A method, apparatus, storage medium and electronic device for roadside testing latency

Abstract

Some embodiments of the present application provide a method, device, storage medium and electronic equipment for roadside test latency, the method comprising: obtaining positioning information obtained by a true value vehicle itself positioning in a fixed time period, wherein the positioning information comprises position coordinates and speed at each timestamp in the fixed time period; taking any one of the position coordinates at the timestamp and the corresponding position coordinate as a reference timestamp and a reference position; based on the reference timestamp, finding a target perception frame matching the reference position from a roadside perception frame, and taking the difference between the reference timestamp and the timestamp corresponding to the target perception frame as a target latency value. Some embodiments of the present application can obtain a target latency value with high accuracy, so that the roadside perception vehicle and the true value vehicle are synchronized without latency.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent transportation, in particular to a roadside test time delay method, device, storage medium and electronic equipment. BACKGROUND

[0002] Roadside perception is to use cameras, millimeter wave radars, laser radars and other sensors, and combine roadside edge computing to realize instantaneous intelligent perception of traffic participants and road conditions of the road section. Roadside perception can expand the perception range of autonomous vehicles and drivers, and realize integrated operation monitoring of man-vehicle-road-cloud through vehicle-road cooperation technology. When the road traffic anomaly is found in the first time, vehicle-road cooperation, vehicle-cloud cooperation, regional road-cloud cooperation and other intelligent applications are realized to meet the intelligent travel needs of autonomous vehicles and social vehicles.

[0003] At present, the accuracy and recall rate test of roadside perception can be obtained by comparing the predicted frame on the picture with the manually labeled frame. However, manual marking of vehicles by artificial is prone to positioning error. In the traditional scheme, the vehicle can carry related equipment as a true value vehicle to co-locate with the roadside perception device to realize the accuracy of vehicle positioning. However, due to the different data output frequencies of the true value vehicle and the roadside perception device, there is a time delay problem, which further affects the transmission of data and the accuracy of positioning.

[0004] Therefore, how to provide a technical scheme of a roadside test time delay method with high accuracy becomes a technical problem to be solved. SUMMARY

[0005] Some embodiments of the present application aim to provide a roadside test time delay method, device, storage medium and electronic equipment. The technical scheme of the embodiments of the present application can realize accurate determination of the time delay value between the true value vehicle and the roadside perception device, and further realize accurate positioning of the target vehicle, so that the data of roadside perception and the data of the true value vehicle remain synchronous without time delay.

[0006] In a first aspect, some embodiments of the present application provide a roadside test time delay method, comprising: obtaining positioning information obtained by a true value vehicle itself within a fixed time period, wherein the positioning information comprises position coordinates and speed at each timestamp within the fixed time period; taking any one of the position coordinates at the timestamp and the corresponding position coordinate as a reference timestamp and a reference position; based on the reference timestamp, finding a target perception frame matching the reference position from a roadside perception frame, and taking the difference between the reference timestamp and the timestamp corresponding to the target perception frame as a target time delay value.

[0007] Some embodiments of the present application can achieve accurate determination of the time delay value between the true value vehicle and the roadside perception device, and thus can achieve accurate positioning of the target vehicle, so that the data of the roadside perception and the data of the true value vehicle remain synchronized without time delay.

[0008] In some embodiments, the obtaining the positioning information obtained by the true value vehicle itself within the fixed time period comprises: obtaining initial positioning information of the true value vehicle in adjacent two frames of images, wherein a time interval of the adjacent two frames of images is the fixed time period, and the initial positioning information comprises a first time stamp of a first frame of image, a first position and a first speed of the true value vehicle, and a second time stamp of a second frame of image, a second position and a second speed of the true value vehicle; and obtaining position coordinates at each time stamp and speeds at the each time stamp of the true value vehicle between the first time stamp and the second time stamp based on the initial positioning information.

[0009] Some embodiments of the present application can achieve filling of the true value vehicle data by obtaining initial positioning information of adjacent two frames of images within a time interval, calculating position coordinates and speeds at each time stamp within a time interval, and provide effective data support for subsequent confirmation of the target time delay value.

[0010] In some embodiments, the fixed time period is determined by a frame rate of a real-time kinematic carrier phase difference measurement instrument deployed by the true value vehicle.

[0011] In some embodiments, the finding the target perception frame matching the reference position from the roadside perception frames based on the reference time stamp comprises: obtaining a corresponding test frame set existing in the roadside perception frame at a test time stamp T, wherein the test frame set comprises a plurality of test frames corresponding to a plurality of test time intervals, and the T is a sum of the reference time stamp Tc and n*t, wherein t is a test time interval, n is a time interval number and is a positive integer; detecting each test frame in the test frame set to obtain a to-be-matched frame containing the target vehicle; and comparing the target vehicle in the to-be-matched frame with the true value vehicle to determine the target perception frame.

