Radar-driven detection method, apparatus, device, and storage medium

Abstract

The application relates to the field of automatic driving control and discloses a radar-driven detection method, device, equipment and storage medium.The method comprises the following steps: acquiring a first point cloud structure output by a first version radar driver and a second point cloud structure output by a second version radar driver; extracting a first string containing all point cloud attribute information in the first point cloud structure and extracting a second string containing all point cloud attribute information in the second point cloud structure; performing consistency comparison on the first string and the second string according to preset character granularity to obtain a comparison result; and generating a detection result of the first version radar driver and the second version radar driver based on the comparison result.The comparison calculation consumption when each attribute value is compared is reduced, the comparison time consumption is reduced, and finally the detection efficiency of the radar driver is improved.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving control, and more particularly to a detection method, apparatus, device, and storage medium for radar-driven systems. Background Technology

[0002] The point cloud data consistency test for radar drivers is a test item that checks whether the output point cloud data is consistent after the same original radar data packet is processed by different versions of radar drivers. It aims to ensure that the processing of point cloud data is consistent with the previous version after the radar driver version is upgraded.

[0003] The typical testing approach involves processing the raw input radar data packet through the radar driver to output a point cloud structure (pointcloud). The point cloud data within this structure is then parsed into individual points, and the attribute values ​​of each point are compared to determine the consistency of the entire point cloud data. The attribute values ​​are the actual data of the current frame's point cloud, data type bytes, and tens of thousands of characters in length. This attribute value is extracted in 32-bit increments and parsed into individual point cloud attribute values ​​(point(x, y, z...). The attribute values ​​of each point are then compared for equality; if all seven attribute values ​​of all points are equal, the point cloud data is considered consistent. Since the attribute values ​​in the point cloud structure are tens of thousands of bytes long, parsing them into points in 32-bit increments and comparing the seven attribute values ​​of each point results in 700,000 comparison operations per frame (assuming 100,000 points), leading to a long overall testing time and low testing efficiency for the radar driver. Summary of the Invention

[0004] The main objective of this invention is to solve the technical problem of low detection efficiency in existing radar-driven detection methods for vehicles.

[0005] The first aspect of the present invention provides a detection method for radar-driven systems, comprising: acquiring a first point cloud structure output by a first version of a radar driver and a second point cloud structure output by a second version of a radar driver; extracting a first string containing all point cloud attribute information from the first point cloud structure and a second string containing all point cloud attribute information from the second point cloud structure; performing a consistency comparison on the first string and the second string according to a preset character fineness to obtain a comparison result; and generating detection results for the first version of the radar driver and the second version of the radar driver based on the comparison result.

[0006] Optionally, in a first implementation of the first aspect of the present invention, the step of comparing the consistency of the first string and the second string according to a preset character fineness to obtain a comparison result includes: if the preset character fineness is a first fineness, then comparing the consistency between the first string and the second string to obtain a first comparison result; if the preset character fineness is a second fineness, then according to the second fineness, dividing the first string and the second string into multiple strings respectively to obtain a first string group and a second string group; comparing the consistency between the corresponding strings obtained from the division of the first string group and the second string group to obtain a second comparison result, wherein the comparison result includes the first comparison result and the second comparison result.

[0007] Optionally, in a second implementation of the first aspect of the present invention, the step of comparing the consistency between the first string and the second string to obtain a first comparison result includes: determining whether the number of attribute values ​​in the first string and the second string exceeds a preset first quantity threshold; if the number of attribute values ​​in the first string and the second string exceeds the preset first quantity threshold, then using a preset indirect comparison algorithm to compare the first string and the second string, and using this as the first comparison result; if the number of attribute values ​​in the first string and the second string does not exceed the preset first quantity threshold, then using a preset direct comparison algorithm to compare the first string and the second string, and using this as the first comparison result.

[0008] Optionally, in a third implementation of the first aspect of the present invention, the step of dividing the first string and the second string into multiple strings according to the second fine-grainedness to obtain a first string group and a second string group includes: if the second fine-grainedness is a fixed number of attribute values, then the first string and the second string are divided into multiple strings of the fixed number of attribute values ​​to obtain a first string group and a second string group; if the second fine-grainedness is a fixed number of groups, then the first string and the second string are divided into multiple strings with the same number of attribute values ​​according to the fixed number of groups to obtain a first string group and a second string group.

