, the tracking performance of this method drops sharply, and it cannot effectively track the target. Another trajectory prediction method is the expectation-maximization algorithm. The expectation-maximization algorithm takes into account the implicit dependencies of the model in the prediction process, that is, the interdependencies between motion attributes such as position, velocity, acceleration, and angular velocity. It predicts the trajectory by estimating the motion state of the maneuvering target, thus having good adaptability to nonlinear motion. However, this prediction method requires complex calculations on trajectory data, the time spent interrupting trajectory association is large, and the association effect of the algorithm will be severely deteriorated under dense and complex environmental conditions.

[0047] Similarity measurement methods determine whether two track segments belong to the same target by using the similarity between them. The key to these algorithms is finding a suitable track similarity measurement method to make the judgment more accurate. Traditional methods, such as those based on structured features of local tracks to measure the similarity between track segments, and those based on the concept of fuzzy similarity proposed by fuzzy mathematics theory, all rely on the environment of the target to interrupt track association as needed. They interrupt track association based on predefined similarity measurement formulas and lack a similarity measurement method that can adapt to the environment.

[0048] To address the issues of high computational complexity in trajectory prediction and the lack of environmental adaptability in similarity-based interrupted trajectory association, embodiments of this application provide an interrupted trajectory association method, apparatus, computer-readable storage medium, and electronic device. When associating interrupted trajectories, this method can extract association features between the pre-interruption and post-interruption trajectories and associate them based on these extracted features. This eliminates the need for complex calculations of trajectory data and allows for environmental adaptability, thereby improving the accuracy of interrupted trajectory association.

[0049] This application provides a method, apparatus, storage medium, and electronic device for interrupting flight path association.

[0050] Please see Figure 1 , Figure 1 This is a schematic diagram of a scenario for an interrupted flight path association system provided in an embodiment of this application. The interrupted flight path association system may include any interrupted flight path association device provided in an embodiment of this application. The interrupted flight path association device may be integrated into an electronic device such as a terminal or a server. The terminal may be a smartphone, tablet computer, smart Bluetooth device, laptop computer, or personal computer (PC). The server may be a single server or a server cluster composed of multiple servers.

[0051] The aforementioned electronic equipment can acquire the pre-interruption track and the post-interruption track; extract at least one target point information from the pre-interruption track and at least one target point information from the post-interruption track to obtain a point information sequence including multiple target point information; determine the point feature information corresponding to each target point information in the point information sequence; determine the target feature information related to other target point information in the point information sequence based on the point feature information corresponding to each target point information in the point information sequence; and associate the pre-interruption track and the post-interruption track based on the target feature information corresponding to each target point information in the point information sequence.

[0052] The pre-interruption track and the post-interruption track can each be composed of at least one point information, and each point information can include information such as time, longitude, latitude, speed, and heading. Point feature information can be used to characterize the characteristics of its corresponding target point information itself. Target feature information can be used to characterize the characteristics of its corresponding target point information related to other target point information.

[0053] Specifically, such as Figure 1 As shown, the aforementioned interrupted flight path association system may include a terminal and a server communicatively connected to the terminal. When a user wants to associate the pre-interruption and post-interruption flight paths, they can send an interrupted flight path association request to the server through their terminal. The server can then receive the request and retrieve the pre-interruption and post-interruption flight paths accordingly. The server can then extract at least one target point from both the pre-interruption and post-interruption flight paths to obtain a point information sequence containing multiple target point information. Next, the server can determine the point feature information corresponding to each target point in the point information sequence, and based on this feature information, determine the target feature information related to other target points in the sequence. Finally, the server can associate the pre-interruption and post-interruption flight paths based on the target feature information corresponding to each target point in the sequence to form a complete flight path.

[0054] like Figure 2 As shown, Figure 2 This is a flowchart illustrating the interrupted track association method provided in an embodiment of this application. The specific process of the interrupted track association method can be as follows:

[0055] S101. Obtain the track before and after the interruption.

[0056] In this embodiment, the above-described interrupted track association method can be applied, but is not limited to, to interrupted track association scenarios for surface targets (e.g., ships). In some embodiments, the above-described interrupted track association method can also be applied to interrupted track association scenarios for underwater targets (e.g., submarines) and airborne targets (e.g., aircraft). Accordingly, the pre-interruption track and the post-interruption track can be the track of a surface target, the track of an underwater target, or the track of an airborne target.

[0057] Specifically, the aforementioned pre-interruption and post-interruption tracks can each be composed of at least one point of information, and each point of information can include the recording time, longitude, latitude, speed, and heading of the point. Furthermore, in practice, the pre-interruption and post-interruption tracks can be obtained by recording the point of the target (e.g., a ship) at fixed intervals (e.g., 10 seconds) during the target's (e.g., a ship's) navigation. Wherein, as... Figure 3 As shown, the recording time t of each point information used to form the pre-interruption track is before the recording time t of each point information used to form the post-interruption track, and the recording time difference Δt between the latest point information recorded at time t in the pre-interruption track and the earliest point information recorded at time t in the post-interruption track can be greater than the time interval of the recorded point information (i.e., the aforementioned fixed duration).

[0058] S102. Extract at least one target point location information from the pre-interruption track and at least one target point location information from the post-interruption track to obtain a point location information sequence including multiple target point location information.

