Track fusion method and system

Abstract

Disclosed in the present application are a track fusion method and system. The method comprises: receiving reported tracks of a detection target, which tracks carry timestamps; and on the basis of the pieces of reported track data, estimating positioning mean square errors thereof, and on the basis of the estimated positioning mean square errors, performing data fusion on the tracks. By means of the present application, the problems, in a conventional multi-station fusion sensing method, of it being difficult to determine a target track curve which has a relatively small error and to quantitatively calculate a track error thereof are solved, thereby improving the track accuracy.

Description

A track fusion method and system

[0001] The present application claims priority to the Chinese patent application No. 202411309051.2, filed on September 19, 2024, and entitled "A track fusion method and system", the whole content of which is incorporated herein by reference. TECHNICAL FIELD

[0002] The present application belongs to the technical field of wireless communication and system perception, and specifically relates to a track fusion method and system. BACKGROUND

[0003] In 5G-A and future 6G systems, base stations will use communication and perception integrated technology to detect the surrounding environment while performing communication coverage, form tracks and report them to the center unit; in a communication and perception integrated cellular system, the cell covered by each base station is divided into three uniform sectors; the maximum range radius of perception detection is generally greater than the radius of communication coverage, which makes there is a large overlapping coverage and perception area between adjacent base stations, which makes multiple adjacent base stations detect the overlapping perception area and report independent detection results for the same target; the fusion center needs to determine whether it is the same target according to the reported target motion information (including positioning and motion speed information in the motion process) and perform track fusion on the tracks obtained by several independent detections of the same target to obtain a more accurate fused track.

[0004] However, at present, due to the lack of index estimation methods for performance metrics such as mean square error of target tracks obtained by filtering, traditional multi-base fusion perception methods are difficult to determine the detection examples with smaller errors in several curves and the quantitative performance metrics of each curve only by track coordinates, and it is difficult to ensure the track fusion performance. SUMMARY

[0005] To solve the problems in the prior art, the present application provides a track fusion method and system.

[0006] In a first aspect, the present application provides a track fusion method, comprising:

[0007] receiving the reported track of the detected target with a time stamp;

[0008] estimating the positioning mean square error of each reported track data, and performing data fusion on the track according to the estimated positioning mean square error.

[0009] Optionally, after performing data fusion on the track, the method further comprises:

[0010] filtering the fused track using a statistical model based method for the track collected based on the statistical model method; and / or,

[0011] filtering the fused track using a statistical model based method for the track collected based on the deep learning method.

[0012] Optionally, the positioning mean square error of each reported track data is estimated, and the track data is fused according to the estimated positioning mean square error, comprising:

[0013] spatiotemporal registration of the tracks;

[0014] calculating the positioning mean square error of the track;

[0015] arranging the tracks that have been spatiotemporally registered in ascending order according to the positioning mean square error of the track;

[0016] selecting the first N arranged tracks as candidate fusion tracks;

[0017] based on the minimum mean square error criterion, weighting the point coordinates of the candidate fusion tracks at time t using the reciprocal of the positioning mean square error of the candidate fusion tracks to obtain the fused track points;

[0018] filtering the T m fused track points to obtain the fused track.

[0019] Optionally, the step of spatiotemporal registration of the tracks comprises:

[0020] linearly translating the point coordinates of the track collected in the local coordinate system to the global coordinate system with the fusion center unit as the center origin coordinate;

[0021] using Lagrange three-point interpolation method to interpolate the collected track, and unifying the reported track to the synchronous time point.

[0022] Optionally, the step of calculating the positioning mean square error of the track comprises:

[0023] calculating the instantaneous velocity v n and acceleration a n at time t n from any three consecutive time-stamped coordinate points;

[0024] assuming that the track has T m time-stamped coordinate points, then obtaining T m -2 sets of (v n , a n ) values for the mth reported track;

[0025] calculating the variance of the velocity change amount of the mth reported track;

[0026] The positioning mean square error of the track is calculated according to the variance of the speed variation.

[0027] Optionally, the calculation formula of the variance of the speed variation is formula (1).

[0028] Optionally, the positioning mean square error of the track is obtained by the product of the positioning mean square error of the plot and an empirical coefficient obtained by interpolation table lookup.

[0029] The positioning mean square error of the plot is obtained by the product of the variance of the speed variation of the track and an empirical parameter obtained by field test.

