Multi-target association management method and device, electronic equipment and storage medium
By constructing an associative storage network based on graph-edge data structure to associate measurements and trajectories, the problem of low storage and access efficiency in multi-sensor fusion is solved, the efficient and accurate multi-target association management is achieved, and the accuracy of target recognition is improved.
Patent Information
- Application Number
- CN202510685962.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-27
- Publication Date
- 2025-09-26
AI Technical Summary
In existing multi-sensor fusion technology, measurement data association results are stored in a single-point form, resulting in low storage and access efficiency and poor accuracy. This limits the support capability of multi-measurement data association targets and is prone to target omission and false detection.
Through an associative storage network based on graph-edge data structure, we can associate measurements and trajectories, establish an association relationship between targets and trajectories, use Mahalanobis distance and gate type to filter target information, and build a multi-target association management method to achieve fast and accurate target management.
It improves the efficiency and accuracy of multi-target association management, ensures the accuracy of target identification, reduces missed detections and false detections, and supports the efficient fusion of multi-sensor data.
Smart Images
Figure CN120705357A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of multi-target association technology, and in particular to a multi-target association management method, device, electronic device and storage medium. Background Art
[0002] Multi-sensor fusion, as an indispensable technology in intelligent driving perception systems, can integrate the advantages of different sensors, solve the problem of time synchronization between different sensors, and effectively overcome various problems caused by insufficient performance of upstream sensor models, such as false detection, missed detection, and accuracy deviation.
[0003] In this process, the multi-sensor association module forms the core foundation of fusion technology. As a prerequisite for subsequent target identification, efficient and reasonable data storage methods for multi-sensor measurement data association are crucial to ensuring the accuracy of the fusion output results. However, under the existing technical framework, measurement data association is usually performed in a one-to-one manner, and the association results are stored in a single point form. This data structure cannot meet the storage and access requirements of multi-measurement data association results. When storing multi-measurement and multi-target data, the degree of association is low. As a result, when accessing the multi-measurement data of a specific target, the low degree of association leads to low reading efficiency and poor data accuracy. This limits the ability to support the association of targets with different measurement data, which in turn leads to problems such as missed and false detection of targets. Summary of the Invention
[0004] The problem solved by the present invention is how to improve the management efficiency and accuracy of multi-target association results.
[0005] To solve the above problems, the present invention provides a multi-target association management method, comprising: Based on the information source, the obtained target information is correlated between measurements to obtain the correlation results between measurements; constructing an association storage network according to the inter-measurement association results, wherein the association storage network is composed of a graph edge data structure; Establishing a trajectory for the target information of the same information source, and constructing a trajectory association result based on the trajectory and the target information; creating a target based on the association storage network and the association results between the measurements, and associating the target with a corresponding trajectory; updating the trajectory of the target based on the trajectory association result to obtain an updated result; A management strategy for the target is determined according to the update result.
[0006] Optionally, establishing a track for the target information of the same information source, and constructing a track association result based on the track and the target information includes: determining a gate type according to a measurement attribute of the target information; determining a first associated gate for the trajectory according to the gate type; The target information is filtered according to the first correlation gate to obtain the trajectory correlation result.
[0007] Optionally, filtering the target information according to the first correlation gate to obtain the trajectory correlation result includes: calculating a Mahalanobis distance between associated target information and the trajectory, wherein the associated target information includes the target information satisfying the first associated wave gate; The associated target information with the shortest Mahalanobis distance is used as the trajectory association result of the trajectory.
[0008] Optionally, performing inter-measurement correlation on the acquired target information based on the information source to obtain the inter-measurement correlation result includes: constructing a target information pair from the two target information that meet the association requirements, wherein the association requirements include: the target information having different measurement attributes, and the two target information having an association relationship; determining a gate type according to the measurement attributes of the target information pair, and determining a second associated gate according to the gate type; The target information pairs are screened according to the second correlation gate to obtain the inter-measurement correlation result.
[0009] Optionally, screening the target information pair according to the second correlation gate to obtain the inter-measurement correlation result includes: determining information priorities of the two target information according to the information sources; determining a screening priority of the second associated wave gate according to the information priority; screening the target information pairs that meet the second association wave gate as associated target information pairs; Calculating the Mahalanobis distance between two associated target information in the associated target information pair; According to the screening priority, the associated target information pair with the shortest Mahalanobis distance and the highest priority is used as the inter-measurement association result.
