A testing method, device, system and medium for the accuracy of an object classification algorithm

Abstract

An embodiment of the present invention discloses a method, device, system, and medium for testing the accuracy of a target classification algorithm. The method is executed by a server, and the method includes: obtaining the relative position of the algorithm between a first vehicle and a second vehicle, the motion behavior data of the first vehicle, and the motion behavior data of the second vehicle; the relative position of the algorithm is determined based on the target classification algorithm according to the motion behavior data of the first vehicle and the motion behavior data of the second vehicle; based on a high-precision map, determining the standard relative position between the first vehicle and the second vehicle according to the motion behavior data of the first vehicle and the motion behavior data of the second vehicle; determining the accuracy of the target classification algorithm according to the standard relative position and the relative position of the algorithm. The embodiment of the present invention introduces the standard relative position of the high-precision map, which can quickly test and verify the accuracy of the target classification algorithm, facilitate the development and optimization of the target algorithm, and improve the test efficiency.

Description

Technical Field

[0001] The embodiments of the present invention relate to the fields of computer technology and communication, and in particular, to a method, device, system, and medium for testing the accuracy of a target classification algorithm. Background Art

[0002] The V2X (Vehicle to X) warning process can be divided into two important stages from the algorithm: target classification and danger judgment. The host vehicle first needs to judge the relative position of the target vehicle with respect to the host vehicle based on the position and driving behavior-related information transmitted by the remote vehicle through V2X communication, that is, target classification. If the host vehicle's target classification judgment of the remote vehicle is inaccurate, it will lead to false triggering or non-triggering of warnings. During the current V2X in-vehicle test process, when the accuracy of the target classification algorithm is not fully guaranteed, it will greatly affect the test efficiency. Summary of the Invention

[0003] The embodiments of the present invention provide a method, device, system, and medium for testing the accuracy of a target classification algorithm to achieve the judgment of the accuracy of the target classification algorithm.

[0004] In a first aspect, the embodiments of the present invention provide a method for testing the accuracy of a target classification algorithm, which is executed by a server, and the method includes:

[0005] Obtain the algorithm relative position, the first vehicle motion behavior data, and the second vehicle motion behavior data between a first vehicle and a second vehicle; the algorithm relative position is determined based on the target classification algorithm according to the first vehicle motion behavior data and the second vehicle motion behavior data;

[0006] Based on the high-precision map, determine the standard relative position between the first vehicle and the second vehicle according to the first vehicle motion behavior data and the second vehicle motion behavior data;

[0007] Determine the accuracy of the target classification algorithm according to the standard relative position and the algorithm relative position.

[0008] In a second aspect, the embodiments of the present invention further provide a method for testing the accuracy of a target classification algorithm, which is executed by the first vehicle, and the method includes:

[0009] Based on the target classification algorithm, determine the algorithm relative position between the first vehicle and the second vehicle according to the first vehicle motion behavior data and the second vehicle motion behavior data;

[0010] Send the first vehicle motion behavior data, the second vehicle motion behavior data, and the algorithm relative position to the server, and the server performs the following: Based on the high-precision map, determine the standard relative position between the first vehicle and the second vehicle according to the first vehicle motion behavior data and the second vehicle motion behavior data; Determine the accuracy of the target classification algorithm according to the standard relative position and the algorithm relative position.

[0011] In a third aspect, an embodiment of the present invention further provides a test device for the accuracy of a target classification algorithm. The device is executed by a server, and the device includes:

[0012] A vehicle information acquisition module, configured to acquire the algorithm relative position, the first vehicle motion behavior data, and the second vehicle motion behavior data between the first vehicle and the second vehicle; The algorithm relative position is determined based on the target classification algorithm according to the first vehicle motion behavior data and the second vehicle motion behavior data;

[0013] A standard relative position determination module, configured to determine the standard relative position between the first vehicle and the second vehicle based on the high-precision map according to the first vehicle motion behavior data and the second vehicle motion behavior data;

[0014] An accuracy determination module, configured to determine the accuracy of the target classification algorithm according to the standard relative position and the algorithm relative position.

[0015] In a fourth aspect, an embodiment of the present invention further provides a test device for the accuracy of a target classification algorithm. The device is executed by the first vehicle, and the device includes:

[0016] An algorithm relative position determination module, configured to determine the algorithm relative position between the first vehicle and the second vehicle based on the target classification algorithm according to the first vehicle motion behavior data and the second vehicle motion behavior data;

[0017] A vehicle information sending module, configured to send the first vehicle motion behavior data, the second vehicle motion behavior data, and the algorithm relative position to the server, and the server performs the following: Based on the high-precision map, determine the standard relative position between the first vehicle and the second vehicle according to the first vehicle motion behavior data and the second vehicle motion behavior data; Determine the accuracy of the target classification algorithm according to the standard relative position and the algorithm relative position.