[0009] Optionally, in the fourth implementation of the first aspect of the present invention, before comparing the consistency between the first string and the second string to obtain the first comparison result if the preset character granularity is the first granularity, the method further includes: determining whether the number of attribute values ​​in the first string and the number of attribute values ​​in the second string are the same; if they are not the same, generating a comparison result indicating that the first string and the second string are inconsistent, and then jumping to the step of generating a detection result for the first version of the radar driver or the second version of the radar driver based on the comparison result.

[0010] Optionally, in a fifth implementation of the first aspect of the present invention, generating a detection result for the first version of the radar driver and the second version of the radar driver based on the comparison result includes: when the comparison result shows that the first string and the second string are consistent, it is determined that the point cloud computing logic corresponding to the first version of the radar driver and the second version of the radar driver is consistent, and a detection result showing that the point cloud computing logic is consistent is generated; when the comparison result shows that the first string and the second string are inconsistent, it is determined that the point cloud computing logic corresponding to the first version of the radar driver and the second version of the radar driver is inconsistent; according to the comparison result, the attribute values ​​that are inconsistent in the first string and the second string are located to obtain positioning information, and a detection result showing that the point cloud computing logic is inconsistent and the positioning information is generated is generated.

[0011] A second aspect of the present invention provides a detection device for radar-driven systems, comprising: an acquisition module for acquiring a first point cloud structure output by a first version of a radar driver and a second point cloud structure output by a second version of a radar driver; an extraction module for extracting a first string containing all point cloud attribute information from the first point cloud structure and a second string containing all point cloud attribute information from the second point cloud structure; a comparison module for performing a consistency comparison on the first string and the second string according to a preset character fineness to obtain a comparison result; and a generation module for generating detection results for the first version of the radar driver and the second version of the radar driver based on the comparison result.

[0012] Optionally, in a first implementation of the second aspect of the present invention, the comparison module includes: a first comparison unit, configured to compare the consistency between the first string and the second string to obtain a first comparison result if the preset character granularity is a first granularity; a segmentation unit, configured to segment the first string and the second string into multiple strings according to the second granularity if the preset character granularity is a second granularity, thereby obtaining a first string group and a second string group; and a second comparison unit, configured to compare the consistency between the corresponding strings obtained from the segmentation of the first string group and the second string group to obtain a second comparison result, wherein the comparison result includes the first comparison result and the second comparison result.

[0013] Optionally, in a second implementation of the second aspect of the present invention, the first comparison unit is further configured to: determine whether the number of attribute values ​​in the first string and the second string exceeds a preset first quantity threshold; if the number of attribute values ​​in the first string and the second string exceeds the preset first quantity threshold, then use a preset indirect comparison algorithm to compare the first string and the second string, and use the result as a first comparison result; if the number of attribute values ​​in the first string and the second string does not exceed the preset first quantity threshold, then use a preset direct comparison algorithm to compare the first string and the second string, and use the result as a first comparison result.

[0014] Optionally, in a third implementation of the second aspect of the present invention, the segmentation unit is further configured to: if the second fine-grainedness is a fixed number of attribute values, then the first string and the second string are respectively segmented into multiple strings of the fixed number of attribute values, corresponding to a first string group and a second string group; if the second fine-grainedness is a fixed number of groups, then the first string and the second string are respectively segmented into multiple strings of the same number of attribute values ​​according to the fixed number of groups, corresponding to a first string group and a second string group.

[0015] Optionally, in a fourth implementation of the second aspect of the present invention, the comparison module further includes a jump unit, configured to: determine whether the number of attribute values ​​in the first string and the number of attribute values ​​in the second string are the same; if they are not the same, generate a comparison result indicating that the first string and the second string are inconsistent, and jump to execute the step of generating a detection result for the first version of the radar driver or the second version of the radar driver based on the comparison result.