[0012] Some embodiments of the present application can ensure the accuracy of vehicle matching and improve the accuracy of the final result by obtaining a plurality of test frames through a plurality of test time intervals related to the reference time stamp, then detecting the plurality of test frames to obtain a to-be-matched frame, and finally obtaining a target perception frame through comparison.

[0013] In some embodiments, the comparing the target vehicle in the frame to be matched with the ground truth vehicle to determine the target perception frame comprises: determining that the target vehicle is of the same type as the ground truth vehicle, and confirming that there is no other vehicle in the frame to be matched within a preset range of the target vehicle, and that the position of the target vehicle meets a preset condition with the reference position, and then taking the frame to be matched as the target perception frame.

[0014] Some embodiments of the present application can ensure the accuracy of the data by determining the frame to be matched from multiple dimensions to confirm that the frame to be matched is the target perception frame, thereby improving the accuracy of the final result.

[0015] In some embodiments, before taking the difference between the reference timestamp and the timestamp corresponding to the target perception frame as the target time delay value, the method further comprises: taking the difference between the reference timestamp and the timestamp corresponding to the target perception frame as a calibration time delay; obtaining at least two frames in which the target vehicle exists in the road side perception frame; compensating the corresponding time of each of the at least two frames by the calibration time delay to obtain each compensated time; searching for the positioning coordinates obtained by the self-positioning of the ground truth vehicle corresponding to each compensated time; and if it is confirmed that the positioning coordinates obtained by the self-positioning of the ground truth vehicle match the target vehicle successfully, taking the calibration time delay as the target time delay value.

[0016] Some embodiments of the present application can ensure the accuracy of the target time delay value by compensating the road side perception frame by the calibration time delay to obtain the positioning coordinates of the ground truth vehicle, and then comparing the positioning coordinates with the real positioning coordinates of the ground truth vehicle, and determining the target time delay value after a successful comparison.

[0017] In some embodiments, the searching for the target perception frame matching the reference position from the road side perception frame based on the reference timestamp comprises: cyclically performing the following operations until the number of target perception frames in the target perception frame set reaches a preset value: updating the reference timestamp and updating the reference position; searching for the target perception frame from the road side perception frame and adding it to the target perception frame set; wherein the taking the difference between the reference timestamp and the timestamp corresponding to the target perception frame as the target time delay value comprises: calculating the difference between the timestamp of each target perception frame in the target perception frame set and the corresponding reference timestamp to obtain each time delay; and taking the time delay that appears most frequently in the each time delay as the target time delay value.

[0018] Some embodiments of the present application can improve the accuracy of the target time delay value obtained by cyclically performing the operation of obtaining the target perception frame and the time delay value, and selecting the time delay that appears most frequently as the target time delay value.

[0019] In a second aspect, some embodiments of the present application provide a device for roadside test latency, comprising: a data acquisition module configured to acquire positioning information obtained by a ground truth vehicle itself positioning in a fixed time period, wherein the positioning information comprises position coordinates and speed at each timestamp in the fixed time period; a data selection module configured to select any one of the position coordinates at each timestamp and the corresponding position coordinate as a reference timestamp and a reference position; and a data matching module configured to find a target perception frame matching the reference position from a roadside perception frame based on the reference timestamp, and take a difference between the reference timestamp and a timestamp corresponding to the target perception frame as a target latency value.

[0020] In a third aspect, some embodiments of the present application provide a computer readable storage medium having stored thereon a computer program, which, when executed by a processor, can implement the method according to any one of the first aspect.

[0021] In a fourth aspect, some embodiments of the present application provide an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor can implement the method according to any one of the first aspect when executing the program.

[0022] In a fifth aspect, some embodiments of the present application provide a computer program product, comprising a computer program, wherein the computer program, when executed by a processor, can implement the method according to any one of the first aspect. BRIEF DESCRIPTION OF DRAWINGS

[0023] In order to more clearly illustrate the technical solutions of some embodiments of the present application, the following will briefly introduce the drawings needed to be used in some embodiments of the present application. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can also be obtained without creative labor.

[0024] Figure 1 A system diagram for roadside test latency provided by some embodiments of the present application;

[0025] Figure 2 One of the method flowcharts for roadside test latency provided by some embodiments of the present application;

[0026] Figure 3 The second method flowchart for roadside test latency provided by some embodiments of the present application;

[0027] Figure 4An apparatus block diagram for roadside test latency is provided for some embodiments of the present application.