[0059] In this embodiment, the aforementioned point information sequence comprises at least one target point information extracted from the pre-interruption trajectory and at least one target point information extracted from the post-interruption trajectory. The at least one target point information extracted from the pre-interruption trajectory may be some or all of the point information in the pre-interruption trajectory, and the at least one target point information extracted from the post-interruption trajectory may also be some or all of the point information in the post-interruption trajectory. Furthermore, the number of target point information extracted from the pre-interruption trajectory may be equal to the number of target point information extracted from the post-interruption trajectory.

[0060] Specifically, in the aforementioned sequence of location information, the target location information can be arranged in ascending order of recording time. When there are multiple target location information extracted from the pre-interruption track, these multiple target location information extracted from the pre-interruption track can be multiple location information with adjacent recording times in the pre-interruption track. When there are multiple target location information extracted from the post-interruption track, these multiple target location information extracted from the post-interruption track can be multiple location information with adjacent recording times in the post-interruption track.

[0061] Furthermore, in specific implementation, in order to improve the accuracy of associating the pre-interruption track and post-interruption track based on the above-mentioned point information sequence, at least one target point information extracted from the pre-interruption track can be at least one point information with the latest recording time in the pre-interruption track, and at least one target point information extracted from the post-interruption track can be at least one point information with the earliest recording time in the post-interruption track.

[0062] In some examples, the location information of a single point in the pre-interruption track can be represented as x. ai The location information of a single point in the track after the interruption can be represented as x. bj Accordingly, the trajectory before the interruption can be represented as X. a =[x a1 x a2 , ..., x am The track after the interruption can be represented as X. b =[x b1 x b2 , ..., x bn Where m and n are both positive integers, and m equals the number of point information included in the track before the interruption, n equals the number of point information included in the track after the interruption, i is a positive integer not greater than m, j is a positive integer not greater than n, and the larger i is, the more point information x is included. ai The later the recording time, the larger j is, and the corresponding point information x bj The later the recording time.

[0063] Specifically, to obtain the aforementioned point information sequence, the interrupted track association device can extract at least one target point information from the pre-interruption track to obtain a first sequence, and extract at least one target point information from the post-interruption track to obtain a second sequence. Then, the first and second sequences are concatenated to obtain the aforementioned point information sequence. The first sequence can be represented as X. aa =[x a(m-k+1) x a2 , ..., x am The second sequence can be represented as X. bb =[x b1 x b2 , ..., x bk Accordingly, the above sequence of location information can be represented as X c =[x a(m-k+1) x a2 , ..., x am x b1 x b2 , ..., x bk], where k is a positive integer, not greater than m and n, and is equal to the number of target point information in the first sequence and the number of target point information in the second sequence.

[0064] S103. Determine the location feature information corresponding to each target location information in the location information sequence.

[0065] Specifically, the aforementioned point feature information can be used to characterize the features of its corresponding target point information itself. In one example, the aforementioned point feature information can be specifically a feature vector; that is, the aforementioned point feature information can be obtained by converting its corresponding target point information into a feature vector.

[0066] In some embodiments, such as Figure 4 As shown, the above S103 can specifically include:

[0067] S1031. Discretize each target point information in the point information sequence to obtain the corresponding discretized point information.

[0068] Specifically, since the track (e.g., the track before and after the interruption mentioned above) is continuous information, the corresponding point information it includes is a continuous variable. The differences between point information recorded at adjacent times in the track are small, which makes it difficult to learn the overall characteristics of the track. Therefore, in order to better learn the overall characteristics of the track, this embodiment discretizes the target point information to transform it from a continuous variable into a discrete variable.

[0069] In some specific embodiments, the target point information may include continuous variable data of longitude and continuous variable data of latitude, and S1031 may specifically include:

[0070] S1-1. Perform degree-minute-second conversion on the continuous variable data of longitude and latitude of each target point in the point location information sequence to obtain the corresponding integer degree part data, fractional degree part data and second degree part data of longitude, and integer degree part data, fractional degree part data and second degree part data of latitude.

[0071] Specifically, the continuous variable data of the longitude of the above-mentioned point can be a longitude value in decimal form, such as 114.282 degrees. The integer part, fractional part, and second part of the longitude of the above-mentioned point can be the integer part, fractional part, and second part of the longitude value in degrees, minutes, and seconds, respectively. Among them, the integer part is equal to the integer part of the longitude value in decimal form, the fractional part is equal to the integer part of the decimal part of the longitude value in decimal form multiplied by 60, and the second part is equal to the decimal part of the longitude value in decimal form multiplied by 60, and then multiplied by 60 again. Taking the longitude value of 114.282 degrees in decimal form as an example, after converting 114.282 degrees to degrees, minutes, and seconds, we get the longitude value of 114 degrees 16 minutes 55.19999999999988 seconds in degrees, minutes, and seconds, respectively.

[0072] S1-2. Round the second portion of the data to obtain the rounded second portion of the data, thus obtaining the discretized point information including the integer portion of the data, the fractional portion of the data, and the rounded second portion of the data.