[0010] Optionally, constructing an associated storage network according to the inter-measurement association results includes: Determining the order of association of the target information according to information priority; Based on the inter-measurement association results and the association order, the associated measurement relationships corresponding to different information sources are stored to obtain the associated storage network.
[0011] Optionally, updating the trajectory of the target based on the trajectory association result to obtain the update result includes: determining the target information related to the trajectory of the target according to the trajectory association result; Determining a target information combination associated with the target information based on the inter-measurement association result; The target information combination is processed by a preset filtering algorithm, the state of the target is estimated, and an updated result is obtained.
[0012] Optionally, the determining the target management strategy according to the update result includes: When the trajectory corresponding to the target does not have the update result for more than a preset time, the target is deleted; When the update result satisfies a preset release condition, the target corresponding to the update result is released.
[0013] In a second aspect, the present invention further provides a multi-target association management device, comprising: The inter-measurement correlation module is used to perform inter-measurement correlation on the acquired target information based on the information source and obtain the inter-measurement correlation result; A storage module, configured to construct an association storage network based on the inter-measurement association results, wherein the association storage network is composed of a graph edge data structure; a trajectory measurement association module, configured to establish a trajectory for the target information of the same information source, and construct a trajectory association result based on the trajectory and the target information; a creation module, configured to create a target based on the association storage network and the association result between the measurements, and associate the target with a corresponding trajectory; An updating module, configured to update the trajectory of the target based on the trajectory association result to obtain an updated result; A management module is used to determine a management strategy for the target according to the update result.
[0014] In a third aspect, the present invention further provides an electronic device comprising a memory and a processor; The memory is used to store computer programs; The processor is configured to implement the multi-target association management method described above when executing the computer program.
[0015] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the multi-target association management method as described above is implemented.
[0016] The beneficial effects of the multi-objective association management method of the present invention are: By correlating measurements using acquired target information from different information sources (e.g., sensors), the correlation between measurements, i.e., the inter-measurement correlation results, is determined. Based on these inter-measurement correlation results, an associative storage network based on a graph-edge data structure is established, enabling rapid access to associated target information from the storage space when establishing a target. Based on the information source, the target information is correlated with existing trajectories to obtain trajectory correlation results, which are used to update the target's multiple measurement attributes and improve the accuracy of subsequent target fusion. After creating a target based on the inter-measurement correlation results, the target's status is updated based on the trajectory correlation results to obtain an updated result. This updated result is then used to determine the target's subsequent management strategy, enabling rapid and accurate target association management that integrates multiple measurement attributes from multiple sensors. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] Figure 1 Schematic diagram of the process of a multi-target association management method according to an embodiment of the present invention; Figure 2 This is an example diagram of an association storage structure of a multi-target association management method according to an embodiment of the present invention; Figure 3 This is an example diagram of a graph edge structure of a multi-objective association management method according to an embodiment of the present invention; Figure 4 2 is an exemplary diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION
[0018] To make the above-mentioned objects, features, and advantages of the present invention more clearly understood, specific embodiments of the present invention are described in detail below with reference to the accompanying drawings. Although certain embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as being limited to the embodiments described herein. Instead, these embodiments are provided to provide a more thorough and complete understanding of the present invention. It should be understood that the drawings and embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of protection of the present invention.
[0019] It should be understood that the various steps described in the method embodiments of the present invention may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.
[0020] The term "including" and its variations used in this document are open inclusions, that is, "including but not limited to"; the term "based on" means "based at least in part on"; the term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one other embodiment"; the term "some embodiments" means "at least some embodiments"; the term "optionally" means "optional embodiments". The relevant definitions of other terms will be given in the following description. It should be noted that the concepts of "first", "second", etc. mentioned in the present invention are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.
[0021] It should be noted that the modifications of "one" and "multiple" mentioned in the present invention are illustrative rather than restrictive. Those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".
[0022] The names of the messages or information exchanged between multiple devices in the embodiments of the present invention are only used for illustrative purposes and are not used to limit the scope of these messages or information.
[0023] In response to the problems existing in the above-mentioned related technologies, this embodiment provides a multi-target association management method, device, electronic device and storage medium.
[0024] like Figure 1 As shown, an embodiment of the present invention provides a multi-target association management method, including: Step S100 : performing inter-measurement correlation on the acquired target information based on the information source to obtain an inter-measurement correlation result.