[0028] Figure 5 An electronic device schematic diagram is provided for some embodiments of the present application. DETAILED DESCRIPTION

[0029] The technical solutions in the embodiments of the present application will be described below with reference to the drawings in the embodiments of the present application.

[0030] It should be noted that similar reference numerals and letters refer to similar items throughout the accompanying drawings, and thus, once an item is defined in one drawing, it is not necessary to further define and explain it in subsequent drawings. Meanwhile, in the description of the present application, the terms "first", "second", and the like are only used to distinguish descriptions, and cannot be understood as indicating or implying relative importance.

[0031] In the related art, in addition to outputting the detected target type and picture position, the roadside perception device also outputs the position of the target in the world coordinate system. This world coordinate position not only depends on the algorithm detection on the image, but also depends on the calibration file for converting the image position to the world coordinate. Since the target position result of the roadside perception needs to be used for vehicle-side blind filling or digital display, etc., the positioning result of the roadside perception needs to be tested. The accuracy and recall rate test of the roadside perception can be obtained by comparing the predicted box on the picture with the manually labeled box. However, the precise positioning cannot be obtained by manual marking, and the error in the predicted positioning cannot be eliminated. The usual solution is to use a vehicle equipped with RTK (Real-time kinematic, real-time dynamic, real-time dynamic carrier phase difference technology) as a true value vehicle to cooperate with the roadside perception device for positioning. However, there is a time delay problem between the true value vehicle and the roadside perception, and the frequency of the true value vehicle is not consistent with the frequency of the roadside perception, which leads to the problem that the true value vehicle may be mismatched with the vehicle perceived by the roadside perception, thereby calculating an error.

[0032] From the above related art, it can be seen that in the prior art, since there is a time delay between the true value vehicle and the roadside perception, it is particularly important to accurately determine the time delay between the two, so that the true value vehicle and the roadside perception are kept in synchronization without time delay.

[0033] In view of this, some embodiments of the present application provide a method for testing roadside latency, which takes a certain timestamp and corresponding position coordinates in the positioning information obtained by the true value vehicle itself as a reference timestamp and reference position, finds a target perception frame matching the reference position in a roadside perception frame, and finally takes the difference between the target perception frame corresponding timestamp and the reference timestamp as the target latency value. Some embodiments of the present application can realize the testing of the target latency value between the true value vehicle and the roadside perception, with high accuracy, so as to realize the accurate positioning of the target vehicle, so that the data of the roadside perception and the data of the true value vehicle remain synchronized without latency.

[0034] The overall structure of the system for testing roadside latency provided by some embodiments of the present application is described below in conjunction with the accompanying Figure 1 The overall structure of the system for testing roadside latency provided by some embodiments of the present application is described below in conjunction with the accompanying

[0035] As Figure 1 shown, some embodiments of the present application provide a system diagram for testing roadside latency, which includes a true value vehicle 100, a roadside perception end 200, and a data processing end 300. The true value vehicle 100 can be equipped with an RTK and can realize self-positioning. When the true value vehicle 100 is driving within the roadside perception range, the true value vehicle 100 can send the positioning information to the data processing end 300. The roadside perception end 200 can detect the target on the road, obtain the positioning information of the target by means of calibration file or model prediction, and send it to the data processing end 300. The data processing end 300 can obtain the position coordinates at each timestamp based on the positioning information of the true value vehicle 100. Then, any one of the timestamps and the corresponding position coordinates in the position coordinates at each timestamp are taken as a reference timestamp and a reference position. Finally, the data processing end 300 can find a target perception frame matching the reference position from the roadside perception frame based on the reference timestamp, and take the difference between the reference timestamp and the timestamp corresponding to the target perception frame as the target latency value.

[0036] It should be noted that in some other embodiments of the present application, in addition to being equipped with an RTK, the true value vehicle 100 can also be equipped with other software or hardware similar to the RTK function that can realize the positioning of the true value vehicle 100, and the present application is not limited thereto.

[0037] In some embodiments of the present application, the data processing end 300 can be a cloud processing device deployed independently of the true value vehicle 100 and the roadside perception end 200. In some other embodiments of the present application, the data processing end 300 can also be deployed inside the true value vehicle 100 or the roadside perception end 200, so that the true value vehicle 100 or the roadside perception end 200 can process the positioning information to obtain the target latency value. The specific setting can be made according to the actual situation, and the present application is not limited thereto.

[0038] The following describes the implementation process of the method for roadside test latency provided by some embodiments of the present application in combination with the accompanying drawings. Figure 2 The implementation process of the method for roadside test latency provided by some embodiments of the present application is described in combination with the accompanying drawings.

