Target tracking method and device, equipment and medium

Abstract

The invention relates to a target tracking method, device and equipment and a medium, and the method comprises the steps: for each observation moment, obtaining an observation target at the observation moment and the position information of the observation target in a global coordinate system; wherein the global coordinate system is divided into a plurality of grids; performing position matching with the plurality of grids according to the position information, and determining a target grid where the observation target is located from the plurality of grids; determining a tracker corresponding to the observation target in the target grid; and generating a target tracking result of the observation target according to the trackers at the plurality of observation moments. According to the technical scheme of the invention, the method can solve a problem of target jump caused by a matching error, and improves the accuracy of multi-target tracking.

Description

Technical Field

[0001] This disclosure relates to the field of data processing technology, and in particular to a target tracking method, apparatus, device, and medium. Background Technology

[0002] Currently, multi-target tracking tasks are divided into two approaches: tracking models and traditional tracking algorithms. Tracking models take raw signals such as images as input, while traditional tracking algorithms typically take target perception results as input. Due to the interpretability and debuggability of traditional algorithms, a significant number of projects currently employ them.

[0003] In related technologies, traditional tracking algorithms include two main processes: matching and updating. Matching algorithms typically use nearest neighbor matching or Hungarian matching, while updating algorithms typically use Kalman filtering or its variants. For tracking static targets that are densely distributed in spatial locations and whose target perception results can change, the matching algorithm may result in matching errors, leading to changes in the same target identifier. Therefore, the accuracy of target tracking needs to be improved. Summary of the Invention

[0004] To address the aforementioned technical problems, this disclosure provides a target tracking method, apparatus, device, and medium.

[0005] In a first aspect, embodiments of this disclosure provide a target tracking method, including:

[0006] For each observation time, the observed target at that time and the position information of the observed target in the global coordinate system are obtained; wherein, the global coordinate system is divided into multiple grids;

[0007] Based on the location information, position matching is performed with the plurality of grids to determine the target grid where the observed target is located from the plurality of grids;

[0008] Determine the tracker corresponding to the observed target in the target grid;

[0009] Based on the trackers at multiple observation times, the target tracking results of the observed target are generated.

[0010] Secondly, embodiments of this disclosure provide a target tracking device, including:

[0011] The acquisition module is used to acquire, for each observation time, the observation target at the observation time and the position information of the observation target in the global coordinate system; wherein, the global coordinate system is divided into multiple grids;

[0012] A matching module is used to perform position matching with the plurality of grids based on the location information, and to determine the target grid where the observed target is located from the plurality of grids;

[0013] The determination module is used to determine the tracker corresponding to the observed target in the target grid;

[0014] The tracking module is used to generate target tracking results for the observed target based on trackers at multiple observation times.

[0015] Thirdly, embodiments of this disclosure provide an electronic device, including: a processor; a memory for storing executable instructions of the processor; the processor being configured to read the executable instructions from the memory and execute the instructions to implement the target tracking method described in the first aspect above.

[0016] Fourthly, embodiments of this disclosure provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the target tracking method described in the first aspect.

[0017] Compared with the prior art, the technical solution provided in this disclosure has the following advantages: For each observation time, the observed target and its position information in the global coordinate system are obtained at the observation time. Based on the position information, the target grid where the observed target is located is determined from multiple grids, so as to determine the tracker corresponding to the observed target in the target grid. Then, based on the trackers at multiple observation times, the target tracking result of the observed target is generated. Thus, by establishing multiple grids in the global coordinate system and determining the tracker by assigning the observed target to the grid, multi-target tracking of static targets is realized. In view of the problem of target jump caused by matching errors in the current traditional tracking algorithm, matching errors can be avoided and the accuracy of multi-target tracking can be improved. Attached Figure Description

[0018] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0019] To more clearly illustrate the technical solutions in the embodiments of this disclosure or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a schematic flowchart of a target tracking method provided in an embodiment of the present disclosure;

[0021] Figure 2This is a schematic diagram illustrating a matching error in a traditional matching algorithm.