[0073] Specifically, after obtaining the second-degree longitude and second-degree latitude data corresponding to each target point in the aforementioned point information sequence, the interrupted track association device can round (for example, round down) the second-degree longitude and second-degree latitude data corresponding to each target point in the aforementioned point information sequence to obtain the rounded second-degree longitude and second-degree latitude data. For example, if the second-degree longitude data is 55.19999999999988, rounding it down yields a rounded second-degree data of 55 seconds.

[0074] It should be noted that, compared to traditional interrupted track association schemes based on latitude and longitude in decimal form, which heavily rely on the sampling size of track point information, if the sampling is too small, the differences between point information will be small and the trajectory sequence will be too long, making it impossible to learn the overall characteristics of the track. If the sampling is too large, it will be impossible to achieve real-time track interruption association. In this embodiment, the interrupted track association scheme based on latitude and longitude in degree, minute and second form can increase the differences between target track point information, thereby enabling better deep learning of track features and achieving real-time track interruption association.

[0075] In some specific embodiments, the aforementioned target point information may include continuous variable data of point speed and continuous variable data of point heading. Furthermore, S1031 may specifically include:

[0076] S1-3. Round the continuous variable data of point speed and point heading included in each target point information in the point information sequence to obtain the corresponding rounded point speed data and rounded point heading data, thus obtaining discretized point information including the rounded point speed data and rounded point heading data.

[0077] Specifically, the continuous variable data of the speed at the aforementioned points can be speed values ​​in decimal form, and their range can be greater than or equal to 0 knots and less than or equal to 100 knots. The continuous variable data of the heading at the aforementioned points can be heading values ​​in decimal form, and their range can be greater than 0 degrees and less than or equal to 360 degrees.

[0078] Furthermore, in specific implementation, the rounding of the continuous variable data of point speed and point heading included in each target point information in the point information sequence can specifically include: rounding down the continuous variable data of point speed and point heading included in each target point information in the point information sequence. For example, if the continuous variable data of point speed is 30.55 knots, rounding it down yields a rounded point speed of 30 knots. Similarly, if the continuous variable data of point heading is 100.33 degrees, rounding it down yields a rounded point heading of 100 degrees.

[0079] In this embodiment, in the application scenario where the pre-interruption and post-interruption tracks are for surface targets, each target location information can include continuous variable data of longitude, latitude, speed, and heading; that is, it can include target location information in four dimensions. Correspondingly, the discretized location information corresponding to each target location can include discretized location information in eight dimensions, namely, the integer part of longitude, the fractional part of longitude, the rounded second part of longitude, the integer part of latitude, the fractional part of latitude, the rounded second part of latitude, the rounded speed, and the rounded heading.

[0080] In the application scenario where the flight path before and after the interruption is an aerial target, the information for each target point can include continuous variable data of longitude, latitude, speed, and heading, as well as continuous variable data of altitude. Furthermore, S1031 may also include:

[0081] S1-4. Round the continuous variable data of the point height included in each target point information in the point information sequence to obtain the corresponding rounded point height data, and thus obtain the discretized point information including the rounded point height data.

[0082] Specifically, in the application scenario where the flight path before and after the interruption is the flight path of an aerial target, the discretized point information corresponding to each target point information can include nine dimensions of discretized point information, namely, the integer part of the point longitude, the fractional part of the point longitude, the rounded second part of the point longitude, the integer part of the point latitude, the fractional part of the point latitude, the rounded second part of the point latitude, the rounded point speed data, the rounded point heading data, and the rounded point altitude data.

[0083] In the application scenario where the pre-interruption and post-interruption tracks are underwater targets, the information for each target point can include continuous variable data of longitude, latitude, speed, and heading, as well as continuous variable data of depth. Furthermore, S1031 may also include:

[0084] S1-5. Round the continuous variable data of the point depth included in each target point information in the point information sequence to obtain the corresponding rounded point depth data, and thus obtain the discretized point information including the rounded point depth data.

[0085] Specifically, in the application scenario where the track before and after the interruption is the track of an underwater target, the discretized point information corresponding to each target point can include nine dimensions of discretized point information, namely, the integer part of the point's longitude, the fractional part of the point's longitude, the rounded second part of the point's longitude, the integer part of the point's latitude, the fractional part of the point's latitude, the rounded second part of the point's latitude, the rounded point speed data, the rounded point heading data, and the rounded point depth data.

[0086] In this embodiment, when the discretized point information corresponding to each target point information includes discretized point information in eight dimensions, namely, the integer part of the point longitude, the fractional part of the point longitude, the rounded second part of the point longitude, the integer part of the point latitude, the fractional part of the point latitude, the rounded second part of the point latitude, the rounded point speed data, and the rounded point heading data, the discretized point information corresponding to each target point information can be represented as an array of length 8, and each data in the array corresponds to a discretized point information in one dimension, and different data in the array correspond to discretized point information in different dimensions.

[0087] Accordingly, the above point information sequence can be represented as eight matrices, and each of the eight matrices corresponds to a one-dimensional discretized point information corresponding to each target point information in the above point information sequence. Different matrices in the eight matrices correspond to different dimensions of discretized point information corresponding to each target point information in the above point information sequence.

[0088] S1032. Determine the point feature information corresponding to each target point in the point information sequence based on the discretized point information.

[0089] Specifically, after obtaining the discretized point information corresponding to each target point information in the above point information sequence, the above-mentioned interrupted track association device can encode the discretized point information corresponding to each target point information in the above point information sequence to obtain the corresponding point feature information.