[0039] The implementation process of the method for roadside test latency provided by some embodiments of the present application is described in combination with the accompanying drawings. Figure 2 , Figure 2 The implementation process of the method for roadside test latency provided by some embodiments of the present application is described in combination with the accompanying drawings.

[0040] S210, obtaining positioning information obtained by the true value vehicle itself at a fixed time period, wherein the positioning information comprises position coordinates and speed at each time stamp in the fixed time period.

[0041] For example, in some embodiments of the present application, the true value vehicle 100 can use RTK for self-positioning when driving in the roadside sensing range because it is equipped with RTK. Since the coordinates output by RTK are generally not centered on the vehicle, it is necessary to use the calibration relationship between RTK and the center of the vehicle to transform and record the coordinates and speed of the center of the bottom surface of the vehicle at a certain time stamp.

[0042] In some embodiments of the present application, the fixed time period is determined by the frame rate of the real-time dynamic carrier phase difference measurement instrument deployed by the true value vehicle.

[0043] For example, in some embodiments of the present application, the frame rate of RTK is generally 100 Hz, that is, a frame can be output every 10 ms, so the fixed time period is 10 ms.

[0044] In some embodiments of the present application, S210 can include: obtaining initial positioning information of the true value vehicle at two adjacent frames of images, wherein the time interval of the two adjacent frames of images is the fixed time period, and the initial positioning information comprises a first time stamp of a first frame of image, a first position and a first speed of the true value vehicle, and a second time stamp of a second frame of image, a second position and a second speed of the true value vehicle; based on the initial positioning information, obtaining position coordinates of the true value vehicle at each time stamp between the first time stamp and the second time stamp and speed at the each time stamp.

[0045] For example, in some embodiments of the present application, the RTK outputs an image every 10 ms, for example. The first timestamp of the first image is T1 (i.e., representing that the first image is output at time node T1), the first position is (X1, Y1), and the first speed is V1. The second timestamp of the second image is T2 (i.e., representing that the second image is output at time node T2), the second position is (X2, Y2), and the second speed is V2. Wherein T2-T1 = 10 ms. Based on the first position, the second position, and the first timestamp and the second timestamp, interpolation calculation is performed to obtain the position coordinates and the speed at each ms (i.e., each timestamp).

[0046] Specifically, when obtaining the horizontal coordinate of the position coordinates at the i-th timestamp, first, the ratio between the difference between the horizontal coordinate of the first position and the horizontal coordinate of the second position and the difference between the first timestamp and the second timestamp is obtained. Second, the intermediate value is obtained by multiplying the ratio with the difference between the i-th timestamp and the first timestamp. Finally, the horizontal coordinate of the position coordinates at the i-th timestamp is obtained by adding the intermediate value to the horizontal coordinate of the first position. It can be understood that the calculation method of the vertical coordinate of the position coordinates at the i-th timestamp is the same as the calculation method of the horizontal coordinate of the position coordinates at the i-th timestamp, and the calculation method of the speed at the i-th timestamp is similar, which will not be described here for brevity.

[0047] For example, the horizontal coordinate Xi of the position coordinates at the i-th timestamp is (X2-X1) / (T2-T1)*(t-T1)+X1. Similarly, the vertical coordinate Yi of the position coordinates at the i-th timestamp is obtained, and then the position coordinates (Xi, Yi) at the i-th timestamp are obtained, which are inserted into the position at the i-th timestamp between T1 and T2.

[0048] S220, taking any one of the position coordinates at the respective timestamps and the corresponding position coordinates as the reference timestamp and the reference position.

[0049] For example, in some embodiments of the present application, because the positioning error of the target in the near range is small, the time delay is calibrated by taking the range close to the roadside camera within the roadside perception range of the true value car, for example, by taking the target of 0-60 m for calibration. Then, any one of the position coordinates at the respective timestamps of the true value car 100 is selected as the reference timestamp Tc and the reference position Dc.

[0050] Since the speed of the ground truth vehicle 100 is low, it also affects the accuracy of the obtained target latency value, so in some embodiments of the present application, before performing S220, the speed of the ground truth vehicle at each timestamp is first filtered, and the speed lower than the preset speed threshold is filtered. The preset speed threshold can be set according to the actual situation, which is not limited in the present application.

[0051] S230, based on the reference timestamp, finding a target perception frame matching the reference position from the roadside perception frame, and taking the difference between the reference timestamp and the timestamp corresponding to the target perception frame as the target latency value.

[0052] In some embodiments of the present application, before performing S230, the method of roadside test latency further comprises: obtaining the timestamp and positioning information of the target vehicle in the roadside perception frame.

