Target duplicate removal method and device, equipment and storage medium

Abstract

The invention relates to a target duplicate removal method and device, equipment and a storage medium. Comprising the steps that when a new perception target is perceived, the identity state of the new perception target is set to be a newly-added target, and a duplicate removal process for the new perception target is triggered; if the de-duplication result represents that the new sensing target has the adjacent region sensing target having the duplicate target relationship with the new sensing target, setting and updating the identity state of the new sensing target as a non-independent target; if the de-duplication result represents that the new sensing target does not have the adjacent region sensing target having the repeated target relationship with the new sensing target, setting and updating the identity state of the new sensing target as an independent target; wherein the sensing cell where the independent target is located is responsible for sensing data reporting, and the sensing cell where the non-independent target and the newly-added target are located is not responsible for sensing data reporting. By adopting the method and the device, the perception target de-duplication efficiency and de-duplication effect can be improved.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to a method, apparatus, device and storage medium for deduplication of targets. Background Technology

[0002] With the rapid development and promotion of network technology, as well as the rapid development of drones in recent years, the development of sensing services such as low-altitude economy, airway identification, intrusion detection, and electronic fences has been greatly promoted. In order to save frequency domain resources, operators have proposed an integrated communication and sensing network architecture that integrates communication, sensing, and computing.

[0003] In integrated sensing and communication networks, similar to how communication systems need to ensure service continuity, sensing systems also need to ensure the continuity of sensing services. When a sensed target is in motion, such as moving from the sensing range of the current cell to the sensing range of an adjacent cell, if the current cell and the adjacent cell have overlapping sensing areas, multiple cells may simultaneously sense the same target. Therefore, how to deduplicate / associate the trajectory of the same sensed target across multiple cells to ensure the uniqueness of the sensed target and the continuity of sensing services is a key issue that needs to be studied. Summary of the Invention

[0004] Therefore, it is necessary to provide a target deduplication method, apparatus, device, and storage medium to address the aforementioned technical problems, so as to realize the deduplication and association of sensing targets among multiple cells, and ensure the uniqueness of sensing targets and the continuity of sensing services.

[0005] Firstly, this application provides a method for deduplication of a target, including:

[0006] When a new target is detected, the identity status of the new target is set to the new target, and a deduplication process for the new target is triggered.

[0007] If the deduplication result indicates that the new perceived target has a neighboring perceived target with a duplicate target relationship, then the identity status of the new perceived target will be updated to a non-independent target.

[0008] If the deduplication result indicates that the new perceived target does not have any neighboring perceived targets with a duplicate target relationship with it, then the identity status of the new perceived target will be updated to an independent target.

[0009] Among them, the sensing cell where the independent target is located is responsible for reporting sensing data, while the sensing cell where the non-independent target and the newly added target are located is not responsible for reporting sensing data.

[0010] In one embodiment, the deduplication process for a new sensed target includes:

[0011] Determine the set of deduplicated neighboring targets for the new sensing target;

[0012] If the set of deduplicated neighboring targets for the new sensing target is empty, then the deduplication result is obtained, which indicates that the new sensing target does not have any neighboring sensing targets with duplicate target relationships.

[0013] In one embodiment, the neighbor sensing targets in the deduplicated neighbor target set satisfy the following condition:

[0014] The sensing cell where the neighboring sensing target is located and the sensing cell where the new sensing target is located have overlapping sensing areas.

[0015] In one embodiment, the conditions that the neighboring cell sensing target needs to meet also include:

[0016] The identity status of the target perceived in the neighboring cell is either an independent target or a newly added target.

[0017] In one embodiment, when the identity status of a neighboring target is a newly added target, the neighboring target also needs to meet the following conditions:

[0018] The first perception timestamp of the neighboring target is greater than the first perception timestamp of the new target; or, the first perception timestamp of the neighboring target is equal to the first perception timestamp of the new target, and the neighboring target and the new target satisfy the conditions for adding a new target.

[0019] In one embodiment, the conditions that the neighboring cell sensing target needs to meet also include:

[0020] If there are historical deduplication judgment results in the current deduplication window, the neighboring sensing target and the new sensing target are considered to be duplicate targets in the historical deduplication judgment results.

[0021] In one embodiment, the deduplication process for new sensed targets further includes:

[0022] If the set of deduplicated neighboring targets of the new sensing target is not empty, the new sensing target will be paired with each of the neighboring sensing targets in the set of deduplicated neighboring targets.

[0023] For any pair of targets, the distance between the two targets at the same sensing moment is calculated within the sensing time overlap window, and deduplication is performed based on the distance to obtain the deduplication result;

[0024] Update the target consecutive count based on the deduplication judgment result, and update the current deduplication window length based on the updated target consecutive count. Use the deduplication window length before the update as the historical deduplication window length. The target consecutive count is the number of times the deduplication judgment result representing the target pair as a duplicate target relationship is obtained consecutively.

[0025] Based on the relationship between the current deduplication window length and the historical deduplication window length, determine whether to end the deduplication process. When determining to end the deduplication process, determine the deduplication result of the target pair based on the number of consecutive occurrences of the target.

[0026] In one embodiment, determining whether to terminate the deduplication process based on the relationship between the current deduplication window length and the historical deduplication window length includes:

[0027] If the current deduplication window length is greater than the historical deduplication window length, continue the deduplication process for the newly perceived target; if the current deduplication window length is less than or equal to the historical deduplication window length, determine to end the deduplication process for the target pair.

[0028] In one embodiment, updating the target consecutive count based on the deduplication determination result includes:

[0029] When the deduplication result indicates that the two perceived targets are duplicate targets, increment the number of consecutive targets by 1; when the deduplication result indicates that the two perceived targets are not duplicate targets, reset the number of consecutive targets to 0.

[0030] In one embodiment, updating the current deduplication window length based on the updated target consecutive count includes:

[0031] Based on the updated target consecutive count and the base window length, determine the candidate window length; the minimum value between the candidate window length and the preset window length threshold is taken as the current deduplication window length.

[0032] In one embodiment, determining the deduplication result of the target pair based on the number of consecutive occurrences of the target includes:

[0033] If the number of consecutive targets is greater than 0, the result is a deduplication of the two perceived targets in the target pair as repeating targets; if the number of consecutive targets is equal to 0, the result is a deduplication of the two perceived targets in the target pair as non-repeating targets.

[0034] In one embodiment, a deduplication check based on distance is triggered and a deduplication result is obtained when any of the following conditions are met:

[0035] The current deduplication window ends; the trajectory overlap time of the two sensing targets in the target pair no longer changes; the cumulative number of times the distance between the two sensing targets in the target pair exceeds the preset distance threshold exceeds the preset number threshold; the number of sensing moments used to calculate the distance between the two sensing targets in the target pair exceeds the preset number threshold; the interaction state of any sensing target in the target pair is abnormal; the difference between the timestamp of the latest reported sensing data of the new sensing target and the timestamp of the first reported sensing data of the new sensing target exceeds the preset allowed sensing delay.

[0036] In one embodiment, calculating the distance between two sensing targets at the same sensing moment within a sensing time overlap window includes:

[0037] For each sensing time of one of the sensing targets in a target pair, the position information of the other sensing target at each sensing time is determined by an interpolation algorithm;

[0038] Determine the distance between the location information of two sensing targets at the same sensing moment.

[0039] Secondly, embodiments of this application also provide a target deduplication device, comprising:

[0040] The deduplication module is used to set the identity status of the new sensing target to a newly added target when a new sensing target is detected, and to trigger a deduplication process for the new sensing target.

[0041] The identity status change module is used to update the identity status of the new sensing target to a non-independent target if the deduplication result indicates that the new sensing target has a neighboring sensing target with a duplicate target relationship; and to update the identity status of the new sensing target to an independent target if the deduplication result indicates that the new sensing target does not have a neighboring sensing target with a duplicate target relationship.

[0042] The sensing cell where the independent target is located is responsible for reporting sensing data, while the sensing cells where the non-independent target and the newly added target are located are not responsible for reporting sensing data.

[0043] In one embodiment, the deduplication module is specifically used for:

[0044] Determine the set of deduplicated neighboring targets for the new sensing target;

[0045] If the set of deduplicated neighboring targets for the new sensing target is empty, then the deduplication result is obtained, which indicates that the new sensing target does not have any neighboring sensing targets with duplicate target relationships.

[0046] In one embodiment, the neighboring sensing targets in the deduplicated neighboring target set satisfy the following condition: the sensing cell where the neighboring sensing target is located has an overlapping sensing area with the sensing cell where the new sensing target is located.

[0047] In one embodiment, the conditions that the neighboring cell sensing target needs to meet also include: the identity status of the neighboring cell sensing target is an independent target or a newly added target.

[0048] In one embodiment, when the identity status of a neighboring target is a newly added target, the neighboring target also needs to meet the following conditions:

[0049] The first perception timestamp of the neighboring target is greater than the first perception timestamp of the new target; or, the first perception timestamp of the neighboring target is equal to the first perception timestamp of the new target, and the neighboring target and the new target satisfy the conditions for adding a new target.

[0050] In one embodiment, the conditions that the neighboring cell sensing target needs to meet also include:

[0051] If there are historical deduplication judgment results in the current deduplication window, the neighboring sensing target and the new sensing target are considered to be duplicate targets in the historical deduplication judgment results.

[0052] In one embodiment, the deduplication module is further used for:

[0053] If the set of deduplicated neighboring targets of the new sensing target is not empty, the new sensing target will be paired with each of the neighboring sensing targets in the set of deduplicated neighboring targets.

[0054] For any pair of targets, the distance between the two targets at the same sensing moment is calculated within the sensing time overlap window, and deduplication is performed based on the distance to obtain the deduplication result;

[0055] Update the target consecutive count based on the deduplication judgment result, and update the current deduplication window length based on the updated target consecutive count. Use the deduplication window length before the update as the historical deduplication window length. The target consecutive count is the number of times the deduplication judgment result representing the target pair as a duplicate target relationship is obtained consecutively.