[0025] Measurement-Measurement correlation (MM correlation) is defined as correlation between different measurement targets. In the field of intelligent driving perception, sensors are used to measure targets and obtain target information, which can be used to control the vehicle for intelligent driving. Because target information obtained by a single type of sensor is limited in type, using multiple sensors for target measurement has become a trend in the field of intelligent driving perception. Since different sensors can obtain target information with different measurement attributes, they can complement the shortcomings of single-type sensors and effectively address issues such as false detection, missed detection, and inaccurate accuracy caused by insufficient sensor model performance. Fusion of target information between sensors forms the basis for target tracking and correlation.
[0026] In one embodiment of the present invention, data is acquired through various types of sensors. Since different types of sensors acquire different types of data, in order to rationally manage target information, the data is pre-processed, converted into data of a preset type, and stored. For example, the data acquired by the sensor is processed, converted into a SensorFrame, and converted into target information, wherein the target information includes time, target number, position, speed, and size. The target number is used to represent the number of the target actually detected by the sensor, and the position, speed, and size respectively represent the position, speed, and size of the target in the image at the time of detection. In this embodiment, target information with the same measurement attributes can be obtained by different sensors. For example, the position can be obtained by a laser radar (LiDAR), a radar (Radar), a vision sensor (Vision), or a camera sensor (Camera).
[0027] Step S200 : constructing an association storage network according to the inter-measurement association results, wherein the association storage network is composed of a graph edge data structure.
[0028] Specifically, because target information can be of various types, and different types of target information can indicate the same target, when using sensors to measure targets, the acquired target information can be subject to storage confusion and misdetection due to target confusion at multiple times. In embodiments of the present invention, target information from different information sources is correlated using inter-measurement correlation results, and an associative storage network based on a graph edge data structure is constructed using correlation relationships as storage logic. For example, target information from lidar and visual sensors can be correlated, and target information from lidar and radar can be correlated. Target information from different information sources that indicates the same target is then combined into associated measurements and stored in the same association network. For example, after obtaining associated measurements from each information source, associated measurements are formed between the first lidar (LiDAR1) and the first radar (Radar1), and associated measurements between the first radar (Radar1) and the first camera sensor (Camera1). In this embodiment, the preset priority of different information sources is: LiDAR > Radar > Camera sensor. Based on the above association relationship, the association storage network can be constructed as: LiDAR1-Radar1-Camera1. The target information about the target is stored in this order.
[0029] Optionally, the information source includes at least one of a laser radar (LiDAR), a radar (Radar), a vision sensor (Vision), and a camera sensor (Camera).
[0030] Step S300 : establishing a track for the target information of the same information source, and constructing a track association result based on the track and the target information.
[0031] In one embodiment, track-measurement association is defined as Track-Measurement Association (TM Association), which is used to perform multi-target association between tracks (tracks) and sensor measurements. Association is performed based on the target track initially determined based on target information and the newly received sensor-measured target information. Since target information may come from multiple sensors, track-measurement association rules are determined based on the measurement attributes of the target information to obtain a track association result between the track and target information. For example, if the measurement attribute of the target information is from a camera sensor and the attribute is the image, a 2D gate is established using the similarity of the target in the image as one of the rules. This gate is used to calculate the outline similarity between the target in the track and the target in the new target information. For example, the similarity between the two is calculated by calculating the Intersection over Union (IoU) between the two. Target information with an IoU greater than a preset threshold is considered to meet the 2D gate and is associated with the track as the track association result.
[0032] Step S400: creating a target based on the associated storage network and the inter-measurement association result, and associating the target with a corresponding trajectory.
[0033] According to the associative storage network, the associated target information can be quickly read and the target can be created based on the target information. For example, the position, speed, size and other information related to the target can be read from the associative storage network, and the target information that can characterize the position, speed and size from different sensors can be determined based on the correlation results between measurements, thereby creating the target.
[0034] Step S500: updating the trajectory of the target based on the trajectory association result to obtain an updated result.
[0035] In one embodiment, after acquiring new target information, the new target information is associated with the existing trajectory to obtain a trajectory association result, and the latest state of the target is updated according to the trajectory association result as an update result.
[0036] Step S600: Determine the management policy of the target according to the update result.
[0037] Specifically, after the target status is updated, the management strategy of the target can be determined based on the update result. For example, when the target has not been updated for a long time, it means that the target has disappeared and the target is deleted; when the target meets the preset requirements, such as the distance is too close, the target is released and the corresponding intelligent driving strategy is determined based on the target distance.