[0022] Figure 3 This is a flowchart illustrating another target tracking method provided in an embodiment of the present disclosure;

[0023] Figure 4 This is a schematic diagram of a multi-target tracking process provided in an embodiment of the present disclosure;

[0024] Figure 5 This is a schematic diagram of the structure of a target tracking device provided in an embodiment of the present disclosure. Detailed Implementation

[0025] To better understand the above-mentioned objectives, features, and advantages of this disclosure, the solutions disclosed herein will be further described below. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0026] Numerous specific details are set forth in the following description in order to provide a full understanding of this disclosure, but this disclosure may also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only some, and not all, of the embodiments of this disclosure.

[0027] Figure 1 This is a flowchart illustrating a target tracking method provided in an embodiment of the present disclosure. The method provided in this embodiment can be executed by a target tracking device, which can be implemented in software and / or hardware and can be integrated into any electronic device with computing capabilities.

[0028] like Figure 1 As shown, the target tracking method provided in this disclosure embodiment may include:

[0029] Step 101: For each observation time, obtain the observation target at the observation time and the position information of the observation target in the global coordinate system.

[0030] The target tracking method of this disclosure is applied to a conventional tracking algorithm. The input of the conventional tracking algorithm is the target perception result, which includes, but is not limited to, the target detection result of the image, the point cloud data of the radar, etc.

[0031] In this embodiment, the observation target is the target that needs to be tracked, such as traffic cones placed on the road. For each observation time, the observation targets at that time are acquired through target perception, and then the position information of each observation target in the global coordinate system (World Coordinate System) is determined.

[0032] As an example, taking image target detection as an example, one frame of image corresponds to one observation time. For each observation time, target detection is performed on the image to obtain the observed target. Then, the observed target can be projected from the image coordinate system to the vehicle coordinate system to determine the observation position of the observed target in the vehicle coordinate system. Furthermore, the position coordinates of the observed target are transformed from the vehicle coordinate system to the global coordinate system.

[0033] In one embodiment of this disclosure, obtaining the observation target at the observation time and the position information of the observation target in the global coordinate system includes: obtaining the observation position of the observation target in the vehicle coordinate system, converting the observation position from the vehicle coordinate system to the global coordinate system, and determining the position information of the observation target in the global coordinate system based on the center point position corresponding to the observation position in the global coordinate system.

[0034] As an example, the target's position coordinates (x, y) are transformed from the vehicle coordinate system to the global coordinate system, yielding the target's position coordinates in the global coordinate system. Then, the target's center point is determined based on these global coordinates, serving as the target's position information in the global coordinate system. Thus, by using the target's center point as its position information in the global coordinate system, accurate matching between the target and the grid is achieved. It should be noted that the above method using the center point is just one example; other specified points can also be used to represent the target's position information in the global coordinate system, and no specific restrictions are imposed here.

[0035] Step 102: Match the location information with multiple grids to determine the target grid where the observed target is located from the multiple grids.

[0036] In this embodiment, the global coordinate system is divided into multiple grids. Optionally, the grid size is preset, and the global coordinate system is divided into multiple grids according to the size. The grid size can be set according to actual conditions, such as 1 meter, 0.5 meters, etc. The global coordinate system is considered as two-dimensional and compressed into the XY plane; the grid division is implemented based on the XY plane of the global coordinate system.

[0037] For each observation target, the target is matched with multiple grids based on its position information in the global coordinate system in order to determine the target grid from among the multiple grids.

[0038] As an example, taking the center point position of the observed target as an example, the target grid where the observed target is located is determined by the following calculation method: detGridX = floor(detCenterX / resolution)*resolution, detGridY = floor(detCenterY / resolution)*resolution, where resolution is the grid size, detCenterX is the X coordinate of the center point position, detCenterY is the Y coordinate of the center point position, floor is the floor function, detGridX is the X coordinate of the target grid, and detGridY is the Y coordinate of the target grid.

[0039] Step 103: Identify the tracker corresponding to the observed target in the target grid.

[0040] In this embodiment, for each observed target at the current observation time, a tracker corresponding to that observed target is determined in the target grid where the observed target is located, so as to obtain the tracker at the current observation time. The tracker persists for a certain duration.