[0030] In a second aspect, a track fusion system is provided, which is implemented by using the track fusion method of the first aspect, and includes:

[0031] At least two sensory base stations are configured to report the time-stamped track of the detected target to the fusion center unit.

[0032] The fusion center unit is configured to perform data fusion on the track by using the positioning mean square error.

[0033] Optionally, when the track fusion unit is filtered by using the statistical model method, the processing is based on the CPU architecture; and when the track fusion unit is filtered by using the deep learning method, the processing is based on the GPU architecture.

[0034] In a third aspect, an embodiment of the present application further provides a track fusion device for implementing the method of any one of the embodiments of the first aspect of the present application, and includes:

[0035] The acquisition module is configured to receive the reported time-stamped track of the detected target.

[0036] The determination module is configured to estimate the positioning mean square error of each reported track data, and perform data fusion on the track according to the estimated positioning mean square error.

[0037] In an embodiment of the present application, the device further includes a generation module configured to

[0038] For the track collected based on the statistical model method, the post-fusion track is filtered by using the deep learning-based method; and / or,

[0039] For the track collected based on the deep learning method, the post-fusion track is filtered by using the statistical model-based method.

[0040] In a fourth aspect, an embodiment of the present application further provides a computer readable storage medium, which stores a computer program, and the program is executed by a processor to implement the method of any one of the embodiments of the first aspect of the present application.

[0041] In a fifth aspect, the embodiments of the present application further provide an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable by the processor, and the processor implements the method according to any one of the embodiments of the first aspect of the present application when executing the computer program.

[0042] The technical solutions provided by some embodiments of the present application have at least the following beneficial effects:

[0043] The track fusion method provided by the present application solves the problems of being difficult to determine a target track curve with small error and quantitatively calculate the track error in the conventional multi-base fusion perception method, and improves the track accuracy and track fusion accuracy.

[0044] In order to make the above objects, features and advantages of the present application more apparent, the following describes preferred embodiments, and the accompanying drawings are described as follows.

[0045] The advantages of the additional aspects of the present application will be partially given in the following description, partially will become apparent from the following description, or will be understood by the practice of the present application. BRIEF DESCRIPTION OF DRAWINGS

[0046] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments of the present application or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0047] Fig. 1 is a flowchart of a track fusion method according to an embodiment of the present application;

[0048] Fig. 2 is a time registration schematic diagram according to an embodiment of the present application;

[0049] Fig. 3 is a multi-base cooperative perception integrated track fusion system scene schematic diagram according to an embodiment of the present application;

[0050] Fig. 4 is a multi-base cooperative perception integrated track fusion flowchart according to an embodiment of the present application;

[0051] Fig. 5 is an embodiment of a track fusion device according to the present application;

[0052] Fig. 6 is another embodiment of a track fusion device according to the present application. DETAILED DESCRIPTION

[0053] In order to make the objects, technical solutions and advantages of the present application more apparent, the following will further describe the embodiments of the present application in combination with the drawings.

[0054] The following description refers to the accompanying drawings. Unless otherwise noted, like elements in different drawings have the same or similar reference numerals. The following description of exemplary embodiments does not represent all contemplated embodiments consistent with the application. Instead, they are merely examples of methods and apparatus consistent with some aspects of the application as detailed in the appended claims.

[0055] In the description of the present application, it should be understood that the term "multiple" refers to two or more. The association relationship of the associated objects is described, which means that there can be three relationships, for example, A and / or B can represent: A exists alone, A and B exist together, and B exists alone. The character " / " generally represents that the associated objects before and after are in an "or" relationship.

[0056] The present application provides a track fusion method and system, which detects the same piece of space to be detected by using multiple sensing base stations with different physical space positions; each base station reports the detected track to the fusion center unit; the fusion center unit fuses the information reported by multiple stations to obtain a target motion track with higher accuracy; wherein the track fusion system can identify the target according to the fused target track, and the method and system of the present application are suitable for various sensing technology scenes, including emerging communication and sensing integrated technology such as unmanned aerial vehicle detection and tracking scene, vehicle detection scene of Internet of Vehicles, etc.

[0057] The present application only needs each distributed cooperative base station to report the point track / track detected by itself and the base station ID of itself, without reporting other prior information, so that the fusion center can quantitatively determine the accuracy quality of each reported track, select candidate point tracks / tracks with higher accuracy, and fuse the point tracks / tracks of the target to obtain a fused track with higher accuracy.