[0038] In this embodiment, measurements are correlated using acquired target information from different sensors and different measurement attributes. The inter-measurement correlation relationships, i.e., inter-measurement correlation results, are determined one-to-one. Correlated measurements are then determined based on the inter-measurement correlation results. An associated storage network is then established based on the associated measurements, enabling rapid access to associated target information from the storage space when establishing a target. Association conditions are established using measurement attributes, and target information is associated with existing trajectories to update the target's multiple measurement attributes and improve the accuracy of subsequent target fusion. After creating a target based on the inter-measurement correlation results, the target's state is updated based on the trajectory correlation results to obtain an updated result. The target's subsequent management strategy is then determined based on the updated result, enabling rapid and accurate target association management that integrates multiple measurement attributes from multiple sensors.
[0039] Optionally, establishing a track for the target information of the same information source, and constructing a track association result based on the track and the target information includes: The gate type is determined according to the measurement attribute of the target information.
[0040] A first associated gate for the track is determined according to the gate type.
[0041] The target information is filtered according to the first correlation gate to obtain the trajectory correlation result.
[0042] Optionally, filtering the target information according to the first correlation gate to obtain the trajectory correlation result includes: calculating a Mahalanobis distance between associated target information and the trajectory, wherein the associated target information includes the target information satisfying the first associated wave gate; The associated target information with the shortest Mahalanobis distance is used as the trajectory association result of the trajectory.
[0043] In one embodiment, a target trajectory is established based on measured target information. After acquiring new target information, a corresponding association subclass is constructed based on the target information's measured attributes and information source to determine the corresponding gate type and a first association gate. For example, the information source may include target information from a lidar, radar, visual sensor, or camera sensor, and the measured attributes may include speed, position, orientation, and profile. When the target information is speed, a speed gate is established based on the speed and the existing trajectory, using a first speed range as the gate threshold. When the speed meets the first speed range, the target information meets the speed gate requirements and is designated as associated target information. Preliminary screening of target information using the first association gate improves association efficiency.
[0044] After filtering and obtaining the associated target information, the Mahalanobis distance between the associated target information and the trajectory is calculated. The measurement with the shortest Mahalanobis distance is used as the trajectory association result for the trajectory and used to update the target state. That is, the associated target information with the shortest Mahalanobis distance is selected as the trajectory association result for the trajectory.
[0045] In another embodiment, since the target information obtained by sensor measurements is continuous over time, by determining whether the target information in two adjacent frames represents the same target, the association weight of the target information during trajectory measurement association is increased, thereby improving the accuracy of target association. For example, when using Kalman filtering for trajectory measurement association, more sophisticated processing is sometimes required for measurement targets that meet specific conditions. In particular, in this embodiment, when there are multiple sensors or multiple similar targets, a simple Mahalanobis distance calculation may not be sufficient to distinguish different targets. In this case, "IDMatch" (identity matching) is used to further refine the Mahalanobis distance calculation to improve the accuracy of data association. For example, a comparison of target outlines is added to increase the accuracy of trajectory association.
[0046] Optionally, the first associated gates include a transverse distance gate, a longitudinal distance gate, a speed gate, a type gate, a heading gate, and a 2D gate. The transverse distance gate is used to select the transverse distance (y-distance) to the sensor, the longitudinal distance gate is used to select the longitudinal distance (x-distance) to the sensor, the speed gate is used to select the target's speed, the type gate is used to select the target's type (e.g., person, vehicle, fixed object), the heading gate is used to select the target's heading (e.g., in-line, perpendicular, reverse), and the 2D gate is used to select the target's outline.
[0047] Optionally, performing inter-measurement correlation on the acquired target information based on the information source to obtain the inter-measurement correlation result includes: constructing a target information pair from the two target information that meet the association requirements, wherein the association requirements include: the target information having different measurement attributes, and the two target information having an association relationship; determining a gate type according to the measurement attributes of the target information pair, and determining a second associated gate according to the gate type; The target information pairs are screened according to the second correlation gate to obtain the inter-measurement correlation result.
[0048] In one embodiment, inter-measurement correlation is used to associate and store two types of target information from different sources and with different measurement attributes, enabling rapid access to the associated data when needed and improving efficiency. Two types of target information with related and different measurement attributes are constructed into target information pairs. A second correlation gate is used to filter these pairs, and only those that meet the criteria are stored in an associative manner. This inter-measurement correlation result satisfies the storage and access requirements for multi-measurement data correlation results.