[0018] In a fifth aspect, an embodiment of the present invention further provides an electronic device, and the electronic device includes:

[0019] One or more processors;

[0020] A storage device, configured to store one or more programs,

[0021] When the one or more programs are executed by the one or more processors, the one or more processors implement the test method for the accuracy of the target classification algorithm as described in any embodiment of the present invention.

[0022] In a sixth aspect, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the test method for the accuracy of the target classification algorithm as described in any embodiment of the present invention.

[0023] In the embodiments of the present invention, by obtaining the relative position between the first vehicle and the second vehicle in the algorithm, and at the same time, based on the high-precision map, determining the standard relative position between the first vehicle and the second vehicle according to the motion behavior data of the first vehicle and the motion behavior data of the second vehicle, and by comparing the standard relative position with the relative position in the algorithm, the accuracy of the target classification algorithm is determined, so that the accuracy of the target classification algorithm can be quickly tested and verified, which is convenient for developing and optimizing the target algorithm, reducing the warning failure caused by inaccurate target classification algorithm in the real vehicle test process, and improving the test efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a flowchart of a test method for the accuracy of a target classification algorithm provided in Embodiment 1 of the present invention;

[0025] Figure 2 It is a flowchart of a test method for the accuracy of a target classification algorithm provided in Embodiment 2 of the present invention;

[0026] Figure 3 It is a flowchart of a test method for the accuracy of a target classification algorithm provided in Embodiment 3 of the present invention;

[0027] Figure 4 It is a schematic structural diagram of a test device for the accuracy of a target classification algorithm provided in Embodiment 4 of the present invention;

[0028] Figure 5 It is a schematic structural diagram of a test device for the accuracy of a target classification algorithm provided in Embodiment 5 of the present invention;

[0029] Figure 6 It is a schematic structural diagram of an electronic device provided in Embodiment 6 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0030] The present invention will be further described in detail below with reference to the drawings and embodiments. It can be understood that the specific embodiments described herein are only used to explain the present invention, rather than limiting the present invention. Additionally, it should be noted that for the convenience of description, only parts related to the present invention are shown in the drawings rather than all the structures.

[0031] Embodiment 1

[0032] Figure 1 The flowchart of a method for testing the accuracy of a target classification algorithm provided in the first embodiment of the present invention. This embodiment is applicable to the case of testing the accuracy of the target classification algorithm in a V2X real vehicle test. This method can be executed by a test device for the accuracy of the target classification algorithm, and this device can be implemented in a software and / or hardware manner. This device can be configured in a server and / or a vehicle-mounted controller. When this method is executed by the server, it specifically includes:

[0033] S110. Obtain the relative algorithm position between the first vehicle and the second vehicle, the motion behavior data of the first vehicle, and the motion behavior data of the second vehicle.

[0034] Among them, both the first vehicle and the second vehicle are vehicles capable of implementing V2X technology. The relative algorithm position refers to the relative position between the first vehicle and the second vehicle determined by the target classification algorithm. The relative algorithm position is determined based on the target classification algorithm according to the motion behavior data of the first vehicle and the motion behavior data of the second vehicle. At least one of the first vehicle and the second vehicle can implement the target classification algorithm. The motion behavior data of the first vehicle refers to the motion behavior data of the first vehicle during driving, which at least includes data such as the first vehicle timestamp, the first vehicle position information, the first vehicle heading angle, and the first vehicle identity identifier. The first vehicle timestamp refers to the time when the first vehicle triggers the target classification algorithm. The first vehicle position information refers to the longitude and latitude where the first vehicle is located. The first vehicle heading angle refers to the angle between the longitudinal axis of the vehicle and the North Pole of the Earth. The motion behavior data of the second vehicle refers to the motion behavior data of the second vehicle during driving, which at least includes data such as the second vehicle timestamp, the second vehicle position information, the second vehicle heading angle, and the second vehicle identity identifier. The motion behavior data of the first vehicle and the motion behavior data of the second vehicle can be obtained from the LOG (log) information printed in real time by the T-BOX (Telematics BOX, remote information processor).