[0016] Optionally, in a fifth implementation of the second aspect of the present invention, the generation module includes: a consistency determination unit, configured to determine that the point cloud computing logic corresponding to the first version radar driver and the second version radar driver is consistent when the comparison result is that the first string and the second string are consistent, and generate a detection result that the point cloud computing logic is consistent; an inconsistency determination unit, configured to determine that the point cloud computing logic corresponding to the first version radar driver and the second version radar driver is inconsistent when the comparison result is that the first string and the second string are inconsistent; and a positioning unit, configured to locate the inconsistent attribute values ​​in the first string and the second string according to the comparison result, obtain positioning information, and generate a detection result that the point cloud computing logic is inconsistent and the positioning information.

[0017] A third aspect of the present invention provides a radar-driven detection device, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor invokes the instructions in the memory to cause the radar-driven detection device to perform the aforementioned radar-driven detection method.

[0018] A fourth aspect of the present invention provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the aforementioned detection method for radar driving.

[0019] In the technical solution provided by this invention, the first point cloud structure and the second point cloud structure output by the radar driver in processing the original radar data packet are obtained from the two versions of radar driver. Then, the first string and the second string containing the point cloud attribute values ​​are extracted from them. Here, there is no need to further extract the attribute values ​​of each point cloud for comparison. The two complete strings can be compared directly, which reduces the comparison calculation consumption when comparing each attribute value, reduces the comparison time, and ultimately improves the detection efficiency of the radar driver. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the first embodiment of the detection method for radar driving according to the present invention;

[0021] Figure 2 This is a schematic diagram of a second embodiment of the detection method for radar driving according to the present invention;

[0022] Figure 3 This is a schematic diagram of one embodiment of a radar-driven detection device according to an embodiment of the present invention;

[0023] Figure 4This is a schematic diagram of another embodiment of the detection device for radar driving according to the present invention;

[0024] Figure 5 This is a schematic diagram of one embodiment of a radar-driven detection device according to an embodiment of the present invention. Detailed Implementation

[0025] This invention provides a detection method, apparatus, device, and storage medium for radar-driven systems. The method involves acquiring a first point cloud structure output by a first version of the radar driver and a second point cloud structure output by a second version of the radar driver; extracting a first string containing all point cloud attribute information from the first point cloud structure and a second string containing all point cloud attribute information from the second point cloud structure; performing a consistency comparison on the first and second strings according to a preset character fineness to obtain a comparison result; and generating detection results for the first and second versions of the radar driver based on the comparison result. This reduces the computational cost of comparing each attribute value, thus reducing comparison time and ultimately improving the detection efficiency of the radar driver.

[0026] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” or “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0027] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 The first embodiment of the detection method for radar-driven systems in this invention includes:

[0028] 101. Obtain the first point cloud structure output by the first version of the radar driver and the second point cloud structure output by the second version of the radar driver;

[0029] It is understood that the executing entity of this invention can be a detection device for radar driving, or it can be a terminal or a server; no specific limitation is made here. This embodiment of the invention will be described using a server as an example.

[0030] In this embodiment, the terms "first version radar driver" and "second version radar driver" are relative, referring to a newer version and an older version. For example, if the first version radar driver is the newer version, then the second version radar driver is the older version, and vice versa. A preferred application scenario is that the newer version radar driver is the latest updated version, while the older version radar driver is a version preceding this latest version; for example, the first version radar driver is an update of the second version radar driver.

[0031] In this embodiment, the comparison of the two radar driver versions mainly focuses on the comparison of their point cloud calculation logic. After the radar sensor collects scene point cloud information, it obtains the original radar data packet, which is then input into the radar driver for point cloud computing processing, outputting a point cloud structure. This involves inputting the same original radar data packet into both the first and second version radar drivers, resulting in the output of a first and a second point cloud structure, respectively. The consistency of the point cloud calculation logic between the first and second version radar drivers is then compared by performing a consistency comparison on the output first and second point cloud structures.

[0032] Specifically, the point cloud structure after radar-driven processing contains multiple point cloud attributes. Each point cloud stores seven attribute values ​​in 32-bit increments. These seven attribute values ​​for each point cloud include ['x', 'y', 'z', 'intensity', 'timestamp', 'ring', 'azimuth'], representing three three-dimensional coordinate attributes: intensity, timestamp, line bundle, and azimuth. Each point cloud refers to a single point or multiple points in the scene acquired by the aforementioned radar sensor.

