A scenario-based abnormal navigation data extraction method and test system

Abstract

The present invention discloses a scene-based abnormal navigation data extraction method and test system, which classifies video navigation data into scenes and extracts abnormal data for each key scene, thereby completing the extraction of abnormal navigation data in different key scenes, ensuring the pertinence of test data, improving test efficiency, and verifying the performance of the vehicle navigation system in different key scenes.

Description

Technical Field

[0001] The present application relates to vehicle testing, and in particular to a scenario-based abnormal navigation data extraction method and testing system. Background Art

[0002] With the widespread use of vehicle navigation systems in vehicles, the positioning accuracy and system stability of vehicle navigation systems have an important impact on vehicle quality. In order to ensure the quality of license plates, a complete test of the vehicle navigation system is essential.

[0003] At present, for the test of the vehicle navigation system, the navigation data is first collected by using the data acquisition equipment, and then the navigation data is transmitted to the navigation simulator (such as the navigation data playback device, CAN data playback device), and then the navigation simulator is used to play back the navigation data in the laboratory to the vehicle computer for testing.

[0004] However, since the navigation data collected in the early stage is huge and non-targeted, the test is time-consuming and the test efficiency is low. Summary of the invention

[0005] The present invention provides a scenario-based abnormal navigation data extraction method and a test system, so as to solve or partially solve the technical problem that the test data is not targeted when testing the vehicle-mounted navigation system, thereby resulting in low test efficiency.

[0006] In order to solve the above technical problems, the present invention provides a scenario-based abnormal navigation data extraction method, the method comprising:

[0007] Acquire video navigation data; wherein the video navigation data is actual road data captured during navigation;

[0008] The video navigation data is divided according to key scenes to obtain M types of key scene navigation data; M ≥ 1 and is a positive integer;

[0009] Respective abnormal navigation data is extracted from the M-type key scene navigation data for use in vehicle computer testing.

[0010] Preferably, the dividing the video navigation data according to key scenes to obtain M types of key scene navigation data specifically includes:

[0011] The video navigation data is divided according to the scene transit time corresponding to each of the M types of scenes to obtain the M types of key scene navigation data.

[0012] Preferably, the M types of key scene navigation data include N groups of key scene navigation data of the same scene, each group of key scene navigation data has its own scene transit time, N ≥ 1 and is a positive integer;

[0013] The extracting respective abnormal navigation data from the M types of key scene navigation data specifically includes:

[0014] For the N groups of key scene navigation data in each type of scene, abnormal navigation data in the N groups of key scene navigation data are determined based on the number of satellites and / or signal strength corresponding to the key scene navigation data during the scene transit time.

[0015] Preferably, for the N groups of key scene navigation data under each type of scene, determining abnormal navigation data in the N groups of key scene navigation data based on the number of satellites and / or signal strength corresponding to the key scene navigation data within the scene transit time includes:

[0016] For each type of scenario, if the number of satellites corresponding to a data point in the key scene navigation data is less than a quantity threshold, the data point is determined as the abnormal navigation data;

[0017] If the number of satellites corresponding to the data point is greater than or equal to the number threshold, and the number of satellites corresponding to the data point that are above the signal strength threshold is less than the number threshold, the data point is determined as the abnormal navigation data.

[0018] Preferably, the M types of key scene navigation data include N groups of key scene navigation data of the same scene, each group of key scene navigation data has its own scene transit time, N ≥ 1 and is a positive integer;

[0019] The extracting respective abnormal navigation data from the M types of key scene navigation data specifically includes:

[0020] For each type of scene, the N groups of key scene navigation data are segmented into the scene travel time corresponding to each group of key scene navigation data to obtain P time segments for each group of key scene navigation data, where P ≥ 1 and is a positive integer.

[0021] In each time segment, the total data points and the abnormal data points in the time segment are determined, and the abnormal probability is calculated according to the total data points and the abnormal data points, and the abnormal data points are expanded according to the abnormal probability to obtain an abnormal data segment;

[0022] Determine the abnormal data segments corresponding to each group of key scene navigation data according to the abnormal data segments in the P time segments of each group of key scene navigation data;

[0023] Based on the abnormal data segments corresponding to each group of key scene navigation data, abnormal navigation data in the N groups of key scene navigation data are obtained.

[0024] Preferably, the determining of the total data points in the time segment and the abnormal data points in the time segment specifically includes:

[0025] Determine the total data points according to the formula S=Tf; wherein S represents the total data points, T represents the time length of the time segment, and f represents the data collection frequency;

[0026] Based on the number of satellites and / or signal strength corresponding to each data point in the time segment, the abnormal data point in the time segment is determined.