[0053] For example, the roadside camera of the roadside perception end 200 detects the target vehicle on the road, obtains the positioning information of the target vehicle according to the calibration file or model prediction, records the timestamp and positioning information, and sends them to the data processing end 300. The positioning coordinates are the center of the vehicle bottom surface in the world coordinate system.

[0054] The above process is exemplarily described below.

[0055] In some embodiments of the present application, S230 can include:

[0056] S231, obtaining a corresponding test frame set existing in the roadside perception frame at the test timestamp T, wherein the test frame set includes a plurality of test frames corresponding to a plurality of test time intervals, and T is the sum of the reference timestamp Tc and n*t, wherein t is a test time interval, n is the number of time intervals and is a positive integer.

[0057] For example, in some embodiments of the present application, the reference timestamp of the ground truth vehicle 100 is Tc, t=1ms, then it is judged whether there is a test frame corresponding to the timestamp Tc+1ms in the roadside perception frame of T=Tc+1ms. If there is, the test frame is added to the test frame set. If not, it is considered that the matching fails, and then T=T+1ms or -1ms is updated to find whether there is a test frame in the roadside perception frame. That is, through T=Tc±n*1ms, it is found whether there is a test frame corresponding to the timestamp T in the roadside perception frame, so as to obtain the test frame set.

[0058] It should be noted that there can be some interference frames (e.g., no target vehicle or too many vehicles) when searching from the roadside perception frame. Therefore, the test timestamp can be accumulated for multiple times of searching to obtain multiple test frames (i.e., a test frame set).

[0059] S232, detecting each test frame in the test frame set to obtain a to-be-matched frame containing the target vehicle.

[0060] For example, in some embodiments of the present application, since there is uncertainty in the specific situation of the roadside vehicle condition, it is necessary to detect the test frame to obtain a to-be-matched frame containing the target vehicle. For example, if it is detected that the test frame contains the target vehicle, the frame is taken as the to-be-matched frame. If the number of to-be-matched frames obtained is multiple frames, the frame with the earliest timestamp in the time sequence is taken as the to-be-matched frame.

[0061] S233, comparing the target vehicle in the to-be-matched frame with the true value vehicle to determine the target perception frame.

[0062] In some embodiments of the present application, S233 can include: when it is determined that the type of the target vehicle is the same as that of the true value vehicle, and it is confirmed that there is no other vehicle in the preset range of the target vehicle in the to-be-matched frame, and the position of the target vehicle meets the preset condition with the reference position, the to-be-matched frame is taken as the target perception frame.

[0063] For example, in some embodiments of the present application, it is first determined whether the type of the target vehicle is the same as that of the true value vehicle 100. For example, the target vehicle is a truck, and the true value vehicle 100 is a car, at this time it can be determined that the types of the two are different, and the to-be-matched frame is invalid. If the type of the target vehicle is the same as that of the true value vehicle 100, it is necessary to determine whether there is other vehicle in the preset range. For example, the preset range is 2m, if there is other vehicle within the range of 2m around the target vehicle, the to-be-matched frame is invalid. If there is only the target vehicle within the range of 2m, it is necessary to compare the positioning coordinates of the roadside perception output of the target vehicle with the reference position Dc of the true value vehicle 100, if the distance between the two meets the preset condition (e.g., the preset condition is 10cm, 20cm, etc.), it indicates that the two are very close, at this time the to-be-matched frame can be taken as the target perception frame.

[0064] It should be noted that in some embodiments of the present application, after confirming that the to-be-matched frame is invalid, the reference timestamp and the reference position can be reselected for again or multiple times of matching to obtain the to-be-matched frame from the roadside perception frame until the target perception frame is obtained. In addition, the preset range and the preset condition can be set according to the actual situation, which is not limited in the present application.

[0065] To improve the accuracy of the obtained target latency value, in some embodiments of the present application, S230 can further include: taking the difference between the reference timestamp and the timestamp corresponding to the target perception frame as a calibration latency; obtaining at least two frames in which the target vehicle exists in the road-side perception frame; compensating the corresponding time of each of the at least two frames by the calibration latency respectively to obtain each compensated time; searching for the positioning coordinates obtained by the true value car self-positioning corresponding to each compensated time; if it is confirmed that the positioning coordinates obtained by the true value car self-positioning match the target vehicle successfully, taking the calibration latency as the target latency value.

[0066] For example, in some embodiments of the present application, the difference between the reference timestamp of the true value car 100 and the timestamp corresponding to the target perception frame is taken as a provisional latency Td (as one specific example of the calibration latency). Then, the image frames in which the id of the same target vehicle exists in the later frames of the road-side perception frame are searched. After compensating the time corresponding to the image frame by Td respectively, the positioning coordinates of the true value car at the compensated time in the image frame output by the true value car 100 are searched. If the positioning coordinates can match the positioning coordinates of the road-side perception end 200, it is confirmed that Td is the target latency value.