[0049] Optionally, screening the target information pair according to the second correlation gate to obtain the inter-measurement correlation result includes: determining information priorities of the two target information according to the information sources; determining a screening priority of the second associated wave gate according to the information priority; screening the target information pairs that meet the second association wave gate as associated target information pairs; Calculating the Mahalanobis distance between two associated target information in the associated target information pair; According to the screening priority, the associated target information pair with the shortest Mahalanobis distance and the highest priority is used as the inter-measurement association result.
[0050] In one embodiment, due to the varying accuracy and strengths of different sensors, target information with different measurement attributes is correlated. First, target information pairs are created for each sensor, allowing them to correlate with the same target. For example, LiDAR-Radar, LiDAR-Camera, LiDAR-Vision, Radar-Camera, and Radar-Vision associations are established. Specifically, the primary and secondary sensors are determined based on the measurement attributes and information sources. For example, in this embodiment, the primary and secondary sensors are prioritized in the order of LiDAR > Radar > Camera. Based on the target information pairs, an appropriate gate type is selected. Gate types include lateral range gate, longitudinal range gate, velocity gate, type gate, heading gate, and 2D gate. The gate threshold for the second correlation gate is determined based on the gate type and trajectory, preliminarily selecting target information that meets the second correlation gate as the associated target information. When selecting associated target information, it is necessary to ensure that the target information pairs meet the gate requirements.
[0051] After determining the associated target information, the Mahalanobis distance between the associated target information and the trajectory is calculated, and the target information pair with the shortest Mahalanobis distance is selected as the inter-measurement association result.
[0052] Optionally, constructing an associated storage network according to the inter-measurement association results includes: Determining the order of association of the target information according to information priority; Based on the inter-measurement association results and the association order, the associated measurement relationships corresponding to different information sources are stored to obtain the associated storage network.
[0053] In one embodiment, the information priority is determined according to the information source, and the association order of the target information is determined by the information priority. The association storage structure is as follows: Figure 2 As shown, after obtaining the inter-measurement association result, for each measurement attribute, a SensorObject class is constructed. In each SensorObject class, an AssociatedObject class is initialized. This class uses a graph edge data structure to store multiple associated sensor targets, making it easy to obtain associated information. Through the AssociatedObject class, the target information of each other sensor with an inter-measurement association relationship is updated to the corresponding AssociatedObject of the target information SensorObject class to complete the storage of the association result. For example, LiDAR1 has an association relationship with Radar1, and Radar1 has an association relationship with Camera1. In the SensorObject class with LiDAR1, the initialized AssociatedObject class associates and stores the measurement of Radar1, and further stores the measurement of Camera1 associated with Radar1, thereby forming a complete association result storage structure. Accordingly, according to Figure 3 The graph edge structure shown will use Radar1, Radar2, Camera1, Camera2, and Radar3 as measurement SensorObject classes, respectively storing the corresponding associated measurements to obtain an associated storage network.
[0054] In one embodiment, the class diagram of AssociatedObject includes: std::shared+ptr <object>object_: target information of the association; SensorInfo sensor_info_: used to store which sensor the target comes from; double timestamp_: used to store the detection timestamp of the target; std::shared_ptr <associatedobject>next_associated_object_: points to the next sensor object associated with this measurement. If no other measurement is associated, this pointer is nullptr.
[0055] Optionally, the creating a target based on the association storage network and the inter-measurement association result, and associating the target with a corresponding trajectory includes: Based on the association relationship indicated by the inter-measurement association result, extract all the target information associated with the same target from the association storage network to create a target; The trajectory of the target is determined according to the trajectory corresponding to the target information associated with the target.
[0056] In one embodiment, multi-sensor information associated with a target is extracted from an associated storage network through inter-measurement association results to create a target, wherein the multi-sensor information includes information such as the position, speed, and size of the target.
[0057] Optionally, updating the trajectory of the target based on the trajectory association result to obtain the update result includes: determining the target information related to the trajectory of the target according to the trajectory association result; Determining a target information combination associated with the target information based on the inter-measurement association result; The target information combination is processed by a preset filtering algorithm, the state of the target is estimated, and an updated result is obtained.