[0033] 102. Extract the first string containing all point cloud attribute information from the first point cloud structure, and extract the second string containing all point cloud attribute information from the second point cloud structure;

[0034] In this embodiment, the first point cloud structure and the second point cloud structure are the result outputs of the computational logic of the first version of the radar driver and the second version of the radar driver, respectively. The difference between the two computational logics is reflected in the strings storing all point cloud attribute information in the point cloud structures, that is, the first string in the first point cloud structure and the second string in the second point cloud structure. Therefore, comparing the consistency of the first point cloud structure and the second point cloud structure essentially involves comparing the consistency of the first string and the second string.

[0035] In this embodiment, the point cloud attribute information in the point cloud structure is stored in a string, with 7 attribute values ​​of a point cloud stored in 32-bit increments. Therefore, the first string after the first version of the radar driver executes the point cloud computing logic can be extracted from the first point cloud structure, and the second string after the second version of the radar executes the point cloud computing logic can be extracted from the second point cloud structure.

[0036] Specifically, the string contains point cloud attribute information for multiple point clouds (such as the seven mentioned above). For example, if the string contains point cloud attribute information for 10,000 point clouds, then the string contains 70,000 point cloud attribute values ​​with 320,000 digits. Therefore, the first and second strings here contain 320,000 characters.

[0037] 103. Perform a consistency comparison on the first string and the second string according to the preset character fineness to obtain the comparison result;

[0038] In this embodiment, when comparing the first string and the second string, the amount of comparison computing resources consumed mainly depends on the number of comparisons. Instead of comparing each attribute of each point cloud in the two strings separately, the two strings are segmented by a preset larger character granularity before comparison, which reduces the number of comparisons and the consumption of comparison computing resources.

[0039] Specifically, in order to reduce the number of comparisons between two strings, if the number of characters in the preset fine-grained character set is a, and the two strings each contain a number of characters b, since the comparison is performed on the seven attributes of the point cloud as shown above each time, the number of comparisons is b*(7 / 32), so the number of characters in the preset fine-grained character set a>32 / 7.

[0040] Specifically, the finer the character granularity, the fewer the number of splits between the two strings, resulting in fewer split strings and fewer comparisons, thus consuming less computational resources and increasing comparison efficiency. Preferably, a consistency comparison can be performed directly on the first and second strings, requiring only one string comparison, minimizing the computational resources consumed in this case.

[0041] Furthermore, since segmenting and comparing the first and second strings according to the preset fine-grained character settings requires additional segmentation computation resources, we need to consider both the comparison computation resources and the segmentation computation resources. Specifically, the standard for setting the fine-grained character settings is that the total consumption of segmentation and comparison computation resources for the first and second strings is less than the total consumption of comparison computation resources for each attribute of the point clouds of the first and second strings. This ensures that when comparing the first and second strings according to the preset fine-grained character settings, computational resource consumption is lower and comparison efficiency is higher.

[0042] 104. Based on the comparison results, generate detection results for the first version of the radar driver and the second version of the radar driver.

[0043] In this embodiment, if the comparison results between the first string and the second string are the same, it means that the first version of the radar driver and the second version of the radar driver have the same point cloud calculation logic. This means that after updating the older version of the radar driver, the point cloud computing logic of the newer version has not changed; both versions are consistent and will not be affected by the version update. There is no inconsistency in the point cloud computing logic between the two versions of the radar driver. Conversely, if they are different, the point cloud computing logic of the two versions of the radar driver is inconsistent, indicating that there is an anomaly after the update of the newer version of the radar driver. An anomaly detection report can also be generated simultaneously to locate the abnormal point cloud and assist developers in resolving related issues.

[0044] In this embodiment of the invention, by obtaining the first point cloud structure and the second point cloud structure output by the old and new versions of the radar driver when processing the radar raw data packet, and then extracting the first string and the second string containing the point cloud attribute values, it is not necessary to further extract the attribute values ​​of each point cloud for comparison. Instead, the two complete strings can be compared directly, which reduces the comparison calculation consumption when comparing each attribute value, thus reducing the comparison time and ultimately improving the detection efficiency of the radar driver.