[0035] Specifically, after the test starts, when the first vehicle detects the second vehicle, the target classification algorithm on the first vehicle is triggered. At this time, the first vehicle obtains the motion behavior data of the second vehicle through V2X technology. Meanwhile, the T-BOX on the first vehicle sends the LOG information of the first vehicle to the server in real time through the in-vehicle unit. The server obtains information such as the relative position of the algorithms between the first vehicle and the second vehicle, the motion behavior data of the first vehicle, and the motion behavior data of the second vehicle from the LOG information of the first vehicle. Similarly, the second vehicle can achieve the same functions as the first vehicle. That is, after the test starts, when the second vehicle detects the first vehicle, the target classification algorithm on the second vehicle is triggered. At this time, the second vehicle obtains the motion behavior data of the first vehicle through V2X technology. Meanwhile, the T-BOX on the second vehicle sends the LOG information of the second vehicle to the server in real time through the in-vehicle unit. The server obtains information such as the relative position of the algorithms between the first vehicle and the second vehicle, the motion behavior data of the first vehicle, and the motion behavior data of the second vehicle from the LOG information of the second vehicle.

[0036] Further, after the test starts, when the first vehicle detects the second vehicle, the first vehicle obtains the motion behavior data of the first vehicle and the motion behavior data of the second vehicle sent by the second vehicle through V2X technology, and determines whether the second vehicle may pose a threat to the driving safety of the first vehicle based on the heading angle in the motion behavior data of the first vehicle and the heading angle in the motion behavior data of the second vehicle. If so, the target classification algorithm is triggered; otherwise, the target classification algorithm is not triggered. Similarly, the second vehicle can achieve the same functions as the first vehicle.

[0037] S120. Based on the high-precision map, determine the standard relative position between the first vehicle and the second vehicle according to the motion behavior data of the first vehicle and the motion behavior data of the second vehicle.

[0038] Among them, the standard relative position refers to the relative position between the first vehicle and the second vehicle determined through the high-precision map. The high-precision map determines the standard relative position according to the motion behavior data of the first vehicle and the motion behavior data of the second vehicle.

[0039] Specifically, the high-precision map is stored in the server. After receiving the LOG information sent by the first vehicle or the second vehicle, the server parses and stores the LOG information. The server extracts the motion behavior data of the first vehicle and the motion behavior data of the second vehicle from the parsed LOG information, marks points on the high-precision map based on the motion behavior data of the first vehicle and the motion behavior data of the second vehicle, and determines the relative position between the first vehicle and the second vehicle again according to the positions of the marked points on the high-precision map, which is used as the standard relative position.

[0040] S130. Determine the accuracy of the target classification algorithm according to the standard relative position and the relative position of the algorithms.

[0041] The accuracy of the target classification algorithm refers to the accuracy of the target classification algorithm in processing data. The accuracy of the target classification algorithm can be judged by the relative position of the algorithm. It can be understood that by comparing the standard relative position and the relative position of the algorithm, and based on the gap between the relative position of the algorithm and the standard relative position, it is determined whether the relative position of the algorithm is accurate, and then the accuracy of the target classification algorithm is determined.

[0042] Furthermore, determining the accuracy of the target classification algorithm according to the standard relative position and the relative position of the algorithm includes: comparing whether the standard relative position and the relative position of the algorithm are consistent; if they are consistent, it is determined that the target classification algorithm in the test record to which the standard relative position and the relative position of the algorithm belong is accurate; otherwise, it is determined that the target classification algorithm in the test record to which the standard relative position and the relative position of the algorithm belong is incorrect; determining the proportion of accurate test records in at least two test records, and obtaining the accuracy of the target classification algorithm according to the proportion of accurate test records.

[0043] Among them, the test record refers to the LOG information printed in real time by the T-BOX of the first vehicle or the T-BOX of the second vehicle when the target classification algorithm is triggered. Each test record can be located according to the timestamp in the LOG information. The proportion of accurate test records refers to the proportion of the test records in which the target classification algorithm is accurate among all the counted test records. The accuracy of the target classification algorithm refers to the probability that the target classification algorithm is accurate in the test record.

[0044] Specifically, in a test record, based on the high-precision map, the standard relative position is obtained according to the movement behavior data of the first vehicle and the movement behavior data of the second vehicle in the test record. By comparing the standard relative position and the relative position of the algorithm, if the relative position of the algorithm is consistent with the standard relative position, it means that in this test record, the relative position of the algorithm determined by the target classification algorithm according to the movement behavior data of the first vehicle and the movement behavior data of the second vehicle is accurate, and it is further determined that the target classification algorithm in this test record is accurate. Otherwise, it is determined that the target classification algorithm in this test record is incorrect. By counting multiple test records, the proportion of accurate test records in all test records is determined, and then the accuracy of the target classification algorithm is determined.