[0090] In one specific embodiment, the above-mentioned encoding process of the discretized point information corresponding to each target point information in the point information sequence to obtain the corresponding point feature information may specifically include:

[0091] For each target point location information, the discrete point location information in each dimension is embedded and encoded to obtain the point location information embedding vector in each dimension. Then, the interrupted track association device can concatenate the point location information embedding vectors in each dimension corresponding to each target point location information to obtain the corresponding feature vector (that is, the point location feature information).

[0092] Specifically, the embedding vector for each dimension of point information can have a preset dimension, which can be 128 or 256 dimensions. Taking a preset dimension of 128 dimensions, and discretized point information including eight dimensions of discretized point information as an example, the embedding vector for each dimension of discretized point information can be a 128-dimensional feature vector, and the point feature information corresponding to each target point information can be a 1024 (128*8)-dimensional feature vector. Embedding is a mapping used to map the discretized point information of each dimension from its original space to a new multi-dimensional space, that is, embedding the original attribute value space into a new space. In one example, the above embedding encoding of the discretized point information of each dimension can specifically include: performing linear embedding encoding on the discretized point information of each dimension.

[0093] Furthermore, it is understood that in this embodiment, by embedding and encoding the discrete point information of each dimension included in the discrete point information and then concatenating them, the discrete point information is converted into a feature vector with a high dimension, thereby increasing the spatial dimension of the discrete point information, which is beneficial to the point features of deep learning tracks.

[0094] In some specific embodiments, when the discretized point information corresponding to each target point includes the integer part of the point's longitude, the fractional part of the point's longitude, the rounded second part of the point's longitude, the integer part of the point's latitude, the fractional part of the point's latitude, and the rounded second part of the point's latitude, in order to distinguish whether the point's longitude is east or west, and whether the point's latitude is north or south, it can be set that: when the point's longitude is east... The integer, fractional, and rounded second portions of the data are positive. When the longitude of the point is west longitude, the integer, fractional, and rounded second portions are negative. When the latitude of the point is north latitude, the integer, fractional, and rounded second portions are positive. When the latitude of the point is south latitude, the integer, fractional, and rounded second portions are negative. Furthermore, since it is difficult to embed and encode negative information, before embedding and encoding the discrete point information of each dimension corresponding to each target point information, the aforementioned interrupted track association device may include:

[0095] For each target point location, the discrete point location information includes the integer part of the longitude, the fractional part of the longitude, the rounded second part of the longitude, the integer part of the latitude, the fractional part of the latitude, and the rounded second part of the latitude. Temporary encoding is performed on each of these components to obtain the corresponding encodings for the integer part of the longitude, the fractional part of the longitude, the rounded second part of the longitude, the integer part of the latitude, the fractional part of the latitude, and the rounded second part of the latitude.

[0096] Specifically, the ranges for the integer part, fractional part, and rounded second part of longitude in the east can be 0 to 180, 0 to 60, and 0 to 60, respectively; the ranges for the integer part, fractional part, and rounded second part of longitude in the west can be -180 to 0, -60 to 0, and -60 to 0, respectively; the ranges for the integer part, fractional part, and rounded second part of latitude in the north can be 0 to 90, 0 to 60, and 0 to 60, respectively; and the ranges for the integer part, fractional part, and rounded second part of latitude in the south can be -90 to 0, -60 to 0, and -60 to 0, respectively.

[0097] Furthermore, in some examples, the integer part of the longitude data can be encoded by adding 180 to the integer part of the corresponding longitude data; the fractional part of the longitude data can be encoded by adding 60 to the fractional part of the corresponding longitude data; and the rounded second part of the longitude data can be encoded by adding 60 to the rounded second part of the corresponding longitude data. Similarly, the integer part of the latitude data can be encoded by adding 90 to the integer part of the corresponding latitude data; the fractional part of the latitude data can be encoded by adding 60 to the fractional part of the corresponding latitude data; and the rounded second part of the latitude data can be encoded by adding 60 to the rounded second part of the corresponding latitude data. This ensures that the integer part of the longitude, the fractional part of the longitude, the rounded second part of the longitude, the integer part of the latitude, the fractional part of the latitude, and the rounded second part of the latitude are all non-negative.

[0098] Accordingly, the discretized point information of each dimension included in the discretized point information corresponding to each target point information is embedded and encoded to obtain the point information embedding vector of each dimension, which may specifically include:

[0099] For each target point location information, the integer part of the longitude, the fractional part of the longitude, the rounded second part of the longitude, the integer part of the latitude, the fractional part of the latitude, and the rounded second part of the latitude are respectively embedded into the following vectors: integer part of longitude, fractional part of longitude, rounded second part of longitude, integer part of latitude, fractional part of latitude, and rounded second part of latitude.

[0100] S104. Based on the point feature information corresponding to each target point in the point information sequence, determine the target feature information related to other target point information in the point information sequence for each target point in the point information sequence.

[0101] Specifically, in the above sequence of location information, the target feature information corresponding to each target location information can be the features of each target location information that are related to other target location information.