[0045] Please see Figure 2 A second embodiment of the detection method for radar driving in this invention includes:

[0046] 201. Obtain the first point cloud structure output by the first version of the radar driver and the second point cloud structure output by the second version of the radar driver;

[0047] 202. Extract the first string containing all point cloud attribute information from the first point cloud structure, and extract the second string containing all point cloud attribute information from the second point cloud structure;

[0048] 203. If the preset character fineness is the first fineness, then compare the consistency between the first string and the second string to obtain the first comparison result;

[0049] 204. If the preset character fineness is the second fineness, then according to the second fineness, the first string and the second string are divided into multiple strings respectively, resulting in the first string group and the second string group;

[0050] 205. Compare the consistency between the corresponding strings obtained by segmenting the first string group and the second string group to obtain a second comparison result, wherein the comparison result includes the first comparison result and the second comparison result;

[0051] In this embodiment, the fine-grained character selection can be set according to different actual situations to compare the first string and the second string and obtain the comparison result. Specifically, a first fine-grained selection and a second fine-grained selection are preset. The first fine-grained selection is for all characters, meaning that only all characters in the first and second strings are compared once to obtain the first comparison result. Under this condition, the comparison computational resource consumption is minimized, and the comparison efficiency is maximized. The second fine-grained selection is for a subset of characters, meaning that the first and second strings are first divided into multiple strings according to the number of a subset of characters, and then each segmented string is compared for consistency according to the segmentation order to obtain the second comparison result.

[0052] Specifically, regarding the second fine-grained setting, the number of times the string is split according to the second fine-grained setting is d, resulting in d+1 split strings. If the total number of attribute values ​​of all point clouds in the first or second string is n, the resource consumption for each split is c, and the resource consumption for each comparison is k, then after setting the second fine-grained setting, it should satisfy: dc+(d+1)k<nk.

[0053] In one implementation, it is determined whether the number of attribute values ​​in the first string and the second string exceeds a preset first quantity threshold. If the number of attribute values ​​in the first string and the second string exceeds the preset first quantity threshold, a preset indirect comparison algorithm is used to compare the first string and the second string, and the result is taken as the first comparison result. If the number of attribute values ​​in the first string and the second string does not exceed the preset first quantity threshold, a preset direct comparison algorithm is used to compare the first string and the second string, and the result is taken as the first comparison result.

[0054] In this embodiment, the comparison of the first and second strings performed at the first fine-grained level can be further determined by the number of attributes in each string, deciding whether to use a direct or indirect comparison method. For strings with a large number of characters, the strings can be hashed, and the hash values ​​of the two hashed strings can be compared to determine if they are identical. Therefore, a first threshold is preset; a string with more than this threshold is considered to have a large number of characters. For strings with fewer than this threshold, the first and second strings are directly compared using methods such as "=" or "equals" (e.g., "string").

[0055] In one implementation, if the second fine-grainedness is a fixed number of attribute values, then the first string and the second string are respectively divided into multiple strings with the fixed number of attribute values, corresponding to a first string group and a second string group; if the second fine-grainedness is a fixed number of groups, then the first string and the second string are respectively divided into multiple strings with the same number of attribute values ​​according to the fixed number of groups, corresponding to a first string group and a second string group.

[0056] Specifically, if the first and second strings each contain 10,000 characters, and the fixed attribute value is 1,000, then after splitting the first and second strings, each string will contain 10 strings of 1,000 characters. In this case, comparing the first and second strings consumes 9 splitting computations and 10 comparison computations. If the fixed number of groups is 10, then after splitting the first and second strings, each string will contain 10 strings of 1,000 characters, and the comparison process will also consume 9 splitting computations and 10 comparison computations.

[0057] In one implementation, it is determined whether the number of attribute values ​​in the first string and the number of attribute values ​​in the second string are the same. If they are not the same, a comparison result indicating that the first string and the second string are inconsistent is generated, and the process jumps to the step of generating a detection result for the first version of the radar driver or the second version of the radar driver based on the comparison result. That is, before performing the comparison between the first string and the second string, it can be pre-determined whether the number of characters in the two strings is the same. If they are not the same, it is directly determined that the first string and the second string are inconsistent, without further comparison, and the point cloud computing logic of the two versions of the radar driver can be output as inconsistent.