[0058] The following will be further described in combination with embodiments and drawings.

[0059] Embodiment one

[0060] The track fusion method provided by the embodiment of the present application will be described in detail below in combination with Figure 1.

[0061] The flowchart of the track fusion method provided by the embodiment of the present application is shown in Figure 1, and the method of the embodiment of the present application can include the following steps:

[0062] Step S1: receiving the reported time-stamped track of the detected target;

[0063] The point track is the original coordinate point of the detected target position, and the track is the T mThe target motion trajectory is obtained after filtering the original coordinate points; in this embodiment, m sensing base stations are used to report the trajectory of the detected target, and a fusion center unit is used to receive the trajectory.

[0064] Step S2: Estimate the positioning mean square error of each reported track data, and perform data fusion on the track based on the estimated positioning mean square error;

[0065] Specifically, the fusion process includes the following steps:

[0066] Step S2.1: Perform spatiotemporal registration on the tracks;

[0067] It is understandable that spatiotemporal registration is divided into spatial registration and temporal registration. Temporal registration is a necessary step introduced to deal with the incomplete synchronization of timestamps when collecting data. Figure 2 is a schematic diagram of time registration of the track.

[0068] Specifically, the process of spatiotemporal registration of flight tracks is as follows:

[0069] Step S2.1.1: Linearly translate and transform the point coordinates of the track collected in the local coordinate system to the global coordinate system with the fusion center cell as the origin.

[0070] Specifically, in this embodiment, the point coordinates of the tracks of m sensing base stations in the local coordinate system are linearly translated and transformed to the global coordinate system with the fusion center unit as the origin.

[0071] Step S2.1.2: Use the Lagrange three-point interpolation method to interpolate the collected tracks and unify the reported tracks to the synchronization time point.

[0072] Specifically, in this embodiment, the target tracks collected by m sensing base stations are interpolated to unify the reported tracks of the m sensing base stations to the synchronization time point.

[0073] Step S2.2: Calculate the positioning mean square error of the track;

[0074] Both point tracks and flight tracks are sequences of coordinate points with timestamps. Assuming the time interval between any two adjacent reported coordinate points is Δt, the instantaneous velocity v at any given moment can be calculated from any three consecutive coordinate points with timestamps. t and acceleration a t ;

[0075] In one feasible implementation, the steps are as follows:

[0076] Step S2.2.1: Calculate t from any three consecutive coordinate points with timestamps. n instantaneous velocity v at time t n and acceleration a n ;

[0077] Suppose the coordinate point positions of three consecutive time instants t n , t n+1 = t n + Δt' and t n+2 = t n + 2Δt' (Δt' can be several times of Δt, i.e. interval values are taken) are l n , l n+1 , l n+2 respectively, then the velocity at time instant t n can be expressed as: The acceleration can be expressed as:

[0078] Step S2.2.2: Suppose that there are T m coordinate points with time stamps in the track, then T m -2 values of (v n , a n ) are obtained for the mth reported track.

[0079] Step S2.2.3: The variance of the velocity variation of the mth reported track is calculated.

[0080] The variance of the velocity variation of the mth reported track is calculated according to (v n , a n ) and is σ Δv,m 2 Specifically, the calculation formula is:

[0081] wherein n is the nth point of the mth reported track.

[0082] Step S2.2.4: The positioning mean square error of the track is calculated according to the variance of the velocity variation.

[0083] Specifically, the calculation formula of the positioning mean square error of the track according to the variance of the velocity variation is as follows:

[0084] wherein is the positioning mean square error of the track, is the positioning mean square error of the point track, and α m is an empirical coefficient looked up from an interpolation table of α m .

[0085] In this embodiment, the interpolation table of α m needs to be obtained through Monte Carlo simulation or field test;

[0086] Specifically, the target to be tested is placed in a grid space coordinate point centered on the radar of the common sense base station, each point has different distance and direction angle, and the value of a m at the distance and angle parameters is calculated.

[0087] The application interpolates to obtain the fitting value of a m at a certain coordinate point.