[0102] In one specific embodiment, such as Figure 4 As shown, the above S104 can specifically include:

[0103] S1041. Input the point feature information corresponding to each target point in the above point information sequence into the trained attention mechanism model for processing, so as to obtain the target feature information related to other target point information in the point information sequence for each target point in the point information sequence.

[0104] Specifically, the attention mechanism model described above can be used to extract feature information (i.e., the target feature information) related to the point feature information corresponding to other target point information from the point feature information corresponding to each target point information in the above point information sequence. This provides an effective way to select information, improves the efficiency of manual feature extraction, and enables the function of extracting track interruption features under different environments. It can also adapt to the environment.

[0105] In some examples, the attention mechanism model described above can be specifically a Transformer neural network model. The Transformer neural network model is a multi-layered attention mechanism neural network stack technique and is one of the most advanced neural networks currently available. Each layer can be specifically an autoregressive model. In one example, the autoregressive model can be an autoregressive model based on a dot product multi-head self-attention mechanism, as shown in equation (1).

[0106]

[0107] Where Q, K, and V are linear projections of the input sequence (the output of the first layer or the output sequence of the previous layer used for other layers). At each layer, the input sequence is projected onto a new space V, and the output of the block of interest is a weighted sum in V, where the weights indicate the relative contribution of each target point in the aforementioned point information sequence. These weights are calculated as follows: During the training phase, the projection operators Q, K, and V are learned and computations are performed in parallel. Therefore, compared to recurrent neural network models (such as LSTM neural network models), which must process data sequentially and may be unable to extract information from long sequences, the Transformer neural network model has a more powerful learning ability, and its ability to learn sequences is significantly stronger than that of recurrent neural network models.

[0108] S105. Based on the target feature information corresponding to each target point in the point information sequence, associate the pre-interruption trajectory and the post-interruption trajectory.

[0109] In some embodiments, S105 may specifically include:

[0110] S1051. Input the target feature information corresponding to each target point in the point information sequence into the trained classification model to obtain the classification result. The classification result is used to indicate whether there is a correlation between the track before the interruption and the track after the interruption.

[0111] Specifically, the classification model can be used to determine whether there is a correlation between two tracks (i.e., the track before the interruption and the track after the interruption mentioned above).

[0112] In some embodiments, such as Figure 5As shown, the classification model described above can include a fully connected layer and a binary classifier. The fully connected layer is connected to the attention mechanism model and is used to convert the output of the attention mechanism model (i.e., the target feature information corresponding to each target point in the point information sequence) into a two-dimensional matrix. In one example, the fully connected layer can be a linear matrix transformation operator. The binary classifier is connected to the fully connected layer and is used to determine whether there is a correlation between the pre-interruption trajectory and the post-interruption trajectory based on the output of the fully connected layer (i.e., the two-dimensional matrix), and outputs the classification result. In one example, the binary classifier can use the softmax function and can select the position index with the highest probability as the classification result.

[0113] In one example, the classification result can be 0 or 1. When the classification result is 1, it indicates that there is a correlation between the pre-interruption track and the post-interruption track. When the classification result is 0, it indicates that there is no correlation between the pre-interruption track and the post-interruption track.

[0114] Furthermore, in practice, during the training process of the aforementioned binary classifier, cross-entropy can be used as the loss function for training.

[0115] S1052. When the classification result indicates that the pre-interruption track and the post-interruption track are related, the pre-interruption track and the post-interruption track are associated.

[0116] Specifically, when the classification result indicates that the pre-interruption track and the post-interruption track are related, the pre-interruption track and the post-interruption track can belong to the same track, and the interruption track association device can renumber the pre-interruption track and / or the post-interruption track so that the pre-interruption track and the post-interruption track have the same track number, thereby realizing the association of the pre-interruption track and the post-interruption track.

[0117] Furthermore, when the above classification results indicate that the pre-interruption track and the post-interruption track are not related, it can be considered that the pre-interruption track and the post-interruption track do not belong to the same track.

[0118] As can be seen from the above, the interrupted track association method provided in this embodiment obtains the pre-interruption track and the post-interruption track, then extracts at least one target point information from the pre-interruption track and at least one target point information from the post-interruption track to obtain a point information sequence including multiple target point information. Next, it determines the point feature information corresponding to each target point information in the point information sequence, and based on the point feature information corresponding to each target point information in the point information sequence, determines the target feature information related to other target point information in the point information sequence for each target point information. Then, based on the target feature information corresponding to each target point information in the point information sequence, it associates the pre-interruption track and the post-interruption track. Therefore, when associating interrupted tracks, it can extract the association features of the pre-interruption track and the post-interruption track, and perform interrupted track association based on the extracted association features. This eliminates the need for complex calculations of the track data and allows for environmental adaptation, thus improving the accuracy of associating interrupted tracks.

[0119] Based on the methods described in the above embodiments, this embodiment will further describe the process from the perspective of interrupted track association devices. Please refer to [link to relevant documentation]. Figure 6 , Figure 6 This application provides a detailed description of an interrupted flight path association device, which includes: an acquisition module 301, an extraction module 302, a first determination module 303, a second determination module 304, and an association module 305, wherein:

[0120] (1) Obtain module 301

[0121] The acquisition module 301 is used to acquire the track before the interruption and the track after the interruption.

[0122] (2) Extraction module 302

[0123] Extraction module 302 is used to extract at least one target point information from the pre-interruption track and at least one target point information from the post-interruption track to obtain a point information sequence including multiple target point information.