[0045] By comparing the standard relative position and the relative position of the algorithm, it is determined whether the target algorithm is accurate, and according to the proportion of accurate test records in at least two test records, the accuracy of the target classification algorithm is obtained, which is convenient for developers and designers to analyze and optimize the target classification algorithm, and then reduce the misjudgment caused by the inaccurate target classification algorithm.

[0046] In the technical solution of the embodiment of the present invention, by obtaining the algorithmic relative position between the first vehicle and the second vehicle, and at the same time based on the high-precision map, determining the standard relative position between the first vehicle and the second vehicle according to the first vehicle motion behavior data and the second vehicle motion behavior data, and by comparing the standard relative position and the algorithmic relative position, the accuracy of the target classification algorithm can be determined, and the accuracy of the target classification algorithm can be quickly tested and verified, which is convenient for developing and optimizing the target algorithm, reducing the warning failure caused by inaccurate target classification algorithms during the real vehicle test process, and improving the test efficiency.

[0047] On the basis of the above technical solution, based on the high-precision map, determining the standard relative position between the first vehicle and the second vehicle according to the first vehicle motion behavior data and the second vehicle motion behavior data includes: based on the high-precision map, determining the first marked position information of the first vehicle in the high-precision map according to the first vehicle position information and the first vehicle heading angle in the first vehicle motion behavior data; determining the second marked position information of the second vehicle in the high-precision map according to the second vehicle position information and the second vehicle heading angle in the second vehicle motion behavior data; and determining the standard relative position between the first vehicle and the second vehicle according to the first marked position information and the second marked position information.

[0048] Among them, the first marked position information refers to the position information marked by the first vehicle in the high-precision map. The first marked position information is determined by the high-precision map according to the first vehicle position information and the first vehicle heading angle in the first vehicle motion behavior data. The second marked position information refers to the position information marked by the second vehicle in the high-precision map. The second marked position information is determined by the high-precision map according to the second vehicle position information and the second vehicle heading angle in the second vehicle motion behavior data.

[0049] Specifically, the high-precision map determines the first marked position information according to the first vehicle motion behavior data and determines the second marked position information according to the second vehicle motion behavior data. When the high-precision map determines the first marked position information and the first marked position information, it can determine the information of the lanes where the first vehicle and the second vehicle are located. According to the first marked position information and the second marked position information, and combining the lane information, the standard relative position between the first vehicle and the second vehicle is further determined.

[0050] Determining the first marked position information and the second marked position information through the high-precision map, and then determining the standard relative position between the first vehicle and the second vehicle provides a basis for judging the accuracy of the target classification algorithm. At the same time, using the same data as the target classification algorithm reduces data collection and can quickly test and verify the accuracy of the target classification algorithm.

[0051] Embodiment 2

[0052] Figure 2The flowchart of a method for testing the accuracy of a target classification algorithm provided in the second embodiment of the present invention. On the basis of the above embodiments, the present embodiment further expands the way for the server to obtain data. The specific method is as follows:

[0053] S210. Obtain the relative position of the algorithm between the first vehicle and the second vehicle, and the motion behavior data of the first vehicle from the first vehicle.

[0054] Wherein, the relative position of the algorithm is determined by the first vehicle or the second vehicle based on the target classification algorithm according to the motion behavior data of the first vehicle and the motion behavior data of the second vehicle.

[0055] Specifically, when the first vehicle detects the second vehicle during driving and triggers the target classification algorithm, the first vehicle obtains the motion behavior data of the first vehicle. At the same time, the first vehicle obtains the motion behavior data of the second vehicle from the second vehicle through V2X technology. The target classification algorithm in the first vehicle obtains the relative position of the algorithm between the first vehicle and the second vehicle according to the motion behavior data of the first vehicle and the obtained motion behavior data of the second vehicle. The first vehicle sends the relative position of the algorithm between the first vehicle and the second vehicle and the motion behavior data of the first vehicle to the server. Similarly, the second vehicle can achieve the same function as the first vehicle.

[0056] S220. Obtain the motion behavior data of the second vehicle from the second vehicle.