[0058] 206. When the comparison result shows that the first string and the second string are consistent, it is determined that the point cloud computing logic corresponding to the first version radar driver and the second version radar driver is consistent, and a detection result showing that the point cloud computing logic is consistent is generated.

[0059] 207. When the comparison result shows that the first string and the second string are inconsistent, it is determined that the point cloud computing logic corresponding to the first version of the radar driver and the second version of the radar driver are inconsistent.

[0060] 208. Based on the comparison results, locate the inconsistent attribute values ​​in the first string and the second string, obtain the location information, and generate the detection result of the inconsistency of the point cloud computing logic and the location information.

[0061] The detection method for radar driving in the embodiments of the present invention has been described above. The detection device for radar driving in the embodiments of the present invention will be described below. Please refer to [link / reference]. Figure 3 One embodiment of the radar-driven detection device of the present invention includes:

[0062] The acquisition module 301 is used to acquire the first point cloud structure output by the first version of the radar driver and the second point cloud structure output by the second version of the radar driver.

[0063] Extraction module 302 is used to extract a first string containing all point cloud attribute information in the first point cloud structure, and to extract a second string containing all point cloud attribute information in the second point cloud structure;

[0064] The comparison module 303 is used to perform a consistency comparison on the first string and the second string according to a preset character fineness, and obtain a comparison result;

[0065] The generation module 304 is used to generate detection results for the first version of the radar driver and the second version of the radar driver based on the comparison results.

[0066] In this embodiment of the invention, by obtaining the first point cloud structure and the second point cloud structure output by the old and new versions of the radar driver when processing the radar raw data packet, and then extracting the first string and the second string containing the point cloud attribute values, it is not necessary to further extract the attribute values ​​of each point cloud for comparison. Instead, the two complete strings can be compared directly, which reduces the comparison calculation consumption when comparing each attribute value, thus reducing the comparison time and ultimately improving the detection efficiency of the radar driver.

[0067] Please see Figure 4 Another embodiment of the radar-driven detection device in this invention includes:

[0068] The acquisition module 301 is used to acquire the first point cloud structure output by the first version of the radar driver and the second point cloud structure output by the second version of the radar driver.

[0069] Extraction module 302 is used to extract a first string containing all point cloud attribute information in the first point cloud structure, and to extract a second string containing all point cloud attribute information in the second point cloud structure;

[0070] The comparison module 303 is used to perform a consistency comparison on the first string and the second string according to a preset character fineness, and obtain a comparison result;

[0071] The generation module 304 is used to generate detection results for the first version of the radar driver and the second version of the radar driver based on the comparison results.

[0072] Specifically, the comparison module 303 includes:

[0073] The first comparison unit 3031 is used to compare the consistency between the first string and the second string if the preset character fineness is the first fineness, and obtain the first comparison result;

[0074] The segmentation unit 3032 is used to segment the first string and the second string into multiple strings according to the second fineness if the preset character fineness is the second fineness, thereby obtaining the first string group and the second string group respectively;

[0075] The second comparison unit 3033 is used to compare the consistency between the corresponding strings obtained by segmenting the first string group and the second string group to obtain a second comparison result, wherein the comparison result includes a first comparison result and a second comparison result.

[0076] Specifically, the first comparison unit 3031 is also used for:

[0077] Determine whether the number of attribute values ​​in the first string and the second string exceeds a preset first threshold.

[0078] If the number of attribute values ​​in the first string and the second string exceeds a preset first threshold, a preset indirect comparison algorithm is used to compare the first string and the second string, and the result is taken as the first comparison result.

[0079] If the number of attribute values ​​in the first string and the second string does not exceed a preset first quantity threshold, then a preset direct comparison algorithm is used to compare the first string and the second string, and this comparison result is taken as the first comparison result.