[0124] (3) First determination module 303

[0125] The first determining module 303 is used to determine the point feature information corresponding to each target point in the point information sequence.

[0126] In some embodiments, the first determining module 303 may specifically include:

[0127] The first processing unit is used to discretize each target point information in the point information sequence to obtain the corresponding discretized point information.

[0128] The determining unit is used to determine the point feature information corresponding to each target point in the point information sequence based on the discretized point information.

[0129] In some specific embodiments, the determining unit described above can be specifically used for:

[0130] The discretized point information corresponding to each target point in the point information sequence is encoded to obtain the corresponding point feature information.

[0131] In some specific embodiments, the target point information may include continuous variable data of longitude and continuous variable data of latitude, and the first processing unit may be specifically used for:

[0132] For each target point in the point information sequence, the continuous variable data of longitude and latitude are converted into degrees, minutes and seconds to obtain the corresponding integer part data, fractional part data and second part data of longitude and latitude.

[0133] The second portion of the data is rounded down to obtain the rounded second portion of the data, resulting in discretized point information that includes the integer portion of the data, the fractional portion of the data, and the rounded second portion of the data.

[0134] In other specific embodiments, the target point information may further include continuous variable data of point speed and continuous variable data of point heading, and the first processing unit may also be specifically used for:

[0135] The continuous variable data of point speed and continuous variable data of point heading included in each target point information sequence are rounded to obtain the corresponding rounded point speed data and rounded point heading data, thus obtaining discretized point information including the rounded point speed data and rounded point heading data.

[0136] (4) Second determination module 304

[0137] The second determining module 304 is used to determine the target feature information related to other target point information in the point information sequence for each target point information in the point information sequence, based on the point feature information corresponding to each target point information in the point information sequence.

[0138] In some embodiments, the second determining module 304 described above may be specifically used for:

[0139] The point feature information corresponding to each target point in the point information sequence is input into the trained attention mechanism model for processing, so as to obtain the target feature information related to other target points in the point information sequence for each target point in the point information sequence.

[0140] (5) Related module 305

[0141] The association module 305 is used to associate the pre-interruption trajectory and the post-interruption trajectory based on the target feature information corresponding to each target point in the point information sequence.

[0142] In some embodiments, the aforementioned association module 305 may specifically include:

[0143] The input unit is used to input the target feature information corresponding to each target point in the point information sequence into the trained classification model to obtain the classification result. The classification result is used to indicate whether there is a correlation between the track before the interruption and the track after the interruption.

[0144] The association unit is used to associate the pre-interruption track and the post-interruption track when the classification result indicates that there is an association between them.

[0145] In practice, each of the above units and modules can be implemented as an independent entity or can be arbitrarily combined to be implemented as the same or several entities. For the specific implementation of each of the above units and modules, please refer to the previous method implementation examples, which will not be repeated here.

[0146] As can be seen from the above, the interrupted track association device provided in this embodiment includes an acquisition module for acquiring the track before interruption and the track after interruption; an extraction module for extracting at least one target point information from the track before interruption and extracting at least one target point information from the track after interruption, so as to obtain a point information sequence including multiple target point information; a first determination module for determining the point feature information corresponding to each target point information in the point information sequence; and a second determination module for determining each target point information in the point information sequence according to the point feature information corresponding to each target point information in the point information sequence. The target location information corresponds to the target feature information related to other target location information in the location information sequence; the association module is used to associate the pre-interruption track and the post-interruption track according to the target feature information corresponding to each target location information in the location information sequence. Thus, when associating interrupted tracks, it can extract the association features of the pre-interruption track and the post-interruption track, and perform interrupted track association based on the extracted association features of the pre-interruption track and the post-interruption track. There is no need to perform complex calculations on the track data, and it can adapt to the environment, thus improving the accuracy of associating interrupted tracks.

[0147] Accordingly, embodiments of this application also provide an electronic device, wherein the electronic device may be a terminal or a server, such as... Figure 7 As shown, it illustrates a structural schematic diagram of the electronic device involved in the embodiments of this application, specifically:

[0148] The electronic device 400 includes a processor 401 with one or more processing cores, a memory 402 with one or more computer-readable storage media, and a computer program stored in the memory 402 and executable on the processor. The processor 401 and the memory 402 are electrically connected. Those skilled in the art will understand that the electronic device structure shown in the figures does not constitute a limitation on the electronic device, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0149] The processor 401 is the control center of the electronic device 400. It connects various parts of the electronic device 400 through various interfaces and lines. By running or loading software programs and / or modules stored in the memory 402, and calling data stored in the memory 402, it performs various functions of the electronic device 400 and processes data, thereby monitoring the electronic device 400 as a whole.

[0150] In this embodiment, the processor 401 in the electronic device 400 loads the instructions corresponding to the processes of one or more applications into the memory 402 according to the following steps, and the processor 401 runs the applications stored in the memory 402 to realize various functions:

[0151] Obtain the track before and after the interruption;

[0152] Extract at least one target point location information from the track before the interruption and at least one target point location information from the track after the interruption to obtain a point location information sequence including multiple target point location information;

[0153] Determine the location feature information corresponding to each target location in the location information sequence;

[0154] Based on the location feature information corresponding to each target location information in the location information sequence, determine the target feature information related to other target location information in the location information sequence for each target location information in the location information sequence.