[0057] Wherein, when the first vehicle triggers the target classification algorithm, when the second vehicle sends the motion behavior data of the second vehicle to the first vehicle, it can also send the motion behavior data of the second vehicle to the server at the same time. When the server receives the data sent by the first vehicle, it simultaneously receives the motion behavior data of the second vehicle sent by the second vehicle.

[0058] S230. Based on the high-precision map, determine the standard relative position between the first vehicle and the second vehicle according to the motion behavior data of the first vehicle and the motion behavior data of the second vehicle.

[0059] S240. Determine the accuracy of the target classification algorithm according to the standard relative position and the relative position of the algorithm.

[0060] The technical solution of the embodiment of the present invention obtains the motion behavior data from the first vehicle and the second vehicle respectively through the server, which speeds up the data sending time and data volume, and reduces the time for the server to verify the accuracy of the target classification algorithm.

[0061] Embodiment Three

[0062] Figure 3The flowchart of a method for testing the accuracy of a target classification algorithm provided in Embodiment 3 of the present invention is executed by a first vehicle, and the method specifically includes:

[0063] S310. Based on the target classification algorithm, determine the relative position between the first vehicle and the second vehicle according to the motion behavior data of the first vehicle and the motion behavior data of the second vehicle.

[0064] Among them, when the target classification algorithm of the first vehicle is triggered, the first vehicle acquires the motion behavior data of the first vehicle and the motion behavior data of the second vehicle, and uses the target classification algorithm to determine the relative position between the first vehicle and the second vehicle according to the motion behavior data of the first vehicle and the second motion behavior data.

[0065] S320. Send the motion behavior data of the first vehicle, the motion behavior data of the second vehicle, and the relative position of the algorithm to the server, and the server performs the following: Based on the high-precision map, determine the standard relative position between the first vehicle and the second vehicle according to the motion behavior data of the first vehicle and the motion behavior data of the second vehicle; determine the accuracy of the target classification algorithm according to the standard relative position and the relative position of the algorithm.

[0066] Among them, after obtaining the relative position of the algorithm through the target classification algorithm, the first vehicle sends the relative position of the algorithm together with the motion behavior data of the first vehicle and the motion behavior data of the second vehicle to the server. The server determines the standard relative position between the first vehicle and the second vehicle through the high-precision map according to the data sent by the first vehicle.

[0067] The technical solution of the embodiment of the present invention, when the target classification algorithm is triggered by the first vehicle, acquires the motion behavior data of the first vehicle and the motion behavior data of the second vehicle, determines the relative position of the algorithm of the target classification algorithm, and at the same time sends the acquired data and the relative position of the algorithm to the server, so that the server obtains the standard relative position, provides a data basis for judging the accuracy of the target classification algorithm of the first vehicle, and is convenient for testers to analyze and optimize the target classification algorithm.

[0068] Embodiment 4

[0069] Figure 4 The structural schematic diagram of a device for testing the accuracy of a target classification algorithm provided in Embodiment 4 of the present invention can execute the method for testing the accuracy of the target classification algorithm provided in any one of Embodiments 1 and 2 above. The device is executed by the server and can include: a vehicle information acquisition module 401, a standard relative position determination module 402, and an accuracy determination module 403.

[0070] Among them, the vehicle information acquisition module 401 is configured to acquire the algorithmic relative position between the first vehicle and the second vehicle, the first vehicle motion behavior data, and the second vehicle motion behavior data; the algorithmic relative position is determined based on a target classification algorithm according to the first vehicle motion behavior data and the second vehicle motion behavior data.

[0071] The standard relative position determination module 402 is configured to determine the standard relative position between the first vehicle and the second vehicle based on a high-precision map according to the first vehicle motion behavior data and the second vehicle motion behavior data.

[0072] The accuracy determination module 403 is configured to determine the accuracy of the target classification algorithm according to the standard relative position and the algorithmic relative position.

[0073] The technical solution of the embodiment of the present invention can quickly test and verify the accuracy of the target classification algorithm by acquiring the algorithmic relative position between the first vehicle and the second vehicle, and at the same time determining the standard relative position between the first vehicle and the second vehicle based on a high-precision map according to the first vehicle motion behavior data and the second vehicle motion behavior data, and comparing the standard relative position with the algorithmic relative position, which is convenient for developing and optimizing the target algorithm, reducing early warning failures caused by inaccurate target classification algorithms during real vehicle testing, and improving the testing efficiency.