[0088] In the case of reporting the track, the target track is uploaded by each common sense base station in the embodiment, and the track is actually obtained by filtering the point track collected by the common sense base station. The positioning mean square error of the point track is calculated as follows:

[0089] Wherein, is the positioning mean square error of the point track, a is an empirical parameter obtained by field test, and s Δv,m 2 is the variance of the speed change of the mth reported track.

[0090] Step S2.3: arranging the tracks that have been time and space registered in ascending order according to the positioning mean square error of the track;

[0091] Specifically, the positioning MSE (Mean Squared Error, mean square error) of all point tracks of the common sense base station is obtained through step S2.2 or the positioning MSE of the track Then, the m tracks that have been time and space registered are arranged in ascending order according to the positioning MSE from small to large. The smaller the MSE value is, the higher the accuracy of the track is.

[0092] Step S2.4: selecting the first N arranged tracks as candidate fusion tracks;

[0093] In the specific implementation process, the CU (Central Unit, fusion center unit) selects the first N best time and space registered tracks as candidate fusion tracks, and the MSE set of the candidate fusion tracks is denoted as The point coordinate set of the track at time t is denoted as

[0094] Step S2.5: based on the minimum mean square error criterion, the reciprocal of the positioning mean square error of the candidate fusion track is used to weight the point coordinates of the candidate fusion track at time t, to obtain the fused track point;

[0095] Specifically, the fused track point is denoted as Based on the minimum MSE criterion, the reciprocal of the track MSE is used for weighting, to obtain

[0096] Step S2.6: filtering the T fused track points to obtain a fused track. m Step S2.6: filtering the T fused track points to obtain a fused track.

[0097] Specifically, after fusing all T points, a fused track is obtained m Step S2.6: filtering the T fused track points to obtain a fused track.

[0098] In a feasible implementation, after data fusion of the track, the method further comprises: filtering the track collected based on the statistical model method using a deep learning-based method; the deep learning method comprises a convolutional neural network or any similar method.

[0099] In a feasible implementation, after data fusion of the track, the method further comprises: filtering the track collected based on the deep learning method using a statistical model-based method. The statistical model method comprises a support vector machine or any similar method. The above-mentioned methods are applicable to a track fusion system. In this case, the on-site report is a track after filtering of a point track, and the on-site signal collection device (a sensing base station in this embodiment) outputs a track. The point track is only a state in an internal intermediate processing process. The above-mentioned methods are also applicable to a case where the sensing base station reports a point track to the fusion center unit. In this case, the point track is fused by the same method, and the system for point track fusion is each sensing base station reporting a point track. The fusion center unit filters the fused point track to generate a fused track. The filtering method in this embodiment is Kalman filtering.

[0100] By using the track fusion method disclosed in the present application, the problem that the conventional multi-base fusion sensing method is difficult to determine a detection example with smaller error in a plurality of curves in a detected target only by using a point track / track coordinate, and is difficult to quantitatively measure the performance of each curve, and is difficult to ensure the point track / track fusion performance is solved, and the track accuracy and track fusion accuracy are improved.

[0101] It should be noted that the execution subject of each step of the method provided in Embodiment One can be the same device, or the method can also be executed by different devices as the execution subject. For example, the execution subject of step S1 and step S2 can be device 1; for another example, the execution subject of step S1 can be device 1, and the execution subject of step S2 can be device 2; and the like.

[0102] Embodiment Two

[0103] The present application provides a track fusion system, which is implemented by using the track fusion method in Embodiment One, and comprises:

[0104] At least two sensing base stations for reporting a track with a time stamp of a detected target to a fusion center unit, as shown in FIG. 3;

[0105] Specifically, the ISAC (Integrated Sensing And Communication, communication sensing integration) base station, as shown in FIG. 2, is denoted as BS (Base Station), each ISAC base station contains an AAU (Active Antenna Unit) and a BBU (Building Base band Unite) module, and each base station can use any frequency band; the AAU transmits and receives radio frequency signals and echo signals, and the BBU processes the echo signals to obtain target track sequence information and uploads it to the fusion center unit;

[0106] The specific working process is that m adjacent ISAC base stations report the tracks with time stamps to the fusion center unit. The fusion center unit first performs space-time registration and track MSE estimation on the reported tracks. The point track refers to the original coordinate point information of the target position detected by the ISAC base station, and the track refers to the target motion trajectory obtained by filtering (such as using Kalman filtering method) the original target point data detected in continuous time.