[0155] Based on the target feature information corresponding to each target point in the point information sequence, the pre-interruption trajectory and the post-interruption trajectory are correlated.

[0156] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0157] Optional, such as Figure 7 As shown, the electronic device 400 also includes: a touch display screen 403, a radio frequency circuit 404, an audio circuit 405, an input unit 406, and a power supply 407. The processor 401 is electrically connected to the touch display screen 403, the radio frequency circuit 404, the audio circuit 405, the input unit 406, and the power supply 407. Those skilled in the art will understand that... Figure 7 The electronic device structure shown does not constitute a limitation on the electronic device and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0158] The touch display screen 403 can be used to display a graphical user interface (GUI) and receive operation commands generated by the user interacting with the GUI. The touch display screen 403 may include a display panel and a touch panel. The display panel can be used to display information input by the user or information provided to the user, as well as various graphical user interfaces of the electronic device. These graphical user interfaces can be composed of graphics, text, icons, video, and any combination thereof. Optionally, the display panel can be configured using a liquid crystal display (LCD), organic light-emitting diode (OLED), or other similar technologies. The touch panel can be used to collect touch operations performed by the user on or near it (such as operations performed by the user using a finger, stylus, or any suitable object or accessory on or near the touch panel), generate corresponding operation commands, and execute the corresponding program according to the operation commands. Optionally, the touch panel may include two parts: a touch detection device and a touch controller. The touch detection device detects the user's touch location and the signal generated by the touch operation, transmitting the signal to the touch controller. The touch controller receives touch information from the touch detection device, converts it into touch point coordinates, and sends it to the processor 401. It can also receive and execute commands from the processor 401. The touch panel can cover the display panel. When the touch panel detects a touch operation on or near it, it transmits the information to the processor 401 to determine the type of touch event. Subsequently, the processor 401 provides corresponding visual output on the display panel based on the type of touch event. In this embodiment, the touch panel and the display panel can be integrated into the touch display screen 403 to achieve input and output functions. However, in some embodiments, the touch panel and the touch display screen 403 can be implemented as two independent components to achieve input and output functions. That is, the touch display screen 403 can also be used as part of the input unit 406 to achieve input functions.

[0159] In this embodiment of the application, the processor 401 executes a game application to generate a virtual three-dimensional scene on the touch screen 403. The scene includes a graphical user interface (UI) and a second spatial orientation indicator. The second spatial orientation indicator displays a spatial orientation mark corresponding to the target object. The spatial orientation mark is used to indicate the location of the target object.

[0160] The touch display screen 403 can be used to present images of virtual three-dimensional scenes, as well as graphical user interfaces and receive operation commands generated by the user interacting with the graphical user interface.

[0161] The radio frequency circuit 404 can be used to transmit and receive radio frequency signals to establish wireless communication with network devices or other electronic devices, and to transmit and receive signals with network devices or other electronic devices.

[0162] Audio circuitry 405 can be used to provide an audio interface between a user and electronic device 400 via a speaker and microphone. Audio circuitry 405 can convert received audio data into electrical signals and transmit them to the speaker, where the speaker converts them into sound signals for output. Conversely, the microphone converts collected sound signals into electrical signals, which are then received by audio circuitry 405, converted back into audio data, and then processed by processor 401 before being transmitted via radio frequency circuitry 404 to, for example, another electronic device, or output to memory 402 for further processing. Audio circuitry 405 may also include an earphone jack to provide communication between peripheral headphones and electronic device 400.

[0163] The input unit 406 can be used to receive input numbers, characters, or user characteristic information (such as fingerprints, iris, facial information, etc.), and to generate keyboard, mouse, joystick, optical, or trackball signal inputs related to user settings and function control.

[0164] Power supply 407 is used to supply power to various components of electronic device 400. Optionally, power supply 407 can be logically connected to processor 401 through a power management system, thereby enabling functions such as charging, discharging, and power consumption management through the power management system. Power supply 407 may also include one or more DC or AC power supplies, recharging systems, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components.

[0165] although Figure 7 As not shown in the diagram, the electronic device 400 may also include a camera, sensor, wireless fidelity module, Bluetooth module, etc., which will not be described in detail here.

[0166] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0167] As can be seen from the above, the electronic device provided in this embodiment can improve the accuracy of interrupted track association.

[0168] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by instructions, or by instructions controlling related hardware. These instructions can be stored in a computer-readable storage medium and loaded and executed by a processor.

[0169] Therefore, embodiments of this application provide a computer-readable storage medium storing multiple computer programs that can be loaded by a processor to execute steps in any of the interrupted track association methods provided in embodiments of this application. For example, the computer program can execute the following steps:

[0170] Obtain the track before and after the interruption;

[0171] Extract at least one target point location information from the track before the interruption and at least one target point location information from the track after the interruption to obtain a point location information sequence including multiple target point location information;

[0172] Determine the location feature information corresponding to each target location in the location information sequence;

[0173] Based on the location feature information corresponding to each target location information in the location information sequence, determine the target feature information related to other target location information in the location information sequence for each target location information in the location information sequence.