[0041] As an example, determining the tracker corresponding to the observed target in the target grid includes: if the tracker does not exist in the target grid, creating a new tracker corresponding to the observed target in the target grid; if the tracker already exists in the target grid, updating the tracker in the target grid. In this example, all grids with existing trackers are traversed, and if the target grid matches any grid with an existing tracker, it is determined that a tracker already exists in the target grid. Optionally, if the tracker does not exist in the target grid, a new tracker is created based on the position information of the observed target; if the tracker already exists in the target grid, the position of the tracker is updated based on the position information of the observed target.

[0042] Step 104: Generate target tracking results for the observed target based on trackers at multiple observation times.

[0043] In this embodiment, for each observation time, the tracker at that observation time can be determined through the aforementioned steps. Based on this, target tracking of multiple observation targets can be achieved by using trackers at multiple observation times.

[0044] As an example, at time t1, cone 1 and cone 2 are detected, and trackers a and b are newly created in grid A and grid B, respectively. At observation times after time t1, it can be determined that tracker a already exists in grid A where cone 1 is located, so tracker a is updated; and it can be determined that tracker b already exists in grid B where cone 2 is located, so tracker b is updated. Thus, after multiple observation times, based on trackers a and b at each observation time, multi-target tracking of cone 1 and cone 2 is achieved.

[0045] Currently, for tracking static targets that are densely distributed in spatial locations and whose perception results exhibit certain abrupt changes, traditional matching algorithms suffer from matching errors, leading to the same identifier appearing as different targets. For example, when using the Hungarian matching algorithm, which searches for the maximum match in a bipartite graph, when targets are densely packed and there are observational fluctuations, errors may occur. Figure 2 The matching error shown causes large position jumps during attribute updates, resulting in inaccurate multi-target tracking results.

[0046] According to the technical solution of this disclosure, for each observation time, the observed target and its position information in the global coordinate system are obtained at that observation time. Based on the position information, the target grid where the observed target is located is determined from multiple grids. The tracker corresponding to the observed target is then determined in the target grid. Furthermore, based on the trackers at multiple observation times, the target tracking result of the observed target is generated. Thus, by establishing multiple grids in the global coordinate system and determining the tracker by assigning the observed target to the grid, multi-target tracking of static targets is achieved. This avoids the problem of target jump caused by matching errors in current traditional tracking algorithms and improves the accuracy of multi-target tracking.

[0047] Based on the above embodiments, Figure 3 This is a flowchart illustrating another target tracking method provided in an embodiment of the present disclosure, as shown below. Figure 3 As shown, the method includes:

[0048] Step 301: For each observation time, obtain the observation target at the observation time and the position information of the observation target in the global coordinate system.

[0049] Step 302: Determine the target grid where the observation target is located from multiple grids based on the location information, and determine the tracker corresponding to the observation target in the target grid.

[0050] In this embodiment, if there is no tracker in the target grid, a new tracker corresponding to the observed target is created in the target grid; if there is already a tracker in the target grid, the tracker in the target grid is updated.

[0051] Updating the trackers in the target grid includes updating the positions of the trackers in the target grid according to a preset update strategy. Optionally, a first position of the trackers in the target grid and a second position of the observed target are obtained. A third position is obtained by weighted summation based on the first position and a first weight, and the second position and a second weight. The first position is then updated to the third position. The first position is the historical position of the tracker, for example, the position of the tracker at the previous observation time, and the first weight is greater than the second weight.

[0052] As an example, the update strategy employs an exponential averaging approach, updating the tracker using the following calculations: trackPositionX = trackCenterX * 0.9 + detCenterX * 0.1, trackPositionY = trackCenterY * 0.9 + detCenterY * 0.1, where detCenterX is the X-coordinate of the center point, detCenterY is the Y-coordinate of the center point, trackCenterX is the X-coordinate of the tracker's historical position, trackCenterY is the Y-coordinate of the tracker's historical position, trackPositionX is the X-coordinate of the tracker's updated position, and trackPositionY is the Y-coordinate of the tracker's updated position. The first weight is 0.9, and the second weight is 0.1. Therefore, using an exponential averaging strategy to update the tracker's position reduces the impact of jumps on the tracking results, further improving target tracking accuracy.

[0053] Step 303: Perform verification processing for each tracker at the current observation time.

[0054] In this embodiment, the accuracy of target tracking is improved by verifying each tracker at the current observation time and deleting trackers that do not meet the conditions.