[0056] Based on the relationship between the current deduplication window length and the historical deduplication window length, determine whether to end the deduplication process. When determining to end the deduplication process, determine the deduplication result of the target pair based on the number of consecutive occurrences of the target.

[0057] In one embodiment, the deduplication module is further used for:

[0058] If the current deduplication window length is greater than the historical deduplication window length, continue the deduplication process for the newly perceived target; if the current deduplication window length is less than or equal to the historical deduplication window length, determine to end the deduplication process for the target pair.

[0059] In one embodiment, the deduplication module is further used for:

[0060] When the deduplication result indicates that the two perceived targets are duplicate targets, increment the number of consecutive targets by 1; when the deduplication result indicates that the two perceived targets are not duplicate targets, reset the number of consecutive targets to 0.

[0061] In one embodiment, the deduplication module is further used for:

[0062] Based on the updated target consecutive count and the base window length, determine the candidate window length; the minimum value between the candidate window length and the preset window length threshold is taken as the current deduplication window length.

[0063] In one embodiment, the deduplication module is further used for:

[0064] If the number of consecutive targets is greater than 0, the result is a deduplication of the two perceived targets in the target pair as repeating targets; if the number of consecutive targets is equal to 0, the result is a deduplication of the two perceived targets in the target pair as non-repeating targets.

[0065] In one embodiment, a deduplication check based on distance is triggered and a deduplication result is obtained when any of the following conditions are met:

[0066] The current deduplication window ends; the trajectory overlap time of the two sensing targets in the target pair no longer changes; the cumulative number of times the distance between the two sensing targets in the target pair exceeds the preset distance threshold exceeds the preset number threshold; the number of sensing moments used to calculate the distance between the two sensing targets in the target pair exceeds the preset number threshold; the interaction state of any sensing target in the target pair is abnormal; the difference between the timestamp of the latest reported sensing data of the new sensing target and the timestamp of the first reported sensing data of the new sensing target exceeds the preset allowed sensing delay.

[0067] In one embodiment, the deduplication module is further used for:

[0068] For each sensing time of one of the sensing targets in a target pair, the position information of the other sensing target at each sensing time is determined by an interpolation algorithm; the distance between the position information of the two sensing targets at the same sensing time is determined.

[0069] Thirdly, a target deduplication device is provided, comprising: a memory, a transceiver, and a processor.

[0070] Memory is used to store computer programs; transceiver is used to send and receive data under the control of the processor; processor is used to read the computer programs from memory and perform the following operations:

[0071] When a new target is detected, the identity status of the new target is set to the new target, and a deduplication process for the new target is triggered.

[0072] If the deduplication result indicates that the new perceived target has a neighboring perceived target with a duplicate target relationship, then the identity status of the new perceived target will be updated to a non-independent target.

[0073] If the deduplication result indicates that the new perceived target does not have any neighboring perceived targets with a duplicate target relationship with it, then the identity status of the new perceived target will be updated to an independent target;

[0074] Among them, the sensing cell where the independent target is located is responsible for reporting sensing data, while the sensing cell where the non-independent target and the newly added target are located is not responsible for reporting sensing data.

[0075] In one embodiment, the processor is also configured to perform the following operations:

[0076] Determine the set of deduplicated neighboring targets for the new sensing target;

[0077] If the set of deduplicated neighboring targets for the new sensing target is empty, then the deduplication result is obtained, which indicates that the new sensing target does not have any neighboring sensing targets with duplicate target relationships.

[0078] In one embodiment, the neighboring sensing targets in the deduplicated neighboring target set satisfy the following condition: the sensing cell where the neighboring sensing target is located has an overlapping sensing area with the sensing cell where the new sensing target is located.

[0079] In one embodiment, the conditions that the neighboring cell sensing target needs to meet also include: the identity status of the neighboring cell sensing target is an independent target or a newly added target.

[0080] In one embodiment, when the identity status of a neighboring target is a newly added target, the neighboring target also needs to meet the following conditions:

[0081] The first perception timestamp of the neighboring target is greater than the first perception timestamp of the new target; or, the first perception timestamp of the neighboring target is equal to the first perception timestamp of the new target, and the neighboring target and the new target satisfy the conditions for adding a new target.

[0082] In one embodiment, the conditions that the neighboring cell sensing target needs to meet also include:

[0083] If there are historical deduplication judgment results in the current deduplication window, the neighboring sensing target and the new sensing target are considered to be duplicate targets in the historical deduplication judgment results.

[0084] In one embodiment, the processor is also configured to perform the following operations:

[0085] If the set of deduplicated neighboring targets of the new sensing target is not empty, the new sensing target will be paired with each of the neighboring sensing targets in the set of deduplicated neighboring targets.

[0086] For any pair of targets, the distance between the two targets at the same sensing moment is calculated within the sensing time overlap window, and deduplication is performed based on the distance to obtain the deduplication result;

[0087] Update the target consecutive count based on the deduplication judgment result, and update the current deduplication window length based on the updated target consecutive count. Use the deduplication window length before the update as the historical deduplication window length. The target consecutive count is the number of times the deduplication judgment result representing the target pair as a duplicate target relationship is obtained consecutively.

[0088] Based on the relationship between the current deduplication window length and the historical deduplication window length, determine whether to end the deduplication process. When determining to end the deduplication process, determine the deduplication result of the target pair based on the number of consecutive occurrences of the target.

[0089] In one embodiment, the processor is also configured to perform the following operations:

[0090] When the current deduplication window length is longer than the historical deduplication window length, continue to execute the deduplication process for the new sensing target;

[0091] If the current deduplication window length is less than or equal to the historical deduplication window length, the deduplication process for the target pair is terminated.

[0092] In one embodiment, the processor is also configured to perform the following operations:

[0093] When the deduplication result indicates that two perceived targets are related as repeated targets, increment the number of consecutive targets by 1.

[0094] When the deduplication result indicates that the two perceived targets are not duplicate targets, the consecutive count of the targets is set to 0.

[0095] In one embodiment, the processor is also configured to perform the following operations:

[0096] The candidate window length is determined based on the updated target consecutive count and the base window length;

[0097] The minimum value between the candidate window length and the preset window length threshold is used as the current deduplication window length.

[0098] In one embodiment, the processor is also configured to perform the following operations:

[0099] If the number of consecutive targets is greater than 0, the result of deduplication is obtained to represent the relationship between the two perceived targets in the target pair as repeated targets;

[0100] If the number of consecutive targets is 0, then the result is the deduplication result representing the relationship between the two perceived targets in the target pair as non-repeating targets.

[0101] In one embodiment, a deduplication check based on distance is triggered and a deduplication result is obtained when any of the following conditions are met:

[0102] The current deduplication window ends; the trajectory overlap time of the two sensing targets in the target pair no longer changes; the cumulative number of times the distance between the two sensing targets in the target pair exceeds the preset distance threshold exceeds the preset number threshold; the number of sensing moments used to calculate the distance between the two sensing targets in the target pair exceeds the preset number threshold; the interaction state of any sensing target in the target pair is abnormal; the difference between the timestamp of the latest reported sensing data of the new sensing target and the timestamp of the first reported sensing data of the new sensing target exceeds the preset allowed sensing delay.

[0103] In one embodiment, the processor is also configured to perform the following operations:

[0104] For each sensing time of one of the sensing targets in a target pair, the position information of the other sensing target at each sensing time is determined by an interpolation algorithm;

[0105] Determine the distance between the location information of two sensing targets at the same sensing moment.

[0106] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of any target deduplication method.

[0107] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of any target deduplication method.

[0108] The aforementioned target deduplication method, apparatus, device, and storage medium, when a sensing node detects a new sensing target, sets the identity status of the new sensing target to "new target" and triggers a deduplication process for the new sensing target to obtain the deduplication result. If the deduplication result indicates that the new sensing target has neighboring sensing targets with a duplicate target relationship, the identity status of the new sensing target is updated to "non-independent target." If the deduplication result indicates that the new sensing target does not have any neighboring sensing targets with a duplicate target relationship, the identity status of the new sensing target is updated to "independent target." The sensing cell containing the independent target is responsible for reporting sensing data, while the sensing cells containing the non-independent target and the new target are not responsible for reporting sensing data. By employing the target deduplication method, apparatus, device, and storage medium provided in the embodiments of this application, perception data is reported only for perception targets whose identity status is independent. By performing a deduplication process on new perception targets, and obtaining the deduplication result indicating whether there are neighboring perception targets with duplicate target relationships with the new perception target, the identity status of the perception target is updated to independent target or non-independent target based on the deduplication result. Target deduplication can be achieved from the perception data reporting stage, ensuring the uniqueness of the perception data reported, thereby ensuring the uniqueness of the perception target and the continuity of perception services. Attached Figure Description

[0109] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0110] Figure 1 A flowchart illustrating a target deduplication method in one embodiment is shown;

[0111] Figure 2 A schematic diagram illustrating the perception of target identity state transitions in one embodiment is shown;

[0112] Figure 3 A flowchart illustrating the deduplication process for a new sensed target in one embodiment is shown.

[0113] Figure 4 A schematic diagram of a scenario from Example 1 in one embodiment is shown;

[0114] Figure 5 A schematic diagram of a scenario from Example 2 in one embodiment is shown;

[0115] Figure 6 A schematic diagram of a scenario from Example 3 in one embodiment is shown;

[0116] Figure 7 A schematic diagram of a scenario from Example 4 in one embodiment is shown;

[0117] Figure 8 This is a structural block diagram of the target deduplication device in one embodiment;

[0118] Figure 9 This is a structural block diagram of a target deduplication device in one embodiment. Detailed Implementation

[0119] In the embodiments of this application, the term "and / or" describes the relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent three cases: A alone, A and B simultaneously, and B alone. The character " / " generally indicates that the preceding and following associated objects have an "or" relationship.

[0120] In the embodiments of this application, the term "multiple" refers to two or more, and other quantifiers are similar.