[0058] In one embodiment, the sensor information associated with the target is determined through trajectory association results, and the optimal sensor information is selected as the measurement to construct an optimal measurement combination, namely the target information combination. Kalman filtering is then used to update the target. The optimal sensor information is determined by a preset information priority. For example, when both LiDAR1 and Radar1 have location information associated with the target, the measurement acquired by LiDAR1 is considered the superior location information based on the preset information priority. The optimal target information combination is determined based on the information priority, and the target information combination is processed using a Kalman filter algorithm to update the target's status.
[0059] Optionally, the determining the target management strategy according to the update result includes: When the trajectory corresponding to the target does not have the update result for more than a preset time, the target is deleted; When the update result satisfies a preset release condition, the target corresponding to the update result is released.
[0060] In one embodiment, if a target has not been associated for a predetermined period of time, it is considered lost and is deleted. When the target's update result satisfies a predetermined release condition, the target is released, allowing further driving control based on the update result. For example, if the target's update result indicates that the target is too close, the target is released, and subsequent processing allows for vehicle braking and other operations.
[0061] An embodiment of the present invention provides a multi-target association management device, comprising: The inter-measurement correlation module is used to perform inter-measurement correlation on the acquired target information based on the information source and obtain the inter-measurement correlation result; A storage module, configured to construct an association storage network based on the inter-measurement association results, wherein the association storage network is composed of a graph edge data structure; a trajectory measurement association module, configured to establish a trajectory for the target information of the same information source, and construct a trajectory association result based on the trajectory and the target information; a creation module, configured to create a target based on the association storage network and the association result between the measurements, and associate the target with a corresponding trajectory; An updating module, configured to update the trajectory of the target based on the trajectory association result to obtain an updated result; A management module is used to determine a management strategy for the target according to the update result.
[0062] like Figure 4 As shown, an electronic device 400 provided by an embodiment of the present invention includes a memory 410 and a processor 420; the memory 410 is used to store computer programs; the processor 420 is used to implement the multi-target association management method as described above when executing the computer program.
[0063] In other words, an electronic device 400 includes a memory 410 and a processor 420 coupled to the memory 410; the memory 410 is configured to store a computer program; and the processor 420 is configured to perform the following operations when executing the computer program: Based on the information source, the obtained target information is correlated between measurements to obtain the correlation results between measurements; constructing an association storage network according to the inter-measurement association results, wherein the association storage network is composed of a graph edge data structure; Establishing a trajectory for the target information of the same information source, and constructing a trajectory association result based on the trajectory and the target information; creating a target based on the association storage network and the association results between the measurements, and associating the target with a corresponding trajectory; updating the trajectory of the target based on the trajectory association result to obtain an updated result; A management strategy for the target is determined according to the update result.
[0064] An embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the multi-target association management method described above is implemented.
[0065] In other words, a non-volatile computer-readable storage medium stores a computer program, which, when executed by a processor, causes the processor to perform the following operations: Based on the information source, the obtained target information is correlated between measurements to obtain the correlation results between measurements; constructing an association storage network according to the inter-measurement association results, wherein the association storage network is composed of a graph edge data structure; Establishing a trajectory for the target information of the same information source, and constructing a trajectory association result based on the trajectory and the target information; creating a target based on the association storage network and the association results between the measurements, and associating the target with a corresponding trajectory; updating the trajectory of the target based on the trajectory association result to obtain an updated result; A management strategy for the target is determined according to the update result.
[0066] Electronic device 400 includes a computing unit that can perform various appropriate actions and processes based on a computer program stored in a read-only memory (ROM) or a computer program loaded from a storage unit into a random access memory (RAM). The RAM can also store various programs and data required for device operation. The computing unit, ROM, and RAM are interconnected via a bus. An input / output (I / O) interface is also connected to the bus.
[0067] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware through a computer program. The program can be stored in a computer-readable storage medium. When executed, the program can include the processes in the above-described method embodiments. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM). In this application, the units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network elements. Some or all of these units can be selected based on actual needs to achieve the objectives of the embodiments of the present invention. Furthermore, the functional units in the various embodiments of the present invention can be integrated into a single processing unit, each unit can exist physically separately, or two or more units can be integrated into a single unit. These integrated units can be implemented in either hardware or software functional units.