[0107] The fusion center unit is used to estimate the positioning mean square error of each reported track data, and to perform data fusion on the tracks according to the estimated positioning mean square error;

[0108] Specifically, the positioning accuracy index of all distributed ISAC base station reported track sequences is evaluated, the reported track sequences are sorted according to the index evaluation, the track association of the track set is performed, the tracks belonging to the same target are determined, and a number of tracks with high quality of the same target are selected for fusion and filtering; this process can be regarded as a first filtering process to obtain the fused track;

[0109] In the specific implementation process, as shown in FIG. 4, the fusion center unit can further use a secondary filtering and prediction module to filter the fused track, and the specific process is as follows:

[0110] The secondary filtering and prediction module continues to filter the transmitted fused track to obtain a more accurate target trajectory; at this time, if the radar of the base station uses a statistical model based method, the secondary filtering and prediction module uses a deep learning based method; conversely, if the radar of the base station uses a deep learning based method, the secondary filtering and prediction module uses a statistical model based method.

[0111] The specific cooperation process of the fusion center unit and the sensing base station is as follows: the CU (Centralized Unit, centralized unit) receives track information reported by multiple adjacent BSs after independent detection, and each reported track information should have the identification ID of the BS completing the target detection;

[0112] The CU first performs data fusion on the tracks reported by the connected m adjacent sensing base stations, and then inputs the fused target track into a track fusion unit based on a statistical model method or a deep learning method for track fusion.

[0113] In the specific implementation process, when the track fusion unit uses a statistical model method for filtering, it is based on CPU architecture processing; when it uses a deep learning method for filtering, it is based on GPU architecture processing.

[0114] It can be understood that the track fusion system can also include a terminal unit for receiving and displaying the track fusion results of the fusion center unit.

[0115] Specifically, the fusion track result information transmitted by the fusion center unit is received, the track fusion classification result transmitted by the fusion center unit is received, and the received information is displayed on the display screen of the terminal unit.

[0116] Specifically, the track fusion module implements the track fusion unit function, and is combined with the data fusion processing module as a fusion center unit to realize the track fusion function of the system; the terminal unit is a visualization and control unit, which is composed of a visualization display module, a human-computer interaction module, and a remote automatic control module.

[0117] The visualization display module receives the fusion track result transmitted by the CU data fusion processing module, receives the track fusion classification result transmitted by the CU track fusion unit module, and displays the received information on the display screen; the human-computer interaction module provides an interactive interface between a person and a remote control module; the remote automatic control module controls external systems to perform flight control, network configuration, and other operations according to human-computer interaction instructions.

[0118] The track fusion system provided in the embodiment is suitable for various sensing technology scenarios, including emerging communication and sensing integrated technologies such as unmanned aerial vehicle detection and tracking scenarios, vehicle networking vehicle detection scenarios, etc. By using the track fusion method provided in Embodiment 1, the same technical features as the track fusion method provided in Embodiment 1 are achieved, so the same technical problems can be solved, and the same technical effects can be achieved. Therefore, it is also suitable for a track fusion system.

[0119] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific implementation process of the functions of the above-described system and units can refer to the corresponding process in the foregoing method embodiments, which will not be described here.

[0120] Embodiment three

[0121] FIG. 5 is an embodiment of a track fusion device of the present application, which is used to implement the track fusion method of the present application, and further implements the fusion center unit of the present application. The device comprises:

[0122] The acquisition module 510 is configured to receive the reported track of the detected target with a time stamp;

[0123] The determination module 520 is configured to estimate the positioning mean square error of each reported track data, and perform data fusion on the track according to the estimated positioning mean square error.

[0124] In an embodiment of the present application, the device further comprises a generation module 530 configured to filter the T fused track points to obtain the fused track. m

[0125] In an embodiment of the present application, the generation module is further configured to: filter the fused track using a deep learning-based method based on the track collected by a statistical model method; and / or filter the fused track using a statistical model-based method based on the track collected by a deep learning method.

[0126] In an embodiment of the present application, the device further comprises a registration module configured to time-control registration of the track, preferably linearly translating the point coordinates of the track collected in a local coordinate system to a global coordinate system with the fusion center unit as the center origin coordinate; and using Lagrange three-point interpolation method to interpolate the collected track to synchronize the reported track to a time point.