[0080] Specifically, the segmentation unit 3032 is also used for:

[0081] If the second fine-grainedness is a fixed number of attribute values, then the first string and the second string are respectively divided into multiple strings of the fixed number of attribute values, corresponding to the first string group and the second string group;

[0082] If the second fine-grainedness is a fixed number of groups, then the first string and the second string are respectively divided into multiple strings with the same number of attribute values ​​according to the fixed number of groups, thus obtaining the first string group and the second string group.

[0083] Specifically, the comparison module 303 further includes a jump unit 3034, used for:

[0084] Determine whether the number of attribute values ​​in the first string is the same as the number of attribute values ​​in the second string;

[0085] If they are not the same, a comparison result indicating that the first string and the second string are inconsistent is generated, and the process jumps to the step of generating a detection result for the first version of the radar driver or the second version of the radar driver based on the comparison result.

[0086] Specifically, the generation module 304 includes:

[0087] The consistency determination unit 3041 is used to determine that the point cloud computing logic corresponding to the first version radar driver and the second version radar driver is consistent when the comparison result is that the first string and the second string are consistent, and to generate a detection result that the point cloud computing logic is consistent.

[0088] The inconsistency determination unit 3042 is used to determine that the point cloud computing logic corresponding to the first version radar driver and the second version radar driver is inconsistent when the comparison result is that the first string and the second string are inconsistent.

[0089] The positioning unit 3043 is used to locate the inconsistent attribute values ​​between the first string and the second string according to the comparison result, obtain positioning information, and generate a detection result of the inconsistency of the point cloud computing logic and the positioning information.

[0090] above Figure 3 and Figure 4 The detection device for radar driving in the embodiments of the present invention will be described in detail from the perspective of modular functional entities. The detection equipment for radar driving in the embodiments of the present invention will be described in detail from the perspective of hardware processing.

[0091] Figure 5This is a schematic diagram of a radar-driven detection device 500 provided in an embodiment of the present invention. The radar-driven detection device 500 can vary significantly due to different configurations or performance characteristics. It may include one or more central processing units (CPUs) 510 (e.g., one or more processors) and a memory 520, and one or more storage media 530 (e.g., one or more mass storage devices) storing application programs 533 or data 532. The memory 520 and storage media 530 can be temporary or persistent storage. The program stored in the storage media 530 may include one or more modules (not shown in the diagram), each module including a series of instruction operations on the radar-driven detection device 500. Furthermore, the processor 510 may be configured to communicate with the storage media 530 and execute the series of instruction operations in the storage media 530 on the radar-driven detection device 500.

[0092] The radar-driven detection device 500 may also include one or more power supplies 540, one or more wired or wireless network interfaces 550, one or more input / output interfaces 560, and / or one or more operating systems 531, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art will understand that... Figure 5 The illustrated structure of a radar-driven detection device does not constitute a limitation on radar-driven detection devices, which may include more or fewer components than illustrated, or combine certain components, or have different component arrangements.

[0093] The present invention also provides a radar-driven detection device, wherein the computer device includes a memory and a processor, the memory storing computer-readable instructions, and when the computer-readable instructions are executed by the processor, the processor performs the steps of the radar-driven detection method described in the above embodiments.

[0094] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when executed on a computer, cause the computer to perform the steps of the detection method for radar driving.

[0095] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0096] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0097] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A detection method for radar-driven systems, the detection method for radar-driven systems comprising: Obtain the first point cloud structure output by the first version of the radar driver and the second point cloud structure output by the second version of the radar driver; Extract a first string containing all point cloud attribute information from the first point cloud structure, and extract a second string containing all point cloud attribute information from the second point cloud structure; According to the preset character fineness, the first string and the second string are compared for consistency to obtain the comparison result; Based on the comparison results, detection results are generated for the first version of the radar driver and the second version of the radar driver.

2. The detection method for radar-driven radar according to claim 1, characterized in that, The step of performing a consistency comparison on the first string and the second string according to a preset character fineness, and obtaining the comparison result includes: If the preset character fineness is the first fineness, then the consistency between the first string and the second string is compared to obtain the first comparison result; If the preset character fineness is the second fineness, then according to the second fineness, the first string and the second string are divided into multiple strings respectively, resulting in the first string group and the second string group; By comparing the consistency between the corresponding strings obtained from the segmentation of the first string group and the second string group, a second comparison result is obtained, wherein the comparison result includes the first comparison result and the second comparison result.