[0068] Although the present invention is disclosed as above, the protection scope of the present invention is not limited thereto. Those skilled in the art may make various changes and modifications without departing from the spirit and scope of the present invention, and these changes and modifications will fall within the protection scope of the present invention.< / associatedobject> < / object>
Claims
1. A multi-objective association management method, characterized in that: include: Based on the information source, the obtained target information is correlated between measurements to obtain the correlation results between measurements; constructing an association storage network according to the inter-measurement association results, wherein the association storage network is composed of a graph edge data structure; Establishing a trajectory for the target information of the same information source, and constructing a trajectory association result based on the trajectory and the target information; creating a target based on the association storage network and the association results between the measurements, and associating the target with a corresponding trajectory; updating the trajectory of the target based on the trajectory association result to obtain an updated result; A management strategy for the target is determined according to the update result.
2. The multi-target association management method according to claim 1, characterized in that: The establishing of a track for the target information of the same information source and constructing a track association result based on the track and the target information includes: determining a gate type according to a measurement attribute of the target information; determining a first associated gate for the trajectory according to the gate type; The target information is filtered according to the first correlation gate to obtain the trajectory correlation result.
3. The multi-objective association management method according to claim 2, characterized in that: The filtering the target information according to the first correlation gate to obtain the trajectory correlation result includes: calculating a Mahalanobis distance between associated target information and the trajectory, wherein the associated target information includes the target information satisfying the first associated wave gate; The associated target information with the shortest Mahalanobis distance is used as the trajectory association result of the trajectory.
4. The multi-objective association management method according to claim 1, characterized in that: The inter-measurement correlation of the acquired target information based on the information source to obtain the inter-measurement correlation result includes: constructing a target information pair from the two target information that meet the association requirements, wherein the association requirements include: the target information having different measurement attributes, and the two target information having an association relationship; determining a gate type according to the measurement attributes of the target information pair, and determining a second associated gate according to the gate type; The target information pairs are screened according to the second correlation gate to obtain the inter-measurement correlation result.
5. The multi-target association management method according to claim 4, characterized in that: The step of screening the target information pair according to the second correlation gate to obtain the inter-measurement correlation result includes: determining information priorities of the two target information according to the information sources; determining a screening priority of the second associated wave gate according to the information priority; screening the target information pairs that meet the second association wave gate as associated target information pairs; Calculating the Mahalanobis distance between two associated target information in the associated target information pair; According to the screening priority, the associated target information pair with the shortest Mahalanobis distance and the highest priority is used as the inter-measurement association result.
6. The multi-target association management method according to claim 1, characterized in that: The constructing of the association storage network according to the inter-measurement association results includes: Determining the order of association of the target information according to information priority; Based on the inter-measurement association results and the association order, the associated measurement relationships corresponding to different information sources are stored to obtain the associated storage network.
7. The multi-objective association management method according to any one of claims 1 to 6, characterized in that: The updating of the target trajectory based on the trajectory association result to obtain the updated result includes: determining the target information related to the trajectory of the target according to the trajectory association result; Determining a target information combination associated with the target information based on the inter-measurement association result; The target information combination is processed by a preset filtering algorithm, the state of the target is estimated, and an updated result is obtained.
8. The multi-objective association management method according to any one of claims 1 to 6, characterized in that: The management strategy for determining the target according to the update result includes: When the trajectory corresponding to the target does not have the update result for more than a preset time, the target is deleted; When the update result satisfies a preset release condition, the target corresponding to the update result is released.
9. A multi-target association management device, characterized in that: include: The inter-measurement correlation module is used to perform inter-measurement correlation on the acquired target information based on the information source and obtain the inter-measurement correlation result; A storage module, configured to construct an association storage network based on the inter-measurement association results, wherein the association storage network is composed of a graph edge data structure; a trajectory measurement association module, configured to establish a trajectory for the target information of the same information source, and construct a trajectory association result based on the trajectory and the target information; a creation module, configured to create a target based on the association storage network and the association result between the measurements, and associate the target with a corresponding trajectory; An updating module, configured to update the trajectory of the target based on the trajectory association result to obtain an updated result; A management module is used to determine a management strategy for the target according to the update result.
10. An electronic device, characterized in that: including memory and processor; The memory is used to store computer programs; The processor is configured to implement the multi-target association management method according to any one of claims 1 to 8 when executing the computer program.
11. A computer-readable storage medium, characterized in that The storage medium stores a computer program, and when the computer program is executed by a processor, the multi-target association management method according to any one of claims 1 to 8 is implemented.