[0067] In some embodiments of the present application, S230 can further include:

[0068] S234, the following operations are repeatedly performed until the number of the target perception frame set reaches a preset value:

[0069] updating the reference timestamp and updating the reference position; searching for the target perception frame from the road-side perception frame and adding it to the target perception frame set.

[0070] For example, in some embodiments of the present application, in order to improve the accuracy of the obtained target latency value, a plurality of target perception frames can be obtained to confirm the target latency value. For example, Tc can be updated by 1ms in each loop to update the parameter timestamp, and the position coordinates corresponding to the parameter timestamp can be updated to achieve the purpose of updating the reference position. Then, the target perception frame set is obtained through the implementation process in S220-S230.

[0071] S235, the difference between the timestamp of each target perception frame in the target perception frame set and the corresponding reference timestamp is obtained to obtain each latency; the latency that appears most frequently in each latency is taken as the target latency value.

[0072] For example, in some embodiments of the present application, a plurality of delay values can be recorded by solving the difference between the target perception frame set and the corresponding reference timestamp. Finally, the number of occurrences of each delay value in the plurality of delay values is counted, and the corresponding delay value with the highest number of occurrences is taken as the final target delay value of the calibration output.

[0073] The specific process of the roadside test delay provided by some embodiments of the present application will be described below in conjunction with the accompanying Figure 3 The specific process of the roadside test delay provided by some embodiments of the present application will be described below in conjunction with the accompanying

[0074] The specific process of the roadside test delay provided by some embodiments of the present application will be described below in conjunction with the accompanying Figure 3 , Figure 3 A method flowchart of the roadside test delay provided by some embodiments of the present application.

[0075] It should be noted that the data processing end 300 can process the time and positioning information of the vehicle returned by the real-time roadside perception end 200 and the RTK-equipped true value vehicle 100 in real time to obtain the target delay value between the two, or can process offline to obtain the target delay value. The specific process of the roadside test delay will be described below in the scenario of the data processing end 300 being a cloud processing device processing the time and positioning information of the vehicle returned by the roadside perception end 200 and the true value vehicle 100 offline to obtain the target delay value.

[0076] S310, obtaining the initial positioning information of the true value vehicle at adjacent two frames of images, and based on the initial positioning information, obtaining the position coordinates of the true value vehicle at each timestamp between the first timestamp and the second timestamp and the speed at each timestamp.

[0077] S320, determining the reference timestamp Tc and the reference position.

[0078] S330, determining whether there is a test frame corresponding to Tc in the roadside perception frame, if yes, executing S340, otherwise executing S331.

[0079] S331, Tc=Tc+1, returning to S330.

[0080] S340, detecting the test frame to obtain a to-be-matched frame containing the target vehicle.

[0081] S350, determining whether the type of the target vehicle in the to-be-matched frame is the same as the true value vehicle, if yes, executing S360, otherwise returning to S331.

[0082] S360, determining whether there is another vehicle in the preset range of the target vehicle, if no, executing S370, otherwise executing S331.

[0083] S370, determining whether the position of the target vehicle and the reference position satisfy a preset condition, if yes, performing S380, otherwise performing S331.

[0084] S380, adding the target perception frame to the target perception frame set.

[0085] S381, determining whether the number of the target perception frame set is greater than M, if yes, performing S390, otherwise setting Tc = Tc + 1 and returning to S320.

[0086] S390, obtaining the difference between the timestamp of each target perception frame in the target perception frame set and the corresponding reference timestamp, to obtain each time delay. And the time delay that appears most frequently in each time delay is taken as the target time delay value.

[0087] It should be noted that the specific implementation process of S310-S391 can refer to the method embodiment provided in the foregoing Figure 2 For brevity, the detailed description is appropriately omitted here.

[0088] Please refer to Figure 4 , Figure 4 The composition block diagram of the roadside test time delay device provided by some embodiments of the present application is shown. It should be understood that the roadside test time delay device corresponds to the method embodiments described above, and can perform each step involved in the method embodiments described above. The specific functions of the roadside test time delay device can be referred to the description in the foregoing, and for brevity, the detailed description is appropriately omitted here.

[0089] Figure 4 The roadside test time delay device includes at least one software function module which can be stored in the form of software or firmware in the memory or solidified in the roadside test time delay device. The roadside test time delay device includes: a data acquisition module 410 configured to acquire positioning information obtained by the true value vehicle itself positioning in a fixed time period, wherein the positioning information includes: position coordinates and speed at each timestamp in the fixed time period; a data selection module 420 configured to take any one of the position coordinates at each timestamp and the corresponding position coordinates as a reference timestamp and a reference position; a data matching module 430 configured to find a target perception frame matched with the reference position from the roadside perception frame based on the reference timestamp, and take the difference between the reference timestamp and the timestamp corresponding to the target perception frame as a target time delay value.