[0055] In one embodiment of this disclosure, the verification process includes deleting aging trackers. In this embodiment, for each tracker at the current observation time, the number of times the tracker is updated is obtained. Then, an aging threshold for the tracker is determined based on the number of updates. If the unupdated duration of any tracker exceeds the aging threshold, the tracker is deleted. In this embodiment, since trackers exist continuously for a certain duration, for each observation time, the trackers at that observation time include a first tracker determined at that observation time and a second tracker determined at historical observation times but not determined at that observation time. Taking the current observation time as an example, a first tracker corresponding to the observed target is determined in the target grid. The first tracker includes newly created trackers and updated trackers. A second tracker determined at several previous observation times but not corresponding to the observed target at the current observation time is obtained. The tracker at the current observation time is determined based on the first and second trackers. Therefore, aging trackers need to be deleted.

[0056] The update count represents the number of times the tracker is observed. For example, when a tracker already exists in the target grid and is updated, the update count of that tracker is incremented by one. The time without update can be determined based on the time interval between the current observation time and the last update time of the tracker. A pre-defined correspondence between the update count and the aging threshold is provided; optionally, the update count and the aging threshold are positively correlated.

[0057] As an example, for each observation time, the aging threshold of the tracker is dynamically determined based on the number of tracker updates. When the number of updates ∈ [1, 5], the aging threshold is 1 second; when the number of updates > 5, the aging threshold is 2 seconds. Therefore, by dynamically determining the aging threshold, aging trackers can be deleted more accurately.

[0058] In one embodiment of this disclosure, the verification process includes cross-verification of the trackers. In this embodiment, for multiple trackers at the current observation time, the distance between the multiple trackers is calculated, and a group of trackers with a distance less than a first distance threshold is determined from the multiple trackers. Each tracker group includes two trackers. Furthermore, if one of the two trackers has a corresponding observed target at the current observation time, and the other tracker does not have a corresponding observed target at the current observation time, then the other tracker is controlled to move a specified distance towards the first tracker. Further, if the other tracker is in the same grid as the first tracker after moving, the two trackers are merged. Therefore, in cases where the target perception result is jittery, when the position information of the observed target alternately falls in adjacent grids, multiple trackers may be created for the same observed target. Since these trackers are actually the same observed target, it is necessary to determine the above situation and control these trackers to move closer to each other during the update process, ultimately merging them into the same tracker, thereby avoiding the situation where one observed target is incorrectly output as multiple trackers and improving accuracy.

[0059] As an example, for the current observation time, after identifying the trackers in the target grid, the Euclidean distances between all trackers are calculated. If the Euclidean distance between any two trackers Trk1 and Trk2 is less than 0.7m, and Trk1 is a newly created or updated tracker at the current observation time, while Trk2 is neither updated nor newly created at the current observation time, then Trk2 is moved a certain distance towards Trk1. Optionally, the moving distance can be one-tenth of the Euclidean distance between Trk1 and Trk2. After Trk2 moves, if Trk2 moves to the grid where Trk1 is located, then Trk1 is deleted.

[0060] Figure 4 This is a schematic diagram of a multi-target tracking process provided in an embodiment of the present disclosure, with reference to... Figure 4 After deleting aging trackers and cross-checking the trackers, the system determines and outputs targets that have been tracked for a specified duration based on the trackers. Deduplication can also be performed before outputting the results.

[0061] In one embodiment of this disclosure, a target tracking result is generated based on trackers at multiple observation times. This includes: determining candidate observation targets based on trackers at multiple observation times; if the distance between any two candidate observation targets is less than a second distance threshold, detecting the distance between the two candidate observation targets and the vehicle, and deleting the candidate observation target that is farther from the vehicle from the two candidate observation targets. In this embodiment, the target tracking result is determined based on trackers in a global coordinate system. Candidate observation targets include targets that have been tracked for a specified duration. The second distance threshold can be set according to the actual application scenario, for example, a second distance threshold of 0.7m. After determining that the distance between two candidate observation targets is less than 0.7m, the distance between the two candidate observation targets and the vehicle is further determined, and the candidate observation target that is farther from the vehicle is deleted. After performing deduplication processing, the target tracking result is output. Deduplication can be implemented based on non-maximum suppression.