[0174] Based on the target feature information corresponding to each target point in the point information sequence, the pre-interruption trajectory and the post-interruption trajectory are correlated.

[0175] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0176] The storage medium may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0177] Since the computer program stored in the storage medium can execute the steps in any of the interrupted track association methods provided in the embodiments of this application, the beneficial effects that any of the interrupted track association methods provided in the embodiments of this application can achieve can be realized. For details, please refer to the previous embodiments, which will not be repeated here.

[0178] The foregoing has provided a detailed description of the interrupted track association method, apparatus, storage medium, and electronic device provided in the embodiments of this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for associating interrupted flight paths, characterized in that, include: Obtain the track before and after the interruption; At least one target point information is extracted from the pre-interruption trajectory, and at least one target point information is extracted from the post-interruption trajectory to obtain a point information sequence including multiple target point information. Determine the point feature information corresponding to each target point in the point information sequence; Based on the point feature information corresponding to each of the target point information in the point information sequence, determine the target feature information related to other target point information in the point information sequence for each of the target point information in the point information sequence; wherein, the step of determining the target feature information related to other target point information in the point information sequence for each of the target point information in the point information sequence based on the point feature information corresponding to each of the target point information in the point information sequence includes: inputting the point feature information corresponding to each of the target point information in the point information sequence into a trained attention mechanism model for processing, so as to obtain the target feature information related to other target point information in the point information sequence for each of the target point information in the point information sequence; The pre-interruption trajectory and the post-interruption trajectory are associated based on the target feature information corresponding to each target point in the point information sequence; wherein, the step of associating the pre-interruption trajectory and the post-interruption trajectory based on the target feature information corresponding to each target point in the point information sequence includes: inputting the target feature information corresponding to each target point in the point information sequence into a trained classification model to obtain a classification result, the classification result being used to indicate whether the pre-interruption trajectory and the post-interruption trajectory are associated; when the classification result indicates that the pre-interruption trajectory and the post-interruption trajectory are associated, the pre-interruption trajectory and the post-interruption trajectory are associated.

2. The method for associating interrupted flight paths according to claim 1, characterized in that, The step of determining the point feature information corresponding to each target point in the point information sequence includes: Each target point in the point information sequence is discretized to obtain the corresponding discretized point information; Based on the discretized point information, determine the point feature information corresponding to each target point in the point information sequence.

3. The method for associating interrupted flight paths according to claim 2, characterized in that, The target point information includes continuous variable data of longitude and continuous variable data of latitude; the step of discretizing each target point information in the point information sequence to obtain the corresponding discretized point information includes: The continuous variable data of the longitude and the continuous variable data of the latitude of each target point in the point information sequence are converted into degrees, minutes and seconds to obtain the corresponding integer degree part data, fractional degree part data and second degree part data of the longitude and the integer degree part data, fractional degree part data and second degree part data of the latitude. The second portion of the data is rounded down to obtain the rounded second portion of the data, thus obtaining discrete point information including the integer portion of the data, the fractional portion of the data, and the rounded second portion of the data.

4. The method for associating interrupted flight paths according to claim 2, characterized in that, The target point information includes continuous variable data of point speed and continuous variable data of point heading; the step of discretizing each target point information in the point information sequence to obtain the corresponding discretized point information includes: The continuous variable data of the point speed and the continuous variable data of the point heading included in each of the target point information in the point information sequence are rounded to obtain the corresponding rounded point speed data and rounded point heading data, thereby obtaining discretized point information including the rounded point speed data and the rounded point heading data.

5. The method for associating interrupted flight paths according to claim 2, characterized in that, The step of determining the point feature information corresponding to each target point in the point information sequence based on the discretized point information includes: The discretized point information corresponding to each target point information in the point information sequence is encoded to obtain the corresponding point feature information.

6. A device for linking interrupted flight paths, characterized in that, include: The acquisition module is used to acquire the track before and after the interruption; An extraction module is used to extract at least one target point information from the pre-interruption trajectory and at least one target point information from the post-interruption trajectory, so as to obtain a point information sequence including multiple target point information. The first determining module is used to determine the point feature information corresponding to each target point in the point information sequence; The second determining module is used to determine, based on the point feature information corresponding to each of the target point information in the point information sequence, target feature information related to other target point information in the point information sequence; wherein, the second determining module is specifically used to: input the point feature information corresponding to each of the target point information in the point information sequence into a trained attention mechanism model for processing, so as to obtain the target feature information related to other target point information in the point information sequence for each of the target point information in the point information sequence; An association module is used to associate the pre-interruption trajectory and the post-interruption trajectory based on the target feature information corresponding to each target point in the point information sequence. The association module includes: an input unit, used to input the target feature information corresponding to each target point in the point information sequence into a trained classification model to obtain a classification result, the classification result indicating whether there is an association between the pre-interruption trajectory and the post-interruption trajectory; and an association unit, used to associate the pre-interruption trajectory and the post-interruption trajectory when the classification result indicates that there is an association between them.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program adapted for loading by a processor to perform the steps in the interrupted track association method as described in any one of claims 1-5.

8. An electronic device, characterized in that, The electronic device includes a memory and a processor. The memory stores a computer program, and the processor executes the steps in the interrupted track association method as described in any one of claims 1-5 by calling the computer program stored in the memory.

Images