[0090] In some embodiments of the present application, the data acquisition module 410 is configured to acquire initial positioning information of the ground truth vehicle in two adjacent frames of images, wherein the time interval of the two adjacent frames of images is the fixed time period, and the initial positioning information includes a first timestamp, a first position and a first speed of the ground truth vehicle in a first frame of image, and a second timestamp, a second position and a second speed of the ground truth vehicle in a second frame of image; based on the initial positioning information, the position coordinates of the ground truth vehicle at each timestamp between the first timestamp and the second timestamp and the speed at each timestamp are acquired.

[0091] In some embodiments of the present application, the fixed time period is determined by the frame rate of a real-time dynamic carrier phase difference measurement instrument deployed by the ground truth vehicle.

[0092] In some embodiments of the present application, the data matching module 430 is configured to acquire a corresponding test frame set existing in the roadside perception frame at a test timestamp T, wherein the test frame set includes a plurality of test frames corresponding to a plurality of test time intervals, and the T is the sum of the reference timestamp Tc and n*t, wherein t is a test time interval, n is the number of time intervals and is a positive integer; each test frame in the test frame set is detected to obtain a to-be-matched frame containing a target vehicle; the target vehicle in the to-be-matched frame is compared with the ground truth vehicle to determine the target perception frame.

[0093] In some embodiments of the present application, the data matching module 430 is configured to, when it is determined that the target vehicle and the ground truth vehicle are of the same type, and it is confirmed that there is no other vehicle in a preset range of the target vehicle in the to-be-matched frame, and the position of the target vehicle and the reference position satisfy a preset condition, then the to-be-matched frame is taken as the target perception frame.

[0094] In some embodiments of the present application, the data matching module 430 is configured to take the difference between the reference timestamp and the timestamp corresponding to the target perception frame as a calibration delay; acquire at least two frames in which the target vehicle exists in the roadside perception frame; compensate the corresponding time of each frame in the at least two frames by the calibration delay respectively to obtain each compensation time; find the positioning coordinates obtained by the self-positioning of the ground truth vehicle corresponding to each compensation time; if it is confirmed that the positioning coordinates obtained by the self-positioning of the ground truth vehicle match the target vehicle successfully, then the calibration delay is taken as the target delay value.

[0095] In some embodiments of the present application, the data matching module 430 is configured to perform the following operations in a loop until the number of the target perception frame set reaches a preset value: updating the reference timestamp and updating the reference position; finding the target perception frame from the roadside perception frame and adding it to the target perception frame set; calculating the difference between the timestamp of each target perception frame in the target perception frame set and the corresponding reference timestamp to obtain each time delay; and taking the time delay that appears most frequently in the time delays as the target time delay value.

[0096] Some embodiments of the present application also provide a computer readable storage medium having a computer program stored thereon, wherein the program, when executed by a processor, can implement the operations of the method corresponding to any of the above embodiments.

[0097] Some embodiments of the present application also provide a computer program product, which comprises a computer program, wherein the computer program, when executed by a processor, can implement the operations of the method corresponding to any of the above embodiments.

[0098] As shown in Figure 5 Some embodiments of the present application provide an electronic device 500, which comprises a memory 510, a processor 520, and a computer program stored in the memory 510 and executable on the processor 520, wherein the processor 520 reads the program from the memory 510 through a bus 530 and executes the program to implement the method of any of the above embodiments.

[0099] The processor 520 can process digital signals and can include various computing structures, such as a complex instruction set computer structure, a reduced instruction set computer structure, or a structure implementing a combination of multiple instruction sets. In some examples, the processor 520 can be a microprocessor.

[0100] The memory 510 can be used to store instructions executed by the processor 520 or data related to the execution of the instructions. These instructions and / or data can include code for implementing some or all of the functions of one or more modules described in the embodiments of the present application. The processor 520 of the embodiments of the present disclosure can be used to execute the instructions in the memory 510 to implement the above-described method. The memory 510 includes a dynamic random access memory, a static random access memory, a flash memory, an optical memory, or other memories well known to those skilled in the art.

[0101] The above merely provides an example of the present application and is not intended to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application. It should be noted that similar reference numerals and letters represent similar items in the following drawings, and thus, once an item is defined in one drawing, it need not be further defined and explained in subsequent drawings.

[0102] The above merely provides an example of the present application and is not intended to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application. It should be noted that similar reference numerals and letters represent similar items in the following drawings, and thus, once an item is defined in one drawing, it need not be further defined and explained in subsequent drawings.