[0062] In this embodiment of the disclosure, the tracker is updated using an exponential averaging strategy to reduce the impact of jumps on the tracking results. The accuracy of target tracking is further improved by deleting aging trackers, performing cross-validation on trackers, and deduplication.

[0063] Figure 5 This is a schematic diagram of the structure of a target tracking device provided in an embodiment of the present disclosure, as shown below. Figure 5 As shown, the target tracking device includes: an acquisition module 51, a matching module 52, a determination module 53, and a tracking module 54.

[0064] The acquisition module 51 is used to acquire the observation target and the position information of the observation target in the global coordinate system at each observation time; wherein the global coordinate system is divided into multiple grids;

[0065] Matching module 52 is used to perform position matching with multiple grids based on location information, and to determine the target grid where the observed target is located from the multiple grids;

[0066] The determination module 53 is used to determine the tracker corresponding to the observed target in the target grid;

[0067] The tracking module 54 is used to generate target tracking results for the observed target based on the tracker at multiple observation times.

[0068] In one embodiment of this disclosure, the acquisition module 51 is specifically used for:

[0069] Obtain the observation position of the target in the vehicle coordinate system, and transform the observation position from the vehicle coordinate system to the global coordinate system;

[0070] The location information is determined based on the center point position corresponding to the observation position in the global coordinate system.

[0071] In one embodiment of this disclosure, the determining module 53 is specifically used for:

[0072] If no tracker exists in the target grid, a new tracker corresponding to the observed target will be created in the target grid.

[0073] If a tracker already exists in the target grid, update the tracker in the target grid.

[0074] In one embodiment of this disclosure, the determining module 53 is specifically used for:

[0075] Obtain the first position of the tracker in the target grid and observe the second position of the target;

[0076] The third position is obtained by weighting and summing the first position with its first weight and the second position with its second weight; where the first weight is greater than the second weight.

[0077] Update the first position to the third position.

[0078] In one embodiment of this disclosure, the device further includes:

[0079] The delete module is used to obtain the number of times a tracker has been updated for each tracker at the current observation time;

[0080] The aging threshold of the tracker is determined based on the number of times the tracker is updated;

[0081] If the tracker's unupdated duration exceeds the aging threshold, the tracker is deleted.

[0082] In one embodiment of this disclosure, the device further includes:

[0083] The merging module is used to calculate the distance between multiple trackers at the current observation time.

[0084] From a plurality of trackers, determine a combination of trackers whose distance is less than a first distance threshold; each combination of trackers includes two trackers;

[0085] If one of the two trackers has a corresponding observation target at the current observation time, and the other tracker does not have a corresponding observation target at the current observation time, then control the other tracker to move a specified distance toward the first tracker.

[0086] If another tracker is in the same grid as a tracker after moving, the two trackers are merged.

[0087] In one embodiment of this disclosure, the tracking module 54 is specifically used for:

[0088] Candidate observation targets are determined based on trackers at multiple observation times;

[0089] If the distance between any two candidate observation targets is less than the second distance threshold, then the distance between the two candidate observation targets and the vehicle is detected respectively.

[0090] Remove the candidate observation target that is farther away from the vehicle from the two candidate observation targets.

[0091] The target tracking device provided in this disclosure can execute any target tracking method provided in this disclosure, and has the corresponding functional modules and beneficial effects for executing the method. Content not described in detail in the device embodiments of this disclosure can be referred to the description in any method embodiment of this disclosure.

[0092] This disclosure also provides an electronic device including one or more processors and a memory. The processor may be a central processing unit (CPU) or other processing unit with data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions. The memory may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor may execute the program instructions to implement the methods of the embodiments of this disclosure above and / or other desired functions. Various contents such as input signals, signal components, and noise components may also be stored in the computer-readable storage medium.

[0093] In one example, the electronic device may also include input and output devices, which are interconnected via a bus system and / or other forms of connection. Furthermore, the input device may include, for example, a keyboard, a mouse, etc. The output device can output various information to the outside, including determined distance information, direction information, etc. The output device may include, for example, a display, a speaker, a printer, and a communication network and its connected remote output devices, etc. In addition, depending on the specific application, the electronic device may include any other suitable components such as a bus, input / output interfaces, etc.