[0121] 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 a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0122] The technical solutions provided in this application can be applied to a variety of systems. For example, applicable systems may include Long Term Evolution (LTE) systems, LTE Frequency Division Duplex (FDD) systems, LTE Time Division Duplex (TDD) systems, Long Term Evolution Advanced (LTE-A) systems, Universal Mobile Telecommunications System (UMTS), Worldwide Interoperability for Microwave Access (WiMAX) systems, 5G New Radio (NR) systems, and their evolved communication systems. These systems may include terminal equipment, network equipment, and core network components, such as Evolved Packet System (EPS) and 5G systems (5GS).

[0123] The terminal devices involved in the embodiments of this application can be devices that provide voice and / or data connectivity to users, handheld devices with wireless connectivity, or other processing devices connected to a wireless modem. The names of the terminal devices may differ in different systems; for example, in a 5G system, a terminal device can be called User Equipment (UE). Wireless terminal devices can be USB storage devices, other personal computer memory devices, and dongles. They can also communicate with one or more core networks (CNs) via a Radio Access Network (RAN). Wireless terminal devices can be mobile terminal devices, such as mobile phones (or "cellular" phones) and computers with mobile terminal devices. For example, they can be portable, pocket-sized, handheld, computer-embedded, or vehicle-mounted mobile devices that exchange voice and / or data with the radio access network. Examples of such devices include Personal Communication Service (PCS) phones, cordless phones, Session Initiated Protocol (SIP) phones, Wireless Local Loop (WLL) stations, Personal Digital Assistants (PDAs), personal computers, tablets, and Machine-type Communication (MTC) terminal devices. Wireless terminal devices can also be referred to as systems, subscriber units, subscriber stations, mobile stations, mobile terminals, remote stations, access points, remote terminals, access terminals, user terminals, user agents, user devices, and wireless access devices and routers / modems that meet the limitations of this definition; however, this application does not limit the scope of the embodiments described.

[0124] The network device involved in the embodiments of this application can be a base station, which may include multiple cells providing services to terminals. Depending on the specific application, the base station may also be called an access point, or a device in the access network that communicates with wireless terminal devices through one or more sectors on the air interface, or other names. The network device can be used to exchange received air frames with Internet Protocol (IP) packets, acting as a router between the wireless terminal device and the rest of the access network, where the rest of the access network may include an Internet Protocol (IP) communication network. The network device can also coordinate the attribute management of the air interface. For example, the network device involved in the embodiments of this application may be an evolved Node B (eNB or e-NodeB) in a long term evolution (LTE) system, a 5G base station (gNB) in a next generation system, or a Home evolved Node B (HeNB), relay node, femto, pico, network testing equipment, etc., and is not limited in the embodiments of this application. In some network architectures, network devices may include centralized unit (CU) nodes and distributed unit (DU) nodes, which may also be geographically separated.

[0125] The target deduplication method provided in this application can be applied to sensing nodes, which may include devices such as base stations and centralized processing nodes. The base stations manage multiple sensing cells, and the centralized processing nodes manage multiple sensing cells under multiple base stations. This includes the base stations reporting sensing data to the centralized processing node, which is responsible for deduplication and reporting of sensing data, as well as interacting with sensing information from neighboring centralized processing nodes. Furthermore, the sensing information acquired from neighboring centralized processing nodes does not need to be distributed to the base stations it manages. (Refer to...) Figure 1 As shown, the target deduplication method provided in this application embodiment includes the following steps 101 to 103, wherein:

[0126] Step 101: When a new target is detected, set the identity status of the new target to "new target" and trigger the deduplication process for the new target.

[0127] In this embodiment, the perceived target refers to the specific object that the integrated sensing system or device needs to identify, track, analyze, or operate, such as a person, vehicle, drone, or other object. When a new perceived target is detected in any of the managed sensing cells, the identity status of the new perceived target can be initialized as a newly added target, which represents the perceived target that is reporting sensing data for the first time.

[0128] After initializing the identity status of the new sensing target as a newly added target, the new sensing target will trigger a deduplication process to obtain the corresponding deduplication result. This deduplication process may include sensing data interaction between the current sensing node and neighboring sensing nodes to obtain sensing data from neighboring sensing cells, and determining the deduplication result based on the obtained sensing data. The process of this deduplication will be described in the following embodiments and will not be repeated here. Neighboring sensing cells are sensing cells that have a neighboring relationship with the sensing cell currently sensing the new sensing target. This neighboring relationship refers to overlapping sensing areas between different sensing cells managed by different sensing nodes.

[0129] Step 102: If the deduplication result indicates that the new perceived target has a neighboring perceived target with a duplicate target relationship, then update the identity status of the new perceived target to a non-independent target.

[0130] Step 103: If the deduplication result indicates that the new perceived target does not have any neighboring perceived targets with a duplicate target relationship with it, then update the identity status of the new perceived target to an independent target.

[0131] Among them, the sensing cell where the independent target is located is responsible for reporting sensing data, while the sensing cell where the non-independent target and the newly added target are located is not responsible for reporting sensing data.

[0132] In this embodiment, the identity status of a sensing target can include one of the following: a newly added target, an independent target, and a non-independent target. After the deduplication process, if the deduplication result indicates that the new sensing target has a neighboring sensing target with a duplicate target relationship, it can be determined that a sensing target with duplicate information already exists in the adjacent sensing cells. Therefore, the identity status of the new sensing target can be updated from a newly added target to a non-independent target based on the deduplication result. In this way, the current sensing node does not report sensing data for the new sensing target in response to the identity status of the non-independent target. Instead, the sensing node of the sensing cell where the neighboring sensing target is located reports the sensing data, thus achieving the goal of reporting the sensing data of the new sensing target by only one sensing cell. That is, the deduplication of sensing targets can be achieved during the sensing data reporting stage.

[0133] Alternatively, after the deduplication process, if the deduplication result indicates that there are no neighboring sensing targets with duplicate target relationships with the new sensing target, it can be determined that there are no neighboring sensing targets with duplicate relationships with the new sensing target in the adjacent sensing cells. Therefore, the identity status of the new sensing target can be updated from a newly added target to an independent target, and the sensing node managing the sensing cell where the new sensing target is located is responsible for reporting the sensing data of the new sensing target. In this way, the sensing data of the new sensing target can be reported by only one sensing cell, without causing duplicate reporting of sensing data.

[0134] For example, taking the existence of sensing node 1 and sensing node 2 as an example, assume that sensing node 1 manages sensing cells 1-1 and 1-2, and sensing node 2 manages sensing cells 2-1 and 2-2, with overlapping sensing areas between sensing cells 1-2 and 2-2. If sensing node 1 first senses a sensing target in sensing cell 1-1, defining it as sensing target A, and the deduplication result indicates that there are no neighboring sensing targets with duplicate target relationships with sensing target A, then sensing node 1 can update the identity status of sensing target A to independent target, and sensing node 1 will be responsible for reporting the sensing data of sensing target A.

[0135] Suppose the sensing target continues to move into the overlapping sensing area of ​​sensing cell 1-2 and sensing cell 2-2, then sensing node 2 will sense the sensing target for the first time and define it as sensing target B. At this time, the deduplication result determined by the deduplication process indicates that sensing target B has a sensing target A with a duplicate target relationship. Then, sensing node 2 can update the identity status of sensing target B to non-independent target, that is, sensing node 2 will not report the sensing data of the sensing target. The sensing node 1 will still be responsible for reporting the sensing data of the sensing target, thereby realizing the deduplication of the sensing target.

[0136] In other words, a given entity may be detected sequentially or simultaneously by multiple sensing cells. A single sensing cell can categorize detected targets into three types: newly added targets, independent targets, and non-independent targets. Within that sensing cell, each target corresponds to only one state at any given time. Therefore, the entity has different statuses in different sensing cells. The sensing cell containing the independent target is responsible for reporting sensing data, while the sensing cells containing the newly added / non-independent target do not report sensing data.

[0137] Among them, reference Figure 2 As shown, three target states can be transitioned based on the deduplication result, including:

[0138] (1) For a new sensing target that reports sensing data for the first time, initialize its identity status to "new target". (2) For a new target, if the deduplication result shows that the new sensing target has no duplicate target relationship with any neighboring sensing targets, then the identity status of the new sensing target will change from "new target" to "independent target". (3) For a new sensing target, if the deduplication result shows that the new sensing target has a duplicate target relationship with at least one neighboring sensing target, then the identity status of the new sensing target will change from "new target" to "non-independent target". (4) For a sensing target with an identity status of non-independent target, if the independent target with a duplicate target relationship with it disappears, and the sensing target is selected as the next data reporting anchor point, then the identity status of the sensing target will change from "non-independent target" to "independent target".

[0139] Sensing targets with an independent identity status are used as data reporting anchors and reported using a reporting ID (Identification). If a sensing target with an independent identity status cannot be detected within a sensing cell, it is considered that the sensing target has disappeared. From multiple sensing targets with a non-independent identity status that have a duplicate target relationship with the original sensing target, one of them is selected as the new data reporting anchor based on the priority principle. The cell where the original independent target is located needs to transfer the list of duplicate targets associated with the original independent target to the determined next data reporting anchor. This list of duplicate targets is used to record sensing targets that are duplicates of the original independent target. The cell where the next data reporting anchor is located will continue to report sensing data using the same reporting ID as the previous independent target. The priority principle may include, but is not limited to: 1. Prioritizing non-independent targets in the same station as the next data reporting anchor; 2. When the conditions in priority principle (1) are the same, prioritizing non-independent targets with the largest point-of-sight timestamp as the next data reporting anchor.