[0027] Preferably, the determining of abnormal data points in the time segment based on the number of satellites and / or signal strength corresponding to each data point in the time segment specifically includes:

[0028] For each data point in the time segment, if the number of satellites corresponding to the data point is less than a quantity threshold, the data point is determined as the abnormal data point;

[0029] If the number of satellites corresponding to the data point is greater than or equal to the number threshold, and the number of satellites corresponding to the data point that are above the signal strength threshold is less than the number threshold, the data point is determined as the abnormal data point.

[0030] Preferably, performing data expansion on the abnormal data points according to the abnormal probability to obtain abnormal data segments specifically includes:

[0031] If the abnormal probability is less than the probability threshold, the abnormal moment of the abnormal data point is used as the first reference time point for a first forward and backward expansion to obtain a first abnormal time period including the abnormal moment, and the data segment corresponding to the first abnormal time period is determined as the abnormal data segment;

[0032] If the abnormal probability is greater than or equal to the probability threshold, the abnormal moments corresponding to the first and last abnormal data points in the time segment are used as the second reference time points for a second forward and backward expansion to obtain a second abnormal time period including all abnormal data points, and the data segment corresponding to the second abnormal time period is determined as the abnormal data segment.

[0033] Preferably, the expansion time periods of the first front-to-back expansion and the second front-to-back expansion are the same.

[0034] The present invention discloses a testing system, comprising:

[0035] A data acquisition device, used to obtain video navigation data; wherein the video navigation data is actual road data during navigation;

[0036] The data processing device is used to divide the video navigation data according to key scenes to obtain M types of key scene navigation data; M≥1 and is a positive integer; and extract respective abnormal navigation data from the M types of key scene navigation data for vehicle computer testing.

[0037] Through one or more technical solutions of the present invention, the present invention has the following beneficial effects or advantages:

[0038] The present invention discloses a scene-based abnormal navigation data extraction method, which can complete the extraction of abnormal navigation data in different key scenes by classifying video navigation data into scenes and extracting abnormal data for each key scene, thereby ensuring the pertinence of test data, improving test efficiency, and verifying the performance of the vehicle navigation system in different key scenes.

[0039] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0041] Figure 1 A flow chart of a scenario-based abnormal navigation data extraction method according to an embodiment of the present invention is shown;

[0042] Figure 2 A schematic diagram of a testing system according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0043] In order to enable technicians in the technical field to which the present application belongs to understand the present application more clearly, the technical solution of the present application is described in detail below through specific embodiments in conjunction with the accompanying drawings.

[0044] See below Figure 1 , is a flow chart of a scenario-based abnormal navigation data extraction method according to an embodiment of the present invention, the method comprising the following steps:

[0045] Step 101, obtaining video navigation data.

[0046] Specifically, the test data used in testing vehicle navigation information mainly include: actual navigation satellite data and video navigation data.

[0047] The actual navigation satellite data mainly includes: satellite transmission data corresponding to each moment, the number of satellites actually searched at each moment, and the signal strength of each satellite, etc. Satellite transmission data is used for actual position positioning, and the number of satellites and satellite signal strength at each moment are used as the basis for data extraction;

[0048] Video navigation data is the actual road data captured during navigation, and is used for preliminary data extraction in different scenarios. Although video navigation data is actual road data, each actual road data corresponds to actual navigation satellite data. That is, the actual road data at each moment corresponds to satellite transmission data, number of satellites, and satellite signal strength.

[0049] When collecting data, video navigation data is obtained through the camera, and the video navigation data at each moment corresponds to satellite transmission data, number of satellites, and satellite signal strength.

[0050] In addition, since the vehicle's driving route is uncertain, the vehicle navigation system will navigate in various scenarios. Due to the particularity of the scenario, the vehicle navigation system will experience navigation anomalies. For example, in tunnels, underground garages, and under viaducts, the main reason is that the number of satellites actually searched is less than 4 due to the obstruction of satellite signals, resulting in vehicle navigation anomalies (navigation drift); in urban high-rise building scenarios, the navigation signal is reflected by the obstruction of the high-rise buildings, and the actual satellite signal strength is weakened, resulting in vehicle navigation anomalies; in high-voltage substations and similar scenarios, the navigation signal will also cause vehicle navigation anomalies due to the addition of various background noise signals.

[0051] Therefore, in this embodiment, after the test data is collected, the video navigation data is obtained from the test data, the video navigation data is used as the basis, and the actual navigation satellite data corresponding to the video navigation data is used as an auxiliary to divide the video navigation data according to the key scenes, so as to make the test data more targeted. For details on the division according to the key scenes, please refer to the following description.