[0094] In addition to the methods and apparatus described above, embodiments of this disclosure may also be computer program products, which include computer program instructions that, when executed by a processor, cause the processor to perform any of the methods provided in the embodiments of this disclosure.

[0095] Computer program products can be written in any combination of one or more programming languages ​​to perform the operations of embodiments of this disclosure. The programming languages ​​include object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on a user's computing device, partially on a user's computing device, as a standalone software package, partially on a user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0096] Furthermore, embodiments of this disclosure may also be computer-readable storage media storing computer program instructions that, when executed by a processor, cause the processor to perform any of the methods provided in the embodiments of this disclosure.

[0097] Computer-readable storage media may take the form of any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may, for example, include, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: electrical connections having one or more wires, portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0098] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0099] The above description is merely a specific embodiment of this disclosure, enabling those skilled in the art to understand or implement it. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this disclosure. Therefore, this disclosure is not to be limited to the embodiments described herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A target tracking method, characterized in that, The method includes: For each observation time, the observed target at that time and the position information of the observed target in the global coordinate system are obtained; wherein, the global coordinate system is divided into multiple grids; Based on the location information, position matching is performed with the plurality of grids to determine the target grid where the observed target is located from the plurality of grids; Determine the tracker corresponding to the observed target in the target grid; Based on the trackers at multiple observation times, the target tracking results of the observed target are generated.

2. The method as described in claim 1, characterized in that, The step of obtaining the observation target at the observation time and the position information of the observation target in the global coordinate system includes: Obtain the observation position of the target in the vehicle coordinate system, and transform the observation position from the vehicle coordinate system to the global coordinate system; The location information is determined based on the center point position corresponding to the observation position in the global coordinate system.

3. The method as described in claim 1, characterized in that, The step of determining the tracker corresponding to the observed target in the target grid includes: If no tracker exists in the target grid, a new tracker corresponding to the observed target is created in the target grid. If a tracker already exists in the target grid, then update the tracker in the target grid.

4. The method as described in claim 3, characterized in that, Updating the tracker in the target grid includes: Obtain the first position of the tracker in the target grid and the second position of the observed target; The third position is obtained by weighted summation of the first position and the first weight, and the second position and the second weight; wherein the first weight is greater than the second weight. Update the first position to the third position.

5. The method as described in claim 3, characterized in that, The method further includes: For each tracker at the current observation time, obtain the number of updates for that tracker; The aging threshold of the tracker is determined based on the number of times the tracker is updated; If the tracker's unupdated duration exceeds the aging threshold, the tracker is deleted.

6. The method as described in claim 1, characterized in that, After determining the tracker corresponding to the observed target in the target grid, the method further includes: For multiple trackers at the current observation time, calculate the distance between the multiple trackers; From the plurality of trackers, determine the tracker combinations whose distance is less than a first distance threshold; each tracker combination includes two trackers; If one of the two trackers has a corresponding observation target at the current observation time, and the other tracker does not have a corresponding observation target at the current observation time, then the other tracker is controlled to move a specified distance toward the first tracker. If the other tracker is in the same grid as the first tracker after it moves, then the two trackers are merged.

7. The method as described in claim 1, characterized in that, The step of generating target tracking results for the observed target based on trackers at multiple observation times includes: Candidate observation targets are determined based on trackers at multiple observation times; If the distance between any two candidate observation targets is less than the second distance threshold, then the distance between the two candidate observation targets and the vehicle is detected respectively. The candidate observation target that is farther away from the vehicle is removed from the two candidate observation targets.

8. A target tracking device, characterized in that, include: The acquisition module is used to acquire, for each observation time, the observation target at the observation time and the position information of the observation target in the global coordinate system; wherein, the global coordinate system is divided into multiple grids; A matching module is used to perform position matching with the plurality of grids based on the location information, and to determine the target grid where the observed target is located from the plurality of grids; The determination module is used to determine the tracker corresponding to the observed target in the target grid; The tracking module is used to generate target tracking results for the observed target based on trackers at multiple observation times.

9. An electronic device, characterized in that, include: processor; Memory used to store the processor's executable instructions; The processor is configured to read the executable instructions from the memory and execute the instructions to implement the target tracking method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the target tracking method according to any one of claims 1-7.

Images