[0140] The target deduplication method provided in this application involves a sensing node setting the identity status of a new sensing target to "new target" when it detects a new sensing target, and triggering a deduplication process for the new sensing target to obtain a deduplication result. If the deduplication result indicates that the new sensing target has neighboring sensing targets with duplicate target relationships, the identity status of the new sensing target is updated to "non-independent target". If the deduplication result indicates that the new sensing target does not have neighboring sensing targets with duplicate target relationships, the identity status of the new sensing target is updated to "independent target". The sensing cell containing the independent target is responsible for reporting sensing data, while the sensing cells containing the non-independent target and the new target are not responsible for reporting sensing data. The target deduplication method provided in this application only reports perception data for perception targets whose identity status is independent. By performing a deduplication process on new perception targets, the deduplication result is obtained to indicate whether there are neighboring perception targets with duplicate target relationships with the new perception target. Based on the deduplication result, the identity status of the perception target is updated to independent target or non-independent target. Target deduplication can be achieved from the perception data reporting stage, ensuring the uniqueness of the perception data reported, thereby ensuring the uniqueness of the perception target and the continuity of perception services.

[0141] The following examples will provide a detailed description of the deduplication process for new sensing targets.

[0142] In one exemplary embodiment, the deduplication process for a new perceived target may include:

[0143] Determine the set of deduplicated neighboring targets for the new sensing target; if the set of deduplicated neighboring targets for the new sensing target is empty, then the deduplication result is obtained, which indicates that the new sensing target does not have any neighboring sensing targets with duplicate target relationships with it.

[0144] In this embodiment, when a new sensing target is detected, a set of deduplicated neighboring targets for the new sensing target can be determined. This set includes neighboring sensing targets that may have duplicate target relationships with the new sensing target; that is, these neighboring sensing targets are the sensing targets to be deduplicated with the new sensing target. If the set of deduplicated neighboring targets is empty (for example, when the sensing cell where the new sensing target is located does not have overlapping sensing areas with other sensing cells, the set of deduplicated neighboring targets for the new sensing target is empty; or when there are no sensing targets in sensing cells that have overlapping sensing areas with the cell where the new sensing target is located, the set of deduplicated neighboring targets for the new sensing target is empty), it indicates that there are no neighboring sensing targets with duplicate target relationships with the new sensing target. Thus, a deduplication result indicating that there are no neighboring sensing targets with duplicate target relationships with the new sensing target can be obtained. The sensing node can update the identity status of the new sensing target to an independent target based on the deduplication result. In response to the identity status of the independent target, the sensing node can continuously sense the new sensing target and report the sensing data of the new sensing target.

[0145] In this embodiment, when determining the set of deduplicated neighboring targets corresponding to a new sensing target, only sensing targets that meet preset neighboring conditions are allowed to be added to the set of deduplicated neighboring targets and participate in the deduplication calculation. In this way, sensing targets that may have duplicate target relationships with the new sensing target can be filtered out for deduplication calculation through the neighboring conditions, which can effectively reduce the computational load of the deduplication process and improve the deduplication efficiency of sensing targets. The neighboring conditions will be explained in the following embodiments.

[0146] In one example embodiment, when determining the set of deduplicated neighboring targets corresponding to a new sensing target, the neighboring sensing targets in the set of deduplicated neighboring targets need to meet the following conditions:

[0147] The sensing cell where the neighboring sensing target is located and the sensing cell where the new sensing target is located have overlapping sensing areas.

[0148] In this embodiment, taking the presence of sensing node 1 and sensing node 2 in the current network as an example, it is assumed that sensing node 1 manages sensing cells 1-1 and 1-2, and sensing node 2 manages sensing cells 2-1 and 2-2. Sensing cells 1-2 and 2-2 have overlapping sensing areas. Assume that sensing node 1 detects a new sensing target A in sensing cell 1-2, and sensing target C is included in sensing cell 2-1 managed by sensing node 2. Sensing targets B, D, and E are included in sensing cell 2-2 managed by sensing node 2. Since sensing cell 2-2 and 1-2 have overlapping sensing areas, the neighboring sensing targets included in the deduplicated neighbor target set corresponding to sensing target A may include sensing targets B, D, and E.

[0149] In one exemplary embodiment, the conditions that the neighboring cell sensing target needs to meet may further include:

[0150] The identity status of the target perceived in the neighboring cell is either an independent target or a newly added target.

[0151] In this embodiment of the application, since only independent targets can be used to report sensing data, during deduplication calculation, the new sensing target is only deduplicated with the sensing targets whose identity status is independent, in order to determine whether there is a duplicate target relationship between the two. If there is a duplicate target relationship, the identity status of the new sensing target is updated from newly added target to independent target. This can reduce the amount of deduplication calculation and improve the deduplication efficiency of sensing targets.

[0152] There is also a special scenario where a new sensing target is simultaneously perceived for the first time by different sensing nodes. That is, its identity status in different sensing nodes is that of a new target. Therefore, the sensing targets whose identity status is that of a new target in different sensing nodes also need to participate in the deduplication calculation to determine whether there is a duplicate target relationship between the two.

[0153] Taking the previous example, assuming that the identity status of perceived target B is an independent target, the identity status of perceived target D is a non-independent target, and the identity status of perceived target E is a newly added target, then the set of deduplicated neighboring targets of perceived target A can include: perceived target B and perceived target E.

[0154] In an exemplary embodiment, when the identity status of a neighboring target is a newly added target, theoretically, the sensing node that senses the neighboring target will also trigger the deduplication process. To ensure that the deduplication calculation between newly added targets is performed only in one sensing node, the neighboring target must further meet the following conditions when its identity status is a newly added target:

[0155] The first perception timestamp of the neighboring target is greater than the first perception timestamp of the new target; or, the first perception timestamp of the neighboring target is equal to the first perception timestamp of the new target, and the neighboring target and the new target satisfy the conditions for adding a new target.

[0156] When the first sensing timestamp of a neighboring target is greater than the first sensing timestamp of a new target, the neighboring target can be added to the deduplicated neighboring target set of the new target. Correspondingly, the new target will not be added to the deduplicated neighboring target set corresponding to the neighboring target. That is, the deduplication calculation between the neighboring target and the new target is only performed within the sensing cell where the new target is located, thus ensuring the uniqueness of the deduplication result.

[0157] In a specific scenario, the first sensing timestamp of a neighboring target is the same as the first sensing timestamp of a new target. In this case, a new target constraint condition is used to determine whether the neighboring target can be added to the deduplicated neighboring target set of the new target. However, this embodiment does not specifically limit the new target constraint condition; any condition that guarantees the uniqueness of the deduplication calculation between new target pairs is applicable to this embodiment. For example, the cell number of the neighboring target is greater than the cell number of the new target, or the cell number of the neighboring target is greater than the cell number of the new target.

[0158] In this way, the embodiments of this application can ensure that the deduplication calculation of the newly added target pair is performed only in one sensing cell. Therefore, it can not only reduce the amount of deduplication calculation and improve the deduplication efficiency of the sensing target, but also ensure the uniqueness of the deduplication result.

[0159] In one exemplary embodiment, the conditions that the neighboring cell sensing target needs to meet may further include:

[0160] If there are historical deduplication judgment results in the current deduplication window, the neighboring sensing target and the new sensing target are considered to be duplicate targets in the historical deduplication judgment results.

[0161] In this embodiment, the deduplication window is a time window for performing a deduplication judgment. The deduplication process includes at least one deduplication window. Within each deduplication window, a deduplication judgment is performed on the new sensing target based on the neighboring sensing targets in the deduplication neighboring target set, resulting in a deduplication judgment result. This result is used to characterize whether the neighboring sensing target and the new sensing target are duplicate targets within the current deduplication window. The deduplication process and deduplication judgment procedure will be described in detail in the following embodiments, and will not be repeated here.

[0162] In determining the deduplication neighbor target set for a new sensing target, only sensing targets in neighboring areas that are judged to be duplicate targets within the historical deduplication window can be added to the deduplication neighbor target set and participate in the deduplication calculation within the current deduplication window. This effectively deduplicates different sensing targets moving in parallel, significantly improving the accuracy and effectiveness of deduplication.

[0163] In an exemplary embodiment, based on the above conditions, after a new sensing target appears in a sensing cell, triggering deduplication processing, within that sensing cell, for the new sensing target, a set of deduplicated neighboring target cells needs to be determined based on overlapping sensing relationships, identity status, and historical deduplication results. Within the current deduplication window, the new sensing target will form a target pair with each neighboring sensing target in the final deduplicated neighboring target set, and deduplication calculations will be performed on this target pair. The new sensing target and targets outside the final deduplicated neighboring target set are, by default, non-duplicate target relationships.

[0164] In one example, the principle for determining the final deduplicated neighbor set is described as follows:

[0165] Only targets within neighboring sensing cells that have overlapping sensing areas with the current sensing cell can be considered as neighboring sensing targets, forming a deduplicated set of neighboring targets 1.

[0166] In the deduplication neighbor target set 1, only the neighboring sensing cell and the sensing cell have overlapping sensing areas, and only sensing targets located within the overlapping sensing areas can be considered as neighbor sensing targets. Based on the deduplication neighbor target set 1, the deduplication neighbor target set 2 is further defined.

[0167] In the deduplication neighbor target set 2, only the perceived targets whose identity status is independent or newly added can be used as neighbor perceived targets. Based on the deduplication neighbor target set 2, the deduplication neighbor target set 3 is further defined.

[0168] To ensure the uniqueness of the deduplication calculation, a sensing target with the identity status of a newly added target can only be considered as a neighboring sensing target if the following conditions are met: The first sensing time of the new sensing target triggering deduplication processing in this sensing cell is set to t0 (the sensing time is also the sensing timestamp, which will not be explained further in the following embodiments), and the first sensing time of the sensing target with the identity status of a newly added target in the deduplication neighboring target set 2 is t0. neighcell The limiting principle is described as follows:

[0169] (1) t neighcell Greater than t0;

[0170] (2) t neighcell The new target is equal to t0 and meets the preset new target constraint conditions. The purpose of these preset new target constraint conditions is to ensure that the deduplication calculation between new targets is performed only in one sensing cell, avoiding inconsistencies in calculation time / results between different cells, and guaranteeing the uniqueness of the deduplication calculation between new target pairs. For example, the new target constraint condition is set to the condition that the number of the sensing target with the identity status of a new target in the deduplication neighbor target set 2 is greater than the number of the new sensing target in the current sensing cell, etc. In this way, the deduplication neighbor target set 3 is further constrained to form the deduplication neighbor target set 4.