[0074] In the above device, optionally, the standard relative position determination module 402 includes:

[0075] The first marked position information determination unit is configured to determine the first marked position information of the first vehicle in the high-precision map based on the high-precision map according to the first vehicle position information and the first vehicle heading angle in the first vehicle motion behavior data.

[0076] The second marked position information determination unit is configured to determine the second marked position information of the second vehicle in the high-precision map according to the second vehicle position information and the second vehicle heading angle in the second vehicle motion behavior data.

[0077] The standard relative position determination unit is configured to determine the standard relative position between the first vehicle and the second vehicle according to the first marked position information and the second marked position information.

[0078] In the above device, optionally, the accuracy determination module 403 includes:

[0079] A relative position comparison unit for comparing whether the standard relative position and the algorithm relative position are consistent; if they are consistent, it is determined that the target classification algorithm in the test record to which the standard relative position and the algorithm relative position belong is accurate; otherwise, it is determined that the target classification algorithm in the test record to which the standard relative position and the algorithm relative position belong is incorrect;

[0080] A target classification algorithm accuracy determination unit for determining the proportion of accurate test records in at least two test records and obtaining the accuracy of the target classification algorithm based on the proportion of accurate test records.

[0081] In the above device, optionally, the vehicle information acquisition module 401 includes:

[0082] A first vehicle information acquisition unit for acquiring the algorithm relative position between the first vehicle and the second vehicle, and the first vehicle motion behavior data from the first vehicle; the algorithm relative position is determined by the first vehicle or the second vehicle based on the target classification algorithm according to the first vehicle motion behavior data and the second vehicle motion behavior data;

[0083] A second vehicle information acquisition unit for acquiring the second vehicle motion behavior data from the second vehicle.

[0084] The test device for the accuracy of the target classification algorithm provided by the embodiments of the present invention can execute the test method for the accuracy of the target classification algorithm provided by any one of Embodiment 1 and Embodiment 2 of the present invention, and has the corresponding functional modules and beneficial effects for executing the method.

[0085] Embodiment 5

[0086] Figure 5 FIG. 22 is a schematic structural diagram of a test device for the accuracy of a target classification algorithm provided by Embodiment 5 of the present invention. The device can execute the test method for the accuracy of the target classification algorithm provided by the above Embodiment 3. The device is executed by the first vehicle and may include: an algorithm relative position determination module 501 and a vehicle information sending module 502.

[0087] Among them, the algorithm relative position determination module 501 is used to determine the algorithm relative position between the first vehicle and the second vehicle based on the target classification algorithm according to the first vehicle motion behavior data and the second vehicle motion behavior data;

[0088] The vehicle information sending module 502 is used to send the first vehicle motion behavior data, the second vehicle motion behavior data, and the algorithm relative position to the server. The server performs the following: determining the standard relative position between the first vehicle and the second vehicle based on the high-precision map according to the first vehicle motion behavior data and the second vehicle motion behavior data; determining the accuracy of the target classification algorithm according to the standard relative position and the algorithm relative position.

[0089] In the technical solution of the embodiment of the present invention, when the target classification algorithm is triggered by the first vehicle, the motion behavior data of the first vehicle and the motion behavior data of the second vehicle are obtained, the relative position of the algorithm of the target classification algorithm is determined, and at the same time, the obtained data and the relative position of the algorithm are sent to the server, so that the server obtains the standard relative position, providing a data basis for judging the accuracy of the target classification algorithm of the first vehicle, and facilitating testers to analyze and optimize the target classification algorithm.

[0090] Embodiment Six

[0091] Figure 6 It is a schematic structural diagram of an electronic device provided in Embodiment Six of the present invention. As Figure 6 shown, the electronic device includes a processor 60, a memory 61, an input device 62, and an output device 63; the number of processors 60 in the electronic device can be one or more, Figure 6 taking one processor 60 as an example; the processor 60, the memory 61, the input device 62, and the output device 63 in the electronic device can be connected through a bus or other means, Figure 6 taking the connection through the bus as an example.

[0092] The memory 61, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and modules, such as the program instructions and / or modules corresponding to the test method for the accuracy of the target classification algorithm in Embodiment Four of the present invention (for example, the vehicle information acquisition module 401, the standard relative position determination module 402, and the accuracy determination module 403) and the program instructions and / or modules corresponding to the test method for the accuracy of the target classification algorithm in Embodiment Five of the present invention (for example, the algorithm relative position determination module 501 and the vehicle information sending module 502). The processor 60 executes various functional applications and data processing of the electronic device by running the software programs, instructions, and modules stored in the memory 61, that is, implements the above-mentioned test method for the accuracy of the target classification algorithm.