[0127] In an embodiment of the present application, the determination module further comprises an estimation submodule configured to calculate the positioning mean square error of the track. In an embodiment, the positioning mean square error of the track is obtained by the product of the positioning mean square error of the point track and the empirical coefficient obtained by interpolation table lookup; in an embodiment, the positioning mean square error of the point track is obtained by the product of the variance of the speed variation of the track and the empirical parameter obtained by field test. Preferably, the instantaneous speed v n and the acceleration a n at time t n are calculated from any three consecutive coordinate points with time stamps; assuming that the track has T m coordinate points with time stamps, then T m -2 groups of (v​n a n ) of the mth reported track; calculating a variance of the speed variation of the mth reported track; and calculating a positioning mean square error of the track according to the variance of the speed variation.

[0128] In an embodiment of the present application, the determining module further comprises a fusion sub-module, configured to determine the fused track point. Preferably, in an embodiment, the time and space registered tracks are arranged in ascending order according to the positioning mean square error of the tracks; the first N arranged tracks are selected as candidate fused tracks; and the point coordinates of the candidate fused tracks at time t are weighted by the reciprocal of the positioning mean square error of the candidate fused tracks based on the least mean square error criterion, to obtain the fused track point. Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, some embodiments of the present application can be in the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can be in the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk memory, CD-ROM, optical memory, etc.) containing computer-usable program code.

[0129] Therefore, the present application further provides a computer readable storage medium, having stored thereon a computer program, which, when executed by a processor, implements the method according to any of the embodiments of the present application.

[0130] Further, the present application further provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable by the processor, wherein the processor implements the method according to any of the embodiments of the present application when executing the computer program.

[0131] For example, the embodiments of the present application are described in reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general purpose computer, a special purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce the functions specified in one or more flows in the flowcharts and / or one or more blocks in the block diagrams.

[0132] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instructions which implement the function specified in the flowchart or flowsheets and / or block or blocks of the block diagrams.

[0133] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart or flowsheets and / or block or blocks of the block diagrams.

[0134] In one typical arrangement, the computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory. The memory can include non-persistent memory, random access memory (RAM), and / or non-volatile memory, such as read only memory (ROM) or flash memory, among others, in a computer readable medium. The memory is an example of computer readable media.

[0135] FIG. 6 is a structural diagram of an electronic device for another embodiment of the track fusion device provided in the present application. The electronic device 600 shown is merely an example and should not impose any limitation on the function and scope of use of the embodiments of the present application. It includes one or more processors 620; a storage device 610 for storing one or more programs, when the one or more programs are run by the one or more processors 620, the one or more processors 620 implement the track fusion method provided by the embodiments of the present application, the method includes:

[0136] receiving the reported track of the detection target with time stamp;

[0137] estimating the positioning mean square error of each reported track data, and using the estimated positioning mean square error to perform data fusion on the track.

[0138] Further embodiments of the method are described as steps S1-S2.

[0139] The electronic device 600 also includes an input device 630 and an output device 640; the processor 620, the storage device 610, the input device 630 and the output device 640 in the electronic device can be connected through a bus or other means, and in the figure, they are connected through a bus 650.

[0140] The storage device 610, as a computer readable storage medium, can be used to store software programs, computer executable programs and module units, such as program instructions corresponding to the track fusion method in the embodiments of the present application. The storage device 610 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system and application programs required by at least one function; the data storage area can store data created according to the use of the terminal and the like. In addition, the storage device 610 can include a high-speed random access memory, and can also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some examples, the storage device 610 can further include a memory disposed remotely with respect to the processor 620, and these remote memories can be connected through a network. Examples of the above network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0141] The input device 630 can be used to receive input digital, character information or voice information, and generate key signal input related to user settings and function control of the electronic device. The output device 640 can include a display screen, a speaker and the like electronic device.

[0142] It should also be noted that the terms "comprising", "including", or any other variant thereof are intended to cover non-exclusive inclusion, so that processes, methods, articles or devices including a series of elements not only include those elements, but also include other elements not explicitly listed, or further include elements inherent to such processes, methods, articles or devices. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of additional identical elements in the process, method, article or device including the element.