[0171] Within the current deduplication window, if there are historical deduplication judgment results, only the perception targets whose historical deduplication judgment results are duplicate targets with the new perception targets in the current perception cell can be used as neighboring perception targets. Thus, based on the deduplication neighboring target set 4, a deduplication neighboring target set 5 is further defined, which can be used as the final deduplication neighboring target set.

[0172] In one exemplary embodiment, reference is made to Figure 3 As shown, the deduplication process for new perceived targets may further include steps 301 to 304, wherein:

[0173] In step 301, if the set of deduplicated neighboring targets of the new sensing target is not empty, the new sensing target is paired with each of the neighboring sensing targets in the set of deduplicated neighboring targets.

[0174] Taking the above embodiment as an example, assuming that the set of deduplicated neighboring targets of the perceived target A includes the perceived target B and the perceived target E, the perceived target A can be paired with each of the neighboring perceived targets in the set of deduplicated neighboring targets to obtain the target pairs: (perceived target A, perceived target B) and (perceived target A, perceived target E).

[0175] In step 302, for any pair of targets, the distance between the two targets at the same sensing time is calculated within the sensing time overlap window, and deduplication judgment is performed based on the distance to obtain the deduplication judgment result.

[0176] In this embodiment of the application, for any target pair, a perception time overlap window can be determined within the current deduplication window. The perception time overlap window is the time range in which the two target pairs have overlapping perception data. For example, if the perception time range of the perceived target A is t0~t10 and the perception time range of the perceived target B is t1~t12, then the perception time overlap window can be t1~t10.

[0177] Within the overlapping window of perception time, the distance between two perceived targets can be determined based on their perception data, yielding the distance between the two targets at each perception time. Deduplication is then performed based on this distance to obtain a deduplication result. For example, the deduplication process may include: for any given perception time, if the distance between the two perceived targets is greater than a preset threshold, they are determined to be non-repeating targets at that time. If the proportion of perception times where a non-repeating target relationship is determined to be greater than or equal to a preset proportion threshold, the two perceived targets are determined to be non-repeating targets, resulting in a deduplication result indicating that the two perceived targets in the target pair are non-repeating targets. Conversely, if the proportion of perception times where a repeating target relationship is determined to be less than a preset proportion threshold, the two perceived targets are determined to be repeating targets, resulting in a deduplication result indicating that the two perceived targets in the target pair are repeating targets.

[0178] Calculating the distance between two sensing targets at the same sensing moment within the sensing time overlap window may include the following steps:

[0179] For each sensing time of one of the sensing targets in a target pair, the position information of the other sensing target at each sensing time is determined by an interpolation algorithm; the distance between the position information of the two sensing targets at the same sensing time is determined.

[0180] In this embodiment of the application, since the sensing times of each sensing node for the same sensing target may be different, the sensing times of two sensing targets can be aligned by an interpolation algorithm, thereby determining the distance between the location information of the two sensing targets at the same sensing time.

[0181] For example, within the overlapping perception time window, assuming the perception times of target A include 0ms, 40ms, 80ms, ..., and the perception times of target B include 5ms, 45ms, 85ms, ..., then the position information of target B at perception time 40ms can be obtained by interpolating the perception data of target B at perception times 5ms and 45ms. Furthermore, based on the position information of target A and target B at perception time 40ms, the distance between them can be calculated. Similarly, the distance between target A and target B at each perception time can be obtained.

[0182] In step 303, the number of consecutive target occurrences is updated based on the deduplication judgment result, and the current deduplication window length is updated based on the updated number of consecutive target occurrences. The deduplication window length before the update is used as the historical deduplication window length. The number of consecutive target occurrences is the number of times that a deduplication judgment result representing the target pair as a duplicate target relationship is obtained consecutively.

[0183] In this embodiment of the application, after obtaining the deduplication judgment result, the number of times the deduplication judgment result representing the target pair as duplicate target relationship can be obtained consecutively can be determined based on the deduplication judgment result (hereinafter referred to as the target consecutive number for clarity), including: when the deduplication judgment result represents the two perceived targets as duplicate target relationship, the target consecutive number is incremented by 1; when the deduplication judgment result represents the two perceived targets as not duplicate target relationship, the target consecutive number is reset to 0.

[0184] That is, assuming that the deduplication judgment result obtained in the previous deduplication window indicates that the target pair is a duplicate target relationship, if the deduplication judgment result obtained in the current deduplication window still indicates that the target pair is a duplicate target relationship, then the consecutive count of the target can be incremented by 1; or, if the deduplication judgment result obtained in the current deduplication window indicates that the target object is not a duplicate target relationship, then the consecutive count of the target can be reset to 0.

[0185] After updating the target consecutive count, the deduplication window length can be dynamically updated based on the updated target consecutive count to obtain the current deduplication window length, and the deduplication window length before the update can be used as the historical deduplication window length.

[0186] In an exemplary embodiment, updating the current deduplication window length based on the updated target consecutive count may include the following steps: determining a candidate window length based on the updated target consecutive count and the base window length; and taking the minimum value between the candidate window length and a preset window length threshold as the current deduplication window length.

[0187] In this embodiment, a base window length and a preset window length threshold can be pre-set, and a candidate window length can be determined based on the updated target consecutive count and the base window length. For example, the candidate window length = (target consecutive count + 1) × base window length. After obtaining the candidate window length, the minimum value between the candidate window length and the preset window length threshold can be used as the current deduplication window length.

[0188] In step 304, based on the relationship between the current deduplication window length and the historical deduplication window length, it is determined whether to end the deduplication process. When it is determined to end the deduplication process, the deduplication result of the target pair is determined based on the number of consecutive target occurrences.

[0189] In this embodiment, the purpose of dynamically updating the deduplication window length is to extend the time window length for deduplication judgment in special scenarios, to adapt to the deduplication judgment in such special scenarios (e.g., deduplication of two different sensing targets moving in parallel). In an exemplary embodiment, determining whether to end the deduplication process based on the relationship between the current deduplication window length and the historical deduplication window length may include the following steps:

[0190] If the current deduplication window length is greater than the historical deduplication window length, continue the deduplication process for the newly perceived target; if the current deduplication window length is less than or equal to the historical deduplication window length, determine to end the deduplication process for the target pair.

[0191] In this embodiment, a preset window length threshold is set to the maximum window length of the deduplication window. When the current deduplication window length is greater than the historical deduplication window length, it indicates that the deduplication judgment result for the target pair is always a duplicate target relationship, but the deduplication process has not yet reached the maximum window length of the deduplication window. Therefore, it is necessary to perform the deduplication judgment of the next deduplication window based on the current deduplication window length, that is, to continue to execute the deduplication process for the new perceived target.

[0192] Conversely, if the current deduplication window length is less than or equal to the historical deduplication window length, it indicates that the current deduplication process has reached the maximum deduplication window length, but the judgment result of the target pair is still a duplicate target relationship. Alternatively, although the current deduplication process has not reached the maximum deduplication window length, it has already determined that the target pair is not a duplicate target relationship, so the deduplication process for that target pair can be terminated.

[0193] After the deduplication process of the target pair is completed, the deduplication result of the target pair can be determined according to the number of consecutive targets. This can include: if the number of consecutive targets is greater than 0, the two perceived targets in the target pair are considered to be repeating targets; if the number of consecutive targets is equal to 0, the two perceived targets in the target pair are considered to be non-repeating targets.

[0194] In this embodiment of the application, if the current consecutive count of the target is greater than 0 after the deduplication process of the target pair ends, it means that the deduplication judgment result determined in each deduplication window in the deduplication process is a duplicate target relationship. Therefore, it can be determined that the two perceived targets in the target pair are duplicate target relationships, and a deduplication result representing that the two are duplicate target relationships can be obtained. Conversely, if the current consecutive count of the target is equal to 0, it means that the deduplication judgment result determined in the last deduplication window in the deduplication process is a non-duplicate target relationship. That is, the two perceived targets in the target pair are two different perceived targets that have been running side by side for a period of time. Therefore, a deduplication result representing that the two are non-duplicate target relationships can be obtained.

[0195] In this way, for different sensing targets moving in parallel, the embodiments of this application can adaptively perform multiple deduplication judgments on the sensing targets by dynamically adjusting the window length of the deduplication window, which can improve the deduplication accuracy of the sensing targets in this scenario.

[0196] In one example embodiment, during the deduplication process within any deduplication window, a deduplication judgment based on distance can be triggered to obtain the deduplication result if any of the following conditions are met:

[0197] (1) The current deduplication window ends;

[0198] (2) The time of overlap of the trajectories of the two perceived targets in the target pair no longer changes;

[0199] (3) The cumulative number of times the distance between two perceived targets in a target pair exceeds a preset distance threshold exceeds a preset number threshold;

[0200] (4) The number of sensing moments used to calculate the distance between two sensing targets in the target pair is greater than a preset threshold;

[0201] (5) The interaction state of any perceived target in the target pair is an abnormal state;

[0202] (6) The difference between the timestamp of the latest reported sensing data of the new sensing target and the timestamp of the first reported sensing data of the new sensing target is greater than the preset allowable sensing delay.

[0203] In this embodiment of the application, when condition (1) is met, deduplication judgment can be performed; when any of the conditions (2) to (6) are met, deduplication judgment can be performed in advance within the deduplication window, thereby improving deduplication efficiency and improving the timeliness of reporting the perception data of the perception target.

[0204] The following explains conditions (2) to (6):

[0205] Regarding condition (2), the time of overlap between the trajectories of the two sensing targets no longer changes. This can include situations such as either sensing target disappearing or the two sensing targets moving to a sensing cell that does not have a sensing overlap area. In such cases, deduplication judgment can be triggered in advance.