[0093] The memory 61 may mainly include a program storage area and a data storage area. Among them, the program storage area can store an operating system and application programs required for at least one function; the data storage area can store data created according to the use of the terminal, etc. In addition, the memory 61 may include a high-speed random access memory, and may 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 devices. In some instances, the memory 61 may further include a memory remotely set relative to the processor 60, and these remote memories can be connected to the device / terminal / server through a network. Examples of the above network include but are not limited to the Internet, an enterprise internal network, a local area network, a mobile communication network, and their combinations.

[0094] The input device 62 can be used to receive input digital or character information and generate key signal inputs related to user settings and function controls of the electronic device. The output device 63 can include a display device such as a display screen.

[0095] Embodiment Seven

[0096] Embodiment Seven of the present invention further provides a storage medium containing computer-executable instructions, and the computer-executable instructions are used to execute a test method for the accuracy of a target classification algorithm when executed by a computer processor. This method is executed by a server, and the method includes:

[0097] Obtain the algorithm relative position, the first vehicle motion behavior data, and the second vehicle motion behavior data between the first vehicle and the second vehicle; the algorithm relative position is determined based on the target classification algorithm according to the first vehicle motion behavior data and the second vehicle motion behavior data;

[0098] Based on the high-precision map, determine the standard relative position between the first vehicle and the second vehicle according to the first vehicle motion behavior data and the second vehicle motion behavior data;

[0099] According to the standard relative position and the algorithm relative position, determine the accuracy of the target classification algorithm.

[0100] Alternatively, execute a test method for the accuracy of a target classification algorithm. This method is executed by the first vehicle, and the method includes:

[0101] Based on the target classification algorithm, determine the algorithm relative position between the first vehicle and the second vehicle according to the first vehicle motion behavior data and the second vehicle motion behavior data;

[0102] Send the first vehicle motion behavior data, the second vehicle motion behavior data, and the algorithm relative position to the server, and the server performs the following: Based on the high-precision map, determine the standard relative position between the first vehicle and the second vehicle according to the first vehicle motion behavior data and the second vehicle motion behavior data; According to the standard relative position and the algorithm relative position, determine the accuracy of the target classification algorithm.

[0103] Of course, for a storage medium containing computer-executable instructions provided by the embodiments of the present invention, the computer-executable instructions are not limited to the method operations described above, and can also execute relevant operations in the test method for the accuracy of the target classification algorithm provided by any embodiment of the present invention.

[0104] From the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented by means of software and necessary general-purpose hardware. Of course, it can also be implemented by hardware, but in many cases the former is a better implementation. Based on such an understanding, the technical solution of the present invention, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as a floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disc of a computer, etc., including several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments of the present invention.

[0105] It should be noted that in the embodiments of the test device for the accuracy of the above-mentioned target classification algorithm, the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be realized; in addition, the specific names of the functional units are only for the convenience of mutual distinction and do not limit the protection scope of the present invention.

[0106] Note that the above is only the preferred embodiment of the present invention and the applied technical principle. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described here, and various obvious changes, re-adjustments and substitutions can be made by those skilled in the art without departing from the protection scope of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments. Without departing from the concept of the present invention, more other equivalent embodiments can be included, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A test method for the accuracy of an object classification algorithm, characterized in that Executed by a server, the method includes: Obtain the algorithmic relative position between a first vehicle and a second vehicle, the first vehicle motion behavior data, and the second vehicle motion behavior data; the algorithmic relative position is determined based on a target classification algorithm according to the first vehicle motion behavior data and the second vehicle motion behavior data; the first vehicle motion behavior data and the second vehicle motion behavior data are obtained through the LOG log information printed in real time by the T-BOX; Based on a high-precision map, determine the standard relative position between the first vehicle and the second vehicle according to the first vehicle motion behavior data and the second vehicle motion behavior data; Determine the accuracy of the target classification algorithm according to the standard relative position and the algorithmic relative position.

2. The method according to claim 1, characterized in that, Based on a high-precision map, determining the standard relative position between the first vehicle and the second vehicle according to the first vehicle motion behavior data and the second vehicle motion behavior data includes: Based on a high-precision map, determine the first marked position information of the first vehicle in the high-precision map according to the first vehicle position information and the first vehicle heading angle in the first vehicle motion behavior data; Determine the second marked position information of the second vehicle in the high-precision map according to the second vehicle position information and the second vehicle heading angle in the second vehicle motion behavior data; Determine the standard relative position between the first vehicle and the second vehicle according to the first marked position information and the second marked position information.