[0052] Step 102: divide the video navigation data according to key scenes to obtain M types of key scene navigation data.

[0053] Among them, the M-type key scene navigation data actually refers to the scene navigation data under the M-type key navigation scenes. The key navigation scene refers to the scene where abnormal navigation data is prone to appear, such as tunnels, underground garages, and under viaducts. M ≥ 1 and is a positive integer. The value of M depends on the number of key navigation scenes appearing in the video navigation data.

[0054] When dividing according to key scenes, the video navigation data is divided according to the scene transit times corresponding to the M types of key scenes to obtain M types of key scene navigation data.

[0055] Specifically, the moment when the vehicle enters the key scene is taken as the start moment, the moment when the vehicle exits the key scene is taken as the end moment, and the time period between the start moment and the end moment is taken as the scene communication time of the key scene.

[0056] Among them, the M-type key scene navigation data contains N groups of key scene navigation data of the same scene, N ≥ 1 and is a positive integer. Each group of key scene navigation data has its own scene transit time. Taking the tunnel scene as an example, the vehicle passes through three tunnels, and each tunnel has a different length and has its own scene communication time. For example, the scene communication time of the first tunnel is [12:00, 12:05], the scene communication time of the second tunnel is [13:00, 13:10], and the scene communication time of the third tunnel is [14:00, 14:15]. By using the above time period to intercept the video navigation data, the key scene navigation data corresponding to the above time period can be obtained. Then for the tunnel scene, it contains 3 groups of key scene navigation data, and each group of key scene navigation data has its own scene transit time.

[0057] In this embodiment, by dividing the video navigation data into key scene navigation data corresponding to each scene according to the scene, subsequent tests can be made more targeted and the test efficiency is improved.

[0058] Since the test focuses more on the performance of the owner's navigation system when the navigation is abnormal, after the data is divided according to the scene, the abnormal navigation data will be extracted from it for vehicle computer testing. Please see the following description for details.

[0059] Step 103, extracting respective abnormal navigation data from the M-type key scene navigation data for use in vehicle computer testing.

[0060] Abnormal navigation data refers to navigation data generated during the navigation process due to abnormal number of satellites and / or abnormal signal strength, such as navigation data generated due to navigation anomalies in scenarios such as tunnels, underground garages, and under viaducts.

[0061] Since the generation of abnormal navigation data depends on the number of satellites and / or signal strength, in the process of extracting abnormal navigation data, for the N groups of key scene navigation data under each type of scene, the abnormal navigation data in the N groups of key scene navigation data are determined based on the number of satellites and / or signal strength corresponding to the key scene navigation data during the scene transit time.

[0062] During the specific implementation process, the N groups of key scene navigation data under each type of scene are processed as follows: if the number of satellites corresponding to a data point in the key scene navigation data is less than the quantity threshold, the data point is determined as abnormal navigation data; if the number of satellites corresponding to the data point is greater than or equal to the quantity threshold, and the number of satellites corresponding to the data point that are above the signal strength threshold is less than the quantity threshold, the data point is determined as abnormal navigation data.

[0063] Specifically, each set of key scene navigation data contains several data points, and each data point has its own collection time, number of satellites used and signal strength of each satellite. Generally speaking, at least 4 satellites are used when collecting data. Less than 4 satellites means that the positioning conditions are not met and abnormal navigation data will be generated. It can be seen that the number of satellites is one of the factors affecting data collection. Therefore, in this embodiment, it is determined whether the number of satellites corresponding to the data point is less than 4. If it is less than 4, the data point is determined as abnormal navigation data. Furthermore, since each satellite has its own signal strength, signal strength is also one of the factors affecting data collection. Therefore, in this embodiment, a signal strength threshold is set. Even if the number of satellites corresponding to the data point is greater than or equal to 4 satellites, but the number of satellites above the signal strength threshold is less than 4, the data point is also determined as abnormal navigation data.

[0064] Further, after obtaining the abnormal navigation data in each set of key scene navigation data, the abnormal navigation data in the N sets of key scene navigation data can be determined. For example, the abnormal navigation data in each set of key scene navigation data in the N sets of key scene navigation data is used as the abnormal navigation data in the N sets of key scene navigation data.

[0065] Obtaining the abnormal navigation data in the N groups of key scene navigation data means obtaining the abnormal navigation data in this type of scene.

[0066] In this embodiment, the number of satellites and / or signal strength is used to obtain data points of abnormal navigation, and the data points are determined as abnormal navigation data for use in vehicle computer testing.

[0067] In one embodiment of the present invention, when abnormal data is converted into normal data, transition data may exist. Since the transition data cannot be undoubtedly confirmed as normal navigation data, this embodiment performs data expansion based