[0206] Regarding condition (3), the preset distance threshold is a pre-set distance value. When the distance between two perceived targets is greater than the preset distance threshold at any perception moment, the number of times can be accumulated. When the accumulated number of times is greater than the preset number of times threshold, the two are likely to be non-repeating target relationships. At this time, the deduplication judgment can be triggered in advance.

[0207] Regarding condition (4), when the number of sensing moments used to calculate the distance between two sensing targets in a target pair is greater than the preset threshold, it indicates that the sensing data corresponding to the two sensing targets is relatively rich, and deduplication judgment can be triggered in advance.

[0208] Regarding condition (5), abnormal states may include situations where any sensing target cannot be perceived and thus is judged to have disappeared. When the sensing target disappears, there will be no new sensing data interaction, so deduplication judgment can be triggered in advance.

[0209] For condition (6), if the difference between the timestamp of the latest reported sensing data of the new sensing target and the timestamp of the first reported sensing data of the new sensing target is greater than the preset allowed sensing delay, then the deduplication judgment needs to be triggered in advance to avoid problems such as excessive delay in reporting sensing data for the new sensing target.

[0210] By using the above conditions (2) to (6), deduplication can be triggered in advance within the deduplication window, which can effectively improve deduplication efficiency and reduce the reporting delay of new target perception data.

[0211] To enable those skilled in the art to better understand the embodiments of this application, the embodiments of this application are described below through specific examples.

[0212] Example 1: Reference Figure 4As shown, assume that sensing cell 1 and sensing cell 2 have overlapping sensing areas. A drone target is in motion. The target first appears in the independent sensing range of sensing cell 1, then moves to the overlapping sensing area of ​​sensing cell 1 and the adjacent sensing cell 2. The target continues to move out of the sensing range of sensing cell 1 and into the independent sensing range of sensing cell 2. Assume the base window length is W. base W max Preset window length threshold: 3*W base .

[0213] (1) Sensing cell 1 at sensing time t 01 The target is detected for the first time. Within sensing cell 1, the target's status will be initialized to "new target," and its detected target number in sensing cell 1 will be recorded as ID1-0 (hereinafter referred to as sensing target ID1-0 for clarity). This sensing target ID1-0 will trigger the deduplication process. Because sensing cell 2 has not detected any targets at this time, the deduplication neighbor target set for sensing target ID1-0 is empty. That is, sensing target ID1-0 has no duplicate target relationship with any neighboring targets. Therefore, the status of sensing target ID1-0 will change from "new target" to "independent target."

[0214] (2) Sensing cell 1 continuously senses the sensing target ID1-0 and is responsible for reporting the sensing data of the sensing target ID1-0 to the next-level node. The reported ID is ID1-0. report .

[0215] (3) When the target moves to the overlapping sensing area of ​​sensing cell 1 and the adjacent sensing cell 2, sensing cell 2 at time t 02 The target is detected for the first time. Within sensing cell 2, the target's identity status will be initialized to "new target", and its target number in sensing cell 2 will be recorded as ID2-0 (hereinafter referred to as sensing target ID2-0 for clarity). The sensing target ID2-0 will trigger the deduplication process.

[0216] (4) The set of deduplicated neighboring targets corresponding to the perceived target ID2-0 is {perceived target ID1-0}, forming a deduplicated target pair (perceived target ID2-0, perceived target ID1-0).

[0217] (5) For the deduplication target pair (perceived target ID2-0, perceived target ID1-0), the historical deduplication result is the number of consecutive times the target has a duplicate target relationship, which is initialized to N. repeat The value is 0, and the length of the historical deduplication window is 0.

[0218] (6) For the deduplication target pair (perceived target ID2-0, perceived target ID1-0), the current deduplication window starts at time t. 02The end time is t 02 +W base The current deduplication window size is larger than the historical deduplication window size, so the deduplication process needs to continue, and the historical deduplication window size needs to be updated to W. base .

[0219] (7) Assuming that the deduplication judgment is triggered by condition (1) (the end time of the current deduplication window), if the output deduplication result is a duplicate target relationship, then update N. repeat The value is 1.

[0220] (8) The current deduplication window length is updated to 2*W base The current deduplication window starts at time t. 02 The end time is t 02 +2*W base The current deduplication window size is larger than the historical deduplication window size, so the deduplication process needs to continue, and the historical deduplication window size needs to be updated to 2*W. base .

[0221] (9) Assuming that the deduplication process is triggered by condition (1) (the end time of the current deduplication window), if the output deduplication result is a duplicate target relationship, then update N. repeat The value is 2.

[0222] (10) The current deduplication window length is updated to 3*W base The current deduplication window starts at time t. 02 The end time is t 02 +3*W base The current deduplication window size is larger than the historical deduplication window size, so the deduplication process needs to continue, and the historical deduplication window size needs to be updated to 3*W. base .

[0223] (11) Assuming that the deduplication judgment is triggered by condition (1) (the end time of the current deduplication window), if the output deduplication result is a duplicate target relation, then update N. repeat The value is 3.

[0224] (12) The current deduplication window length is updated to 3*W base The current deduplication window starts at time t. 02 The end time is t 02 +3*W base The current deduplication window length equals the historical deduplication window length, so the deduplication process will end. This is because at this point N... repeat Since it is not equal to 0, we obtain the deduplication result that represents the relationship between perceived target ID2-0 and perceived target ID1-0 as duplicate targets.

[0225] (13) Based on the deduplication result, update the identity status of sensing target ID2-0 in sensing cell 2 to "non-independent target", and sensing cell 2 will not report the sensing data of sensing target ID2-0. Sensing target ID1-0 in sensing cell 1 can put sensing target ID2-0 in sensing cell 2 into its duplicate target list.

[0226] (14) When the target continues to move out of the sensing range of sensing cell 1 and enters the independent sensing range of sensing cell 2, sensing cell 1 will be unable to detect the target and will determine that the target ID1-0 has disappeared. Then, the target ID2-0 in sensing cell 2 will be selected as the next data reporting anchor point.

[0227] (15) Sensing cell 2 updates the identity status of sensing target ID2-0 to "independent target" and uses the reported ID (ID report Then continue to report the perceived data.

[0228] Example 2: Reference Figure 5 As shown, assume that sensing cell 1 and sensing cell 2 have overlapping sensing areas. Within sensing cell 1, there exists a sensing target ID1-0 with an independent identity state within its independent sensing range. Figure 5 In sensing cell 2, within its independent sensing range, there are sensing targets with the identity status of independent targets (identified as independent targets), sensing target ID1-1 with the identity status of independent targets, and sensing target ID1-2 with the identity status of independent targets. Figure 5 (The perceived target whose identity status is newly added is identified as a newly added target).

[0229] (1) Sensing cell 2 at time t 02 Upon first detection of a target, its identity status is initialized to "new target". Its target number in sensing cell 2 is recorded as ID2-2 (hereinafter referred to as sensing target ID2-2 for clarity). This sensing target ID2-2 will trigger the deduplication process.

[0230] (2) Since the sensing target ID1-0, sensing target ID1-1 and sensing target ID1-2 are not in the overlapping sensing area of ​​sensing cell 2, the set of deduplicated neighboring targets corresponding to sensing target ID2-2 is empty.

[0231] (3) The deduplication process for sensing target ID2-2 ends. Since there is no duplicate target relationship between sensing target ID2-2 and all neighboring sensing targets, the identity status will change from "new target" to "independent target". Sensing cell 2 continues to sense sensing target ID2-2 and is responsible for reporting the sensing data of sensing target ID2-2 to the next higher level node.

[0232] Example 3: Reference Figure 6 As shown, assume that sensing cell 1 and sensing cell 2 have overlapping sensing areas, and a drone takes off within the overlapping sensing area. Sensing cell 1 and sensing cell 2 will detect the target approximately simultaneously. Let sensing cell 1 at time t... 01 The target was first detected in sensing cell 2 at time t. 02 The target was first perceived, and t 01 Slightly less than t 02 .

[0233] For sensing cell 1:

[0234] (1) Within the sensing cell 1, the identity status of the target will be initialized to "new target", and its target number in the sensing cell 1 will be recorded as ID1-0 (hereinafter referred to as sensing target ID1-0 for clarity). The sensing target ID1-0 will trigger the deduplication process.

[0235] (2) The first sensing time t of the sensing target ID1-0 in sensing cell 1 01 The first sensing time t of the sensing target ID2-0 within sensing cell 2 is smaller than the target ID2-0. 02 Therefore, the set of deduplicated neighboring targets corresponding to the newly added target ID1-0 in sensing cell 1 is {sensing target ID2-0}, forming a deduplicated target pair (sensing target ID1-0, sensing target ID2-0).

[0236] (3) Perform a deduplication process on the target pair (perceived target ID1-0, perceived target ID2-0), and the deduplication result is the duplicate target relationship.

[0237] (4) Based on the deduplication result, update the identity status of the sensing target ID1-0 in sensing cell 1 to "non-independent target", and sensing cell 1 does not report the sensing data of the sensing target ID1-0.

[0238] For Sensing Cell 2:

[0239] (1) In the sensing cell 2, the target status will be initialized to "new target", and its target number in the sensing cell 2 will be recorded as ID2-0 (hereinafter referred to as sensing target ID2-0 for clarity). The sensing target ID2-0 will trigger the deduplication process.

[0240] (2) The first sensing time t of the sensing target ID2-0 in sensing cell 2 02 The first sensing time t of the sensing target ID 1-0 within sensing cell 1 is greater than the target ID 1-0. 01 Therefore, the set of deduplicated neighboring targets corresponding to the sensing target ID2-0 in sensing cell 2 is empty.

[0241] (3) The deduplication process for the perceived target ID2-0 ends. Since there is no duplicate target relationship between the perceived target ID2-0 and all neighboring perceived targets, the identity status of the perceived target ID2-0 will change from "new target" to "independent target".

[0242] (4) Sensing cell 2 is responsible for reporting the sensing data of the sensing target ID2-0.