3. The detection method for radar driving according to claim 2, characterized in that, The comparison of the consistency between the first string and the second string to obtain the first comparison result includes: Determine whether the number of attribute values ​​in the first string and the second string exceeds a preset first threshold. If the number of attribute values ​​in the first string and the second string exceeds a preset first threshold, a preset indirect comparison algorithm is used to compare the first string and the second string, and the result is taken as the first comparison result. If the number of attribute values ​​in the first string and the second string does not exceed a preset first quantity threshold, then a preset direct comparison algorithm is used to compare the first string and the second string, and this comparison result is taken as the first comparison result.

4. The detection method for radar driving according to claim 2, characterized in that, The step of dividing the first string and the second string into multiple strings according to the second fine-grained method, corresponding to obtaining a first string group and a second string group, includes: If the second fine-grainedness is a fixed number of attribute values, then the first string and the second string are respectively divided into multiple strings of the fixed number of attribute values, corresponding to the first string group and the second string group; If the second fine-grainedness is a fixed number of groups, then the first string and the second string are respectively divided into multiple strings with the same number of attribute values ​​according to the fixed number of groups, thus obtaining the first string group and the second string group.

5. The detection method for radar driving according to claim 4, characterized in that, Before comparing the consistency between the first string and the second string to obtain the first comparison result if the preset character granularity is the first granularity, the method further includes: Determine whether the number of attribute values ​​in the first string is the same as the number of attribute values ​​in the second string; If they are not the same, a comparison result indicating that the first string and the second string are inconsistent is generated, and the process jumps to the step of generating detection results for the first version of the radar driver and the second version of the radar driver based on the comparison result.

6. The detection method for radar driving according to claim 4, characterized in that, The step of generating detection results for the first version of the radar driver and the second version of the radar driver based on the comparison results includes: When the comparison result shows that the first string and the second string are consistent, it is determined that the point cloud computing logic corresponding to the first version radar driver and the second version radar driver is consistent, and a detection result showing that the point cloud computing logic is consistent is generated. When the comparison result shows that the first string and the second string are inconsistent, it is determined that the point cloud computing logic corresponding to the first version of the radar driver and the second version of the radar driver are inconsistent. Based on the comparison results, the inconsistency of attribute values ​​between the first string and the second string is identified, location information is obtained, and a detection result of the inconsistency of the point cloud computing logic and the location information is generated.

7. A detection device for radar-driven applications, characterized in that, The detection device for radar driving includes: The acquisition module is used to acquire the first point cloud structure output by the first version of the radar driver and the second point cloud structure output by the second version of the radar driver. The extraction module is used to extract a first string containing all point cloud attribute information from the first point cloud structure, and to extract a second string containing all point cloud attribute information from the second point cloud structure. The comparison module is used to perform a consistency comparison on the first string and the second string according to a preset character fineness, and obtain the comparison result; A generation module is used to generate detection results for the first version of the radar driver and the second version of the radar driver based on the comparison results.

8. The detection device for radar driving according to claim 7, characterized in that, The comparison module includes: The first comparison unit is used to compare the consistency between the first string and the second string if the preset character fineness is the first fineness, and obtain the first comparison result; The segmentation unit is used to segment the first string and the second string into multiple strings according to the second fineness if the preset character fineness is the second fineness, thereby obtaining the first string group and the second string group respectively; The second comparison unit is used to compare the consistency between the corresponding strings obtained by segmenting the first string group and the second string group to obtain a second comparison result, wherein the comparison result includes a first comparison result and a second comparison result.

9. A radar-driven detection device, characterized in that, The radar-driven detection device includes: a memory and at least one processor, wherein the memory stores instructions; The at least one processor invokes the instructions in the memory to cause the radar-driven detection device to perform the steps of the radar-driven detection method as described in any one of claims 1-6.

10. A computer-readable storage medium storing instructions thereon, characterized in that, When the instructions are executed by the processor, they implement the steps of the radar-driven detection method as described in any one of claims 1-6.

Images