[0103] It should be noted that the relational terms herein such as first and second and the like are used solely to distinguish one from another entity or action without necessarily requiring or implying any actual relationship or order between such entities or actions. Moreover, the terms "comprises", "comprising", or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises... a" does not, without more constraints, exclude the existence of additional identical elements in the process, method, article, or apparatus that comprises the element.

Claims

1. A method of roadside testing latency, the method comprising: The method comprises the following steps: acquiring positioning information obtained by self-positioning of a true value vehicle within a fixed time period, wherein the positioning information comprises position coordinates and speed at each timestamp within the fixed time period; taking any one of the position coordinates at each timestamp and the corresponding position coordinate as a reference timestamp and a reference position; based on the reference timestamp, searching for a target perception frame matching the reference position from a roadside perception frame, and taking a difference value between the reference timestamp and a timestamp corresponding to the target perception frame as a target time delay value; The method comprises: obtaining a corresponding test frame set existing in the roadside perception frame at a test timestamp T, wherein the test frame set comprises a plurality of test frames corresponding to a plurality of test time intervals, and the T is a sum of the reference timestamp Tc and n times of a test time interval t, wherein t is the test time interval, n is the number of time intervals and is a positive integer; detecting each test frame in the test frame set to obtain a to-be-matched frame containing a target vehicle; and comparing the target vehicle in the to-be-matched frame with the true value vehicle to determine the target perception frame.

2. The method of claim 1, wherein, the acquiring of the positioning information obtained by self-positioning of the true value vehicle within the fixed time period comprises: acquiring initial positioning information of the true value vehicle at adjacent two frames of images, wherein a time interval of the adjacent two frames of images is the fixed time period, and the initial positioning information comprises a first timestamp of a first frame of image, a first position and a first speed of the true value vehicle, and a second timestamp of a second frame of image, a second position and a second speed of the true value vehicle; based on the initial positioning information, acquiring position coordinates at each timestamp between the first timestamp and the second timestamp of the true value vehicle and speed at each timestamp.

3. The method of claim 1 or 2, wherein, The fixed time period is determined by a frame rate of a real-time dynamic carrier phase difference measurement instrument deployed by the true value vehicle.

4. The method of claim 1, wherein, The method further comprises the following steps before taking the difference value between the reference timestamp and the timestamp corresponding to the target perception frame as the target time delay value: taking the difference value between the reference timestamp and the timestamp corresponding to the target perception frame as a calibration time delay; 5. The method of claim 4, wherein, acquiring at least two frames in which the target vehicle exists in the roadside perception frames; compensating the calibration time delay from respective timestamps of the at least two frames to obtain respective compensation times; searching for positioning coordinates obtained by self-positioning of the true value vehicle corresponding to the respective compensation times; if it is confirmed that the positioning coordinates obtained by self-positioning of the true value vehicle match the target vehicle successfully, taking the calibration time delay as the target time delay value. The method further comprises the following steps before taking the difference value between the reference timestamp and the timestamp corresponding to the target perception frame as the target time delay value: performing the following operations in a loop until a number of a target perception frame set reaches a preset value:

6. The method of claim 1 or 2, wherein, updating the reference timestamp and updating the reference position; searching for the target perception frame from the roadside perception frame and adding the target perception frame to the target perception frame set; wherein, the taking of the difference value between the reference timestamp and the timestamp corresponding to the target perception frame as the target time delay value comprises: obtaining respective time delays by calculating differences between timestamps of each target perception frame in the target perception frame set and corresponding reference timestamps. ​ ​ The target time delay value is the time delay that appears most frequently in the time delays.

7. An apparatus for roadside testing of latency, the apparatus comprising: The apparatus is configured to perform the method of claim 1, comprising: a data acquisition module configured to acquire positioning information obtained by a true value vehicle self-positioning in a fixed time period, wherein the positioning information comprises position coordinates and speed at each timestamp in the fixed time period; a data selection module configured to select any one of the position coordinates at each timestamp and the corresponding position coordinate as a reference timestamp and a reference position; a data matching module configured to find a target perception frame matching the reference position from a roadside perception frame based on the reference timestamp, and take a difference between the reference timestamp and a timestamp corresponding to the target perception frame as a target time delay value.

8. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, wherein the computer program is run by a processor to perform the method of any one of claims 1-6.

9. An electronic device, comprising: A computer readable storage medium stores a computer program, wherein the computer program is run by a processor to perform the method of any one of claims 1-6. A computer readable storage medium stores a computer program, wherein the computer program is run by a processor to perform the method of any one of claims 1-6.

Images