[0243] Example 4: Reference Figure 7 As shown, assuming that sensing cell 1 and sensing cell 2 have a large overlapping sensing area, two drones take off within the overlapping sensing area and run parallel in the same direction for a short distance before separating. When the two targets are relatively close, sensing cell 1 and sensing cell 2 will basically detect one target simultaneously. Let sensing cell 1 at time t... 01 The target was first detected in sensing cell 2 at time t. 02 The target was first perceived, and t 01 Slightly less than t 02 Assume the base window length is W. base W max Preset window length threshold: 3*W base .

[0244] For sensing cell 1:

[0245] (1) Within the sensing cell 1, the identity status of the target will be initialized to "new target", and its target number in the sensing cell 1 will be recorded as ID1-0 (hereinafter referred to as sensing target ID1-0 for clarity). The sensing target ID1-0 will trigger deduplication.

[0246] (2) The first sensing time t of the sensing target ID1-0 in sensing cell 1 01 The first sensing time t of the sensing target ID2-0 within sensing cell 2 is smaller than the target ID2-0. 02 Therefore, the set of deduplicated neighboring targets corresponding to the perceived target ID1-0 in the sensing cell 1 is {perceived target ID2-0}, forming a deduplicated target pair (perceived target ID1-0, perceived target ID2-0).

[0247] (3) For the deduplication target pair (perceived target ID1-0, perceived target ID2-0), the historical deduplication result is the number of consecutive times the duplicate target relationship is initialized to N. repeat The value is 0, and the length of the historical deduplication window is 0.

[0248] (4) For the deduplication target pair (perceived target ID1-0, perceived target ID2-0), the current deduplication window starts at time t. 01 The end time is t 01 +W base The current deduplication window size is larger than the historical deduplication window size, so the deduplication process needs to continue, and the historical deduplication window size needs to be updated to W. base .

[0249] (5) Assuming that the deduplication process is triggered by condition (1) (the end time of the current deduplication window) within the current deduplication window, and the output deduplication result is a duplicate target relationship, then update N. repeat The value is 1.

[0250] (6) The current deduplication window length is updated to 2*W base The current deduplication window starts at time t. 01 The end time is t 01 +2*W base The current deduplication window size is larger than the historical deduplication window size, so the deduplication process needs to continue, and the historical deduplication window size needs to be updated to 2*W. base .

[0251] (7) Assuming that the deduplication process is triggered by condition (1) (the end time of the current deduplication window) within the current deduplication window, and the output deduplication result is a duplicate target relation, then update N. repeat The value is 2.

[0252] (8) The current deduplication window length is updated to 3*W base The current deduplication window starts at time t. 01 The end time is t 01 +3*W base The current deduplication window size is larger than the historical deduplication window size, so the deduplication process needs to continue, and the historical deduplication window size needs to be updated to 3*W. base .

[0253] (9) Assuming that the current deduplication window is triggered by condition (3) (the cumulative value of the Euclidean distance greater than the set high distance threshold HighDistanceTh is greater than the set threshold HighDistanceNumTh), and the output deduplication result is a non-repeating target relation, then N repeat Reset to 0.

[0254] (10) The current deduplication window length is updated to W. base The deduplication process will end if the current deduplication window length is less than the historical deduplication window length. This is because the current N... repeat Since the result equals 0, the deduplication result is determined to be a non-repeating target relation.

[0255] (11) Based on the deduplication result, update the identity status of the sensing target ID1-0 in sensing cell 1 to "independent target", and the sensing cell 1 reports the sensing data of the sensing target ID1-0.

[0256] For Sensing Cell 2:

[0257] (1) Within the sensing cell 2, the identity status of the target will be initialized to "new target", and its target number in the sensing cell 2 will be recorded as ID2-0 (hereinafter referred to as sensing target ID2-0 for clarity). The sensing target ID2-0 will trigger the deduplication process.

[0258] (2) The first sensing time t of the sensing target ID2-0 in sensing cell 2 02 The first sensing time t of the sensing target ID 1-0 within sensing cell 1 is greater than the target ID 1-0. 01 Therefore, the set of deduplicated neighboring targets corresponding to the sensing target ID2-0 in sensing cell 2 is empty.

[0259] (3) The deduplication process of the perceived target ID2-0 ends. The perceived target ID2-0 has no duplicate target relationship with any of the neighboring targets, so the identity status of the perceived target ID2-0 will change from "new target" to "independent target".

[0260] (4) Sensing cell 2 is responsible for reporting the sensing data of the sensing target ID2-0.

[0261] The target deduplication method provided in this application has the following advantages: First, by maintaining and updating the identity status of perceived targets; triggering deduplication based on the identity status of perceived targets and overlapping perception relationships; and determining a limited set of neighboring targets for deduplication based on overlapping perception relationships, the identity status of perceived targets, and historical deduplication results, the data processing volume for deduplication judgment can be effectively reduced, and the deduplication efficiency and accuracy can be improved. Second, the length of the deduplication window can be adaptively adjusted based on the deduplication judgment results within each deduplication window in the deduplication process, which can increase the accuracy of deduplication results in certain special scenarios (such as two drones flying in parallel for a long time and then separating). Third, deduplication judgment can be triggered in advance within the deduplication window based on preset events to obtain deduplication results, and the perceived data of perceived targets can be reported to the next-level node as soon as possible, reducing the initial perception latency index.

[0262] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0263] Based on the same inventive concept, this application also provides a target deduplication device for implementing the target deduplication method described above. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations of one or more target deduplication device embodiments provided below can be found in the limitations of the target deduplication method above, and will not be repeated here.

[0264] In one exemplary embodiment, such as Figure 8 As shown, a target deduplication device 800 is provided, including: a deduplication module 810 and an identity status change module 820, wherein:

[0265] The deduplication module 810 is used to set the identity status of the new sensing target to the newly added target when a new sensing target is detected, and to trigger the deduplication process for the new sensing target.

[0266] The identity status change module 820 is used to update the identity status of the new perceived target to a non-independent target if the deduplication result indicates that the new perceived target has a neighboring perceived target with a duplicate target relationship; and to update the identity status of the new perceived target to an independent target if the deduplication result indicates that the new perceived target does not have a neighboring perceived target with a duplicate target relationship.

[0267] Among them, the sensing cell where the independent target is located is responsible for reporting sensing data, while the sensing cell where the non-independent target and the newly added target are located is not responsible for reporting sensing data.

[0268] In one embodiment, the deduplication module 810 is specifically used for:

[0269] Determine the set of deduplicated neighboring targets for the new sensing target;

[0270] If the set of deduplicated neighboring targets for the new sensing target is empty, then the deduplication result is obtained, which indicates that the new sensing target does not have any neighboring sensing targets with duplicate target relationships.

[0271] In one embodiment, the neighboring sensing targets in the deduplicated neighboring target set satisfy the following condition: the sensing cell where the neighboring sensing target is located has an overlapping sensing area with the sensing cell where the new sensing target is located.

[0272] In one embodiment, the conditions that the neighboring cell sensing target needs to meet also include: the identity status of the neighboring cell sensing target is an independent target or a newly added target.

[0273] In one embodiment, when the identity status of a neighboring target is a newly added target, the neighboring target also needs to meet the following conditions:

[0274] The first perception timestamp of the neighboring target is greater than the first perception timestamp of the new target; or, the first perception timestamp of the neighboring target is equal to the first perception timestamp of the new target, and the neighboring target and the new target satisfy the conditions for adding a new target.

[0275] In one embodiment, the conditions that the neighboring cell sensing target needs to meet also include:

[0276] If there are historical deduplication judgment results in the current deduplication window, the neighboring sensing target and the new sensing target are considered to be duplicate targets in the historical deduplication judgment results.

[0277] In one embodiment, the deduplication module 810 is further configured to:

[0278] If the set of deduplicated neighboring targets of the new sensing target is not empty, the new sensing target will be paired with each of the neighboring sensing targets in the set of deduplicated neighboring targets.

[0279] For any pair of targets, the distance between the two targets at the same sensing moment is calculated within the sensing time overlap window, and deduplication is performed based on the distance to obtain the deduplication result;

[0280] Update the target consecutive count based on the deduplication judgment result, and update the current deduplication window length based on the updated target consecutive count. Use the deduplication window length before the update as the historical deduplication window length. The target consecutive count is the number of times the deduplication judgment result representing the target pair as a duplicate target relationship is obtained consecutively.

[0281] Based on the relationship between the current deduplication window length and the historical deduplication window length, determine whether to end the deduplication process. When determining to end the deduplication process, determine the deduplication result of the target pair based on the number of consecutive occurrences of the target.

[0282] In one embodiment, the deduplication module 810 is further configured to:

[0283] If the current deduplication window length is greater than the historical deduplication window length, continue the deduplication process for the newly perceived target; if the current deduplication window length is less than or equal to the historical deduplication window length, determine to end the deduplication process for the target pair.

[0284] In one embodiment, the deduplication module 810 is further configured to:

[0285] When the deduplication result indicates that the two perceived targets are duplicate targets, increment the number of consecutive targets by 1; when the deduplication result indicates that the two perceived targets are not duplicate targets, reset the number of consecutive targets to 0.

[0286] In one embodiment, the deduplication module 810 is further configured to:

[0287] Based on the updated target consecutive count and the base window length, determine the candidate window length; the minimum value between the candidate window length and the preset window length threshold is taken as the current deduplication window length.

[0288] In one embodiment, the deduplication module 810 is further configured to:

[0289] If the number of consecutive targets is greater than 0, the result is a deduplication of the two perceived targets in the target pair as repeating targets; if the number of consecutive targets is equal to 0, the result is a deduplication of the two perceived targets in the target pair as non-repeating targets.