3. The method according to claim 1, characterized in that, Determining the accuracy of the target classification algorithm according to the standard relative position and the algorithmic relative position includes: Compare whether the standard relative position and the algorithmic relative position are consistent; if they are consistent, determine that the target classification algorithm in the test record to which the standard relative position and the algorithmic relative position belong is accurate; otherwise, determine that the target classification algorithm in the test record to which the standard relative position and the algorithmic relative position belong is incorrect; Determine the proportion of accurate test records in at least two test records, and obtain the accuracy of the target classification algorithm according to the proportion of accurate test records.

4. The method according to claim 1, characterized in that, Obtaining the algorithmic relative position between the first vehicle and the second vehicle, the first vehicle motion behavior data, and the second vehicle motion behavior data includes: Obtain the algorithmic relative position between the first vehicle and the second vehicle, and the first vehicle motion behavior data from the first vehicle; the algorithmic relative position is determined by the first vehicle or the second vehicle based on a target classification algorithm according to the first vehicle motion behavior data and the second vehicle motion behavior data; Obtain the second vehicle motion behavior data from the second vehicle.

5. A method for testing the accuracy of an object classification algorithm, characterized in that, Executed by the first vehicle, the method includes: Based on a target classification algorithm, determine the algorithmic relative position between the first vehicle and the second vehicle according to the first vehicle motion behavior data and the second vehicle motion behavior data; the first vehicle motion behavior data and the second vehicle motion behavior data are obtained through the LOG log information printed in real time by the T-BOX; Send the first vehicle motion behavior data, the second vehicle motion behavior data, and the algorithm relative position to the server, and the server performs the following: Based on the high-precision map, determine the standard relative position between the first vehicle and the second vehicle according to the first vehicle motion behavior data and the second vehicle motion behavior data; According to the standard relative position and the algorithm relative position, determine the accuracy of the target classification algorithm.

6. A test device for the accuracy of an object classification algorithm, characterized in that, Executed by the server, the device includes: A vehicle information acquisition module, configured to acquire the algorithm relative position, the first vehicle motion behavior data, and the second vehicle motion behavior data between the first vehicle and the second vehicle; the algorithm relative position is determined based on the target classification algorithm according to the first vehicle motion behavior data and the second vehicle motion behavior data; the first vehicle motion behavior data and the second vehicle motion behavior data are acquired through the LOG log information printed by the T-BOX in real time; A standard relative position determination module, configured to determine the standard relative position between the first vehicle and the second vehicle based on the high-precision map according to the first vehicle motion behavior data and the second vehicle motion behavior data; An accuracy determination module, configured to determine the accuracy of the target classification algorithm according to the standard relative position and the algorithm relative position.

7. The device according to claim 6, characterized in that, The standard relative position determination module includes: A first marked position information determination unit, configured to determine the first marked position information of the first vehicle in the high-precision map based on the high-precision map according to the first vehicle position information and the first vehicle heading angle in the first vehicle motion behavior data; A second marked position information determination unit, configured to determine the second marked position information of the second vehicle in the high-precision map according to the second vehicle position information and the second vehicle heading angle in the second vehicle motion behavior data; A standard relative position determination unit, configured to determine the standard relative position between the first vehicle and the second vehicle according to the first marked position information and the second marked position information.

8. A test device for the accuracy of an object classification algorithm, characterized in that, Executed by the first vehicle, the device includes: An algorithm relative position determination module, configured to determine the algorithm relative position between the first vehicle and the second vehicle based on the target classification algorithm according to the first vehicle motion behavior data and the second vehicle motion behavior data; the first vehicle motion behavior data and the second vehicle motion behavior data are acquired through the LOG log information printed by the T-BOX in real time; A vehicle information sending module, configured to send the first vehicle motion behavior data, the second vehicle motion behavior data, and the algorithm relative position to the server, and the server performs the following: Based on the high-precision map, determine the standard relative position between the first vehicle and the second vehicle according to the first vehicle motion behavior data and the second vehicle motion behavior data; According to the standard relative position and the algorithm relative position, determine the accuracy of the target classification algorithm.

9. An electronic device, characterized in that, Includes: One or more processors; A storage device, configured to store one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the test method for the accuracy of the target classification algorithm as described in any one of claims 1-5.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by a processor, it implements the test method for the accuracy of the target classification algorithm as described in any one of claims 1-5.

Images