[0143] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present application, which are used to illustrate the technical solutions of the present application, rather than limit the same, and the protection scope of the present application is not limited thereto. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can make modifications or easily think of changes to the technical solutions recorded in the foregoing embodiments within the technical scope disclosed by the present application, or make equivalent replacements to some technical features; and these modifications, changes or replacements do not cause the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0144] The applicant of the present application makes detailed explanation and description on the implementation examples of the present application in combination with the drawings of the specification, but the skilled in the art should understand that the above implementation examples are only the preferred implementation of the present application, and the detailed explanation is only to help the reader better understand the spirit of the present application, and does not limit the protection scope of the present application. On the contrary, any improvement or modification based on the spirit of the present application should fall within the protection scope of the present application.

Claims

1. A track fusion method, characterized by, The method comprises the following steps: receiving the reported track of the detection target with a timestamp; estimating the positioning mean square error of each reported track data, and performing data fusion on the track according to the estimated positioning mean square error.

2. The track fusion method according to claim 1, characterized in that, After the data fusion of the track, the method further comprises the following steps: filtering the fused track by using a deep learning-based method for the track collected based on a statistical model method; and / or filtering the fused track by using a statistical model-based method for the track collected based on a deep learning method.

3. The track fusion method according to claim 1, characterized in that, The step of estimating the positioning mean square error of each reported track data and performing data fusion on the track according to the estimated positioning mean square error comprises the following steps: spatiotemporally registering the track; calculating the positioning mean square error of the track; arranging the track that has been spatiotemporally registered in ascending order according to the positioning mean square error of the track; selecting the first N arranged tracks as candidate fused tracks; weighting the point coordinates of the candidate fused track at time t by using the reciprocal of the positioning mean square error of the candidate fused track to obtain the fused track point; To T m The fused track points are filtered to obtain a fused track.

4. The track fusion method of claim 1, wherein, The method further comprises the following steps of spatiotemporally registering the track: linearly translating the point coordinates of the track collected in the local coordinate system to the global coordinate system with the fusion center unit as the center origin coordinate; using the Lagrange three-point interpolation method to interpolate the collected track, and unifying the reported track to the synchronous time point.

5. The track fusion method of claim 1, wherein, The step of calculating the positioning mean square error of the track comprises the following steps: t is calculated from any three consecutive time-stamped coordinate points n the instantaneous velocity v at the moment n and the acceleration a n ; Assume that there are T m coordinate points with time stamps for the track, then for the mth reported track, obtain T m values of the set (v n , a n ). calculating the variance of the speed variation of the mth reported track; calculating the positioning mean square error of the track according to the variance of the speed variation.

6. The track fusion method according to claim 5, characterized in that, The calculation formula of the variance of the speed variation amount is: where σ Δv,m 2 is the variance of the velocity change, n is the nth point of the mth reported track.

7. The track fusion method of claim 1, wherein, The positioning mean square error of the track is obtained by the product of the positioning mean square error of the point track and the empirical coefficient obtained by interpolation table lookup.

8. The track fusion method according to claim 7, characterized in that, The positioning mean square error of the point track is obtained by the product of the variance of the speed variation of the track and the empirical parameter obtained by field test.

9. A track fusion system implemented using the track fusion method according to any one of claims 1 to 8, characterized by, The method comprises the following steps: at least two sensory base stations are used to report the track of the detection target with a timestamp to the fusion center unit; the fusion center unit is used to estimate the positioning mean square error of each reported track data, and perform data fusion on the track according to the estimated positioning mean square error.

10. The track fusion system of claim 9, wherein, When the track fusion unit filters by using the statistical model method, the processing is based on the CPU architecture; when the track fusion unit filters by using the deep learning method, the processing is based on the GPU architecture.

11. A track fusion device for implementing the method according to any one of claims 1 to 8, characterized in that The method comprises the following steps: an acquisition module is used to receive the reported track of the detection target with a timestamp; a determination module is used to estimate the positioning mean square error of each reported track data, and perform data fusion on the track according to the estimated positioning mean square error.

12. A track fusion device as claimed in claim 11, characterized in that The method further comprises the following steps of a generation module: filtering the fused track by using a deep learning-based method for the track collected based on a statistical model method; and / or filtering the fused track by using a statistical model-based method for the track collected based on a deep learning method.

13. A computer readable storage medium having stored thereon a computer program, characterized in that The program is executed by the processor to implement the method in any one of claims 1-8.

14. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the computer program to implement the method in any one of claims 1-8.

Images