[0290] In one embodiment, a deduplication check based on distance is triggered and a deduplication result is obtained when any of the following conditions are met:

[0291] The current deduplication window ends; the trajectory overlap time of the two sensing targets in the target pair no longer changes; the cumulative number of times the distance between the two sensing targets in the target pair exceeds the preset distance threshold exceeds the preset number threshold; the number of sensing moments used to calculate the distance between the two sensing targets in the target pair exceeds the preset number threshold; the interaction state of any sensing target in the target pair is abnormal; the difference between the timestamp of the latest reported sensing data of the new sensing target and the timestamp of the first reported sensing data of the new sensing target exceeds the preset allowed sensing delay.

[0292] In one embodiment, the deduplication module 810 is further configured to:

[0293] For each sensing time of one of the sensing targets in a target pair, the position information of the other sensing target at each sensing time is determined by an interpolation algorithm; the distance between the position information of the two sensing targets at the same sensing time is determined.

[0294] The target deduplication device provided in this application only reports sensing data for sensing targets whose identity status is independent. By performing a deduplication process on new sensing targets, and obtaining the deduplication result indicating whether there are neighboring sensing targets with duplicate target relationships with the new sensing target, the identity status of the sensing target is updated to independent target or non-independent target based on the deduplication result. Target deduplication can be achieved from the sensing data reporting stage, ensuring the uniqueness of the sensing data reporting, thereby ensuring the uniqueness of the sensing targets and the continuity of sensing services.

[0295] It should be noted that the division of units in the embodiments of this application is illustrative and only represents one logical functional division. In actual implementation, other division methods may be used. Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated units described above can be implemented in hardware or as software functional units.

[0296] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a processor-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods of the various embodiments of this application.

[0297] It should be noted that the apparatus provided in this application embodiment can implement all the method steps implemented in the above method embodiment and can achieve the same technical effect. Here, the parts that are the same as those in the method embodiment and the beneficial effects will not be described in detail.

[0298] In one exemplary embodiment, a target deduplication device is provided, the structure of which can be as follows: Figure 9 As shown. The base station includes a memory 920, a transceiver 910, and a processor 900.

[0299] A transceiver is used to receive and send data under the control of a processor.

[0300] Among them, Figure 9In this context, the bus architecture can include any number of interconnected buses and bridges, specifically linking various circuits together, represented by one or more processors (represented by a processor) and memory (represented by memory). The bus architecture can also link together various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and therefore will not be described further herein. The bus interface provides the interface. A transceiver can be multiple components, including transmitters and receivers, providing a unit for communicating with various other devices over transmission media, including wireless channels, wired channels, optical fibers, etc. The processor is responsible for managing the bus architecture and general processing, while the memory can store data used by the processor during operation.

[0301] The processor can be a central processing unit (CPU), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or a complex programmable logic device (CPLD). The processor can also adopt a multi-core architecture.

[0302] The processor executes any of the methods described above in the embodiments of this application according to the obtained executable instructions by calling a program stored in memory. The processor and memory may also be physically separated.

[0303] It should be noted that the apparatus provided in this application embodiment can implement all the method steps implemented in the above method embodiment and can achieve the same technical effect. Here, the parts that are the same as those in the method embodiment and the beneficial effects will not be described in detail.

[0304] In one exemplary embodiment, a target deduplication device is provided. The target deduplication device may be a terminal device, a base station, or a centralized processing node, etc., and includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0305] In one exemplary embodiment, a processor-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above-described method embodiments.

[0306] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0307] Processor-readable storage media can be any available medium or data storage device that the processor can access, including but not limited to magnetic storage (such as floppy disks, hard disks, magnetic tapes, magneto-optical disks (MOs), etc.), optical storage (such as CDs, DVDs, BDs, HVDs, etc.), and semiconductor storage (such as ROMs, EPROMs, EEPROMs, non-volatile memory (NAND flash), solid-state drives (SSDs)).

[0308] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage) containing computer-usable program code.

[0309] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-executable instructions. These computer-executable instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0310] These processor-executable instructions may also be stored in a processor-readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the processor-readable memory produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0311] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for deduplicating targets, characterized in that, include: When a new target is detected, the identity status of the new target is set to a new target, and a deduplication process for the new target is triggered. If the deduplication result indicates that the new perceived target has a neighboring perceived target with a duplicate target relationship, then the identity status of the new perceived target is updated to a non-independent target. If the deduplication result indicates that the new perceived target does not have any neighboring perceived targets with a duplicate target relationship, then the identity status of the new perceived target is updated to an independent target; The sensing cell where the independent target is located is responsible for reporting sensing data, while the sensing cells where the non-independent target and the newly added target are located are not responsible for reporting sensing data.

2. The method according to claim 1, characterized in that, The deduplication process for the new perceived target includes: Determine the set of deduplicated neighboring targets of the new perceived target; If the set of deduplicated neighboring targets for the new sensing target is empty, then the deduplication result is obtained, which indicates that the new sensing target does not have any neighboring sensing targets with which it has a duplicate target relationship.

3. The method according to claim 2, characterized in that, The neighbor sensing targets in the deduplicated neighbor target set satisfy the following conditions: The sensing cell where the neighboring sensing target is located has an overlapping sensing area with the sensing cell where the new sensing target is located.

4. The method according to claim 3, characterized in that, The conditions that the neighboring cell sensing target needs to meet also include: The identity status of the neighboring target is either an independent target or a newly added target.

5. The method according to claim 4, characterized in that, When the identity status of the neighboring cell sensing target is a newly added target, the neighboring cell sensing target also needs to meet the following conditions: The first perception timestamp of the neighboring target is greater than the first perception timestamp of the new target; or, the first perception timestamp of the neighboring target is equal to the first perception timestamp of the new target, and the neighboring target and the new target satisfy the conditions for adding a new target.

6. The method according to claim 5, characterized in that, The conditions that the neighboring cell sensing target needs to meet also include: If there is a historical deduplication judgment result in the current deduplication window, the neighboring sensing target and the new sensing target are considered to be duplicate targets in the historical deduplication judgment result.

7. The method according to any one of claims 2 to 6, characterized in that, The deduplication process for the new perceived target further includes: If the set of deduplicated neighboring targets of the new sensing target is not empty, the new sensing target is paired with each of the neighboring sensing targets in the set of deduplicated neighboring targets. For any of the target pairs, the distance between the two perceived targets at the same perceived moment is calculated within the perception time overlap window, and deduplication judgment is performed based on the distance to obtain the deduplication judgment result; Update the target consecutive count based on the deduplication judgment result, and update the current deduplication window length based on the updated target consecutive count. Use the deduplication window length before the update as the historical deduplication window length. The target consecutive count is the number of times the deduplication judgment result representing the target pair as a duplicate target relationship is obtained consecutively. Based on the relationship between the current deduplication window length and the historical deduplication window length, it is determined whether to end the deduplication process. When it is determined to end the deduplication process, the deduplication result of the target pair is determined based on the number of consecutive target occurrences.

8. The method according to claim 5, characterized in that, The step of determining whether to end the deduplication process based on the relationship between the current deduplication window length and the historical deduplication window length includes: When the current deduplication window length is greater than the historical deduplication window length, the deduplication process for the new sensing target continues. When the current deduplication window length is less than or equal to the historical deduplication window length, the deduplication process for the target pair is terminated.

9. The method according to claim 7, characterized in that, The step of updating the target consecutive count based on the deduplication judgment result includes: When the deduplication judgment result indicates that the two perceived targets are repeated targets, increment the number of consecutive targets by 1; When the deduplication judgment result indicates that the two perceived targets are not duplicate targets, the consecutive count of the target is set to 0.

10. The method according to claim 9, characterized in that, The step of updating the current deduplication window length based on the updated target consecutive count includes: Based on the updated target consecutive count and the base window length, determine the candidate window length; The minimum value between the candidate window length and the preset window length threshold is used as the current deduplication window length.

11. The method according to claim 7, characterized in that, Determining the deduplication result of the target pair based on the consecutive count of the target includes: If the number of consecutive occurrences of the target is greater than 0, then the deduplication result representing the relationship between the two perceived targets in the target pair as duplicate targets is obtained; If the number of consecutive occurrences of the target is equal to 0, then the deduplication result representing the relationship between the two perceived targets in the target pair as non-repeating targets is obtained.

12. The method according to claim 7, characterized in that, The deduplication process based on the distance is triggered when any of the following conditions are met, and the deduplication result is obtained: The current deduplication window ends at the specified time. The overlap time of the trajectories of the two sensed targets in the target pair no longer changes; The cumulative number of times the distance between two perceived targets in the target pair is greater than a preset distance threshold is greater than a preset number threshold. The number of sensing moments used to calculate the distance between two sensing targets in the target pair is greater than a preset threshold; The interaction state of any of the perceived targets in the target pair is an abnormal state; The difference between the timestamp of the latest reported sensing data from the new sensing target and the timestamp of the first reported sensing data from the new sensing target is greater than the preset allowed sensing delay.

13. The method according to claim 7, characterized in that, The calculation of the distance between two sensing targets at the same sensing moment within the sensing time overlap window includes: For each sensing time of one of the sensing targets in the target pair, the position information of the other sensing target at each sensing time is determined by an interpolation algorithm; Determine the distance between the location information of two sensing targets at the same sensing moment.

14. A target deduplication device, characterized in that, include: The deduplication module is used to set the identity status of the new sensing target to a newly added target when a new sensing target is detected, and to trigger a deduplication process for the new sensing target. The identity status change module is used to update the identity status of the new sensing target to a non-independent target if the deduplication result indicates that the new sensing target has a neighboring sensing target with a duplicate target relationship. Furthermore, if the deduplication result indicates that the new perceived target does not have any neighboring perceived targets with a duplicate target relationship with it, then the identity status of the new perceived target is updated to an independent target; The sensing cell where the independent target is located is responsible for reporting sensing data, while the sensing cells where the non-independent target and the newly added target are located are not responsible for reporting sensing data.

15. A target deduplication device, characterized in that, The target deduplication device includes: a memory, a transceiver, and a processor. A memory for storing a computer program; a transceiver for sending and receiving data under the control of the processor; and a processor for reading the computer program from the memory and executing the steps of the method according to any one of claims 1 to 13.

16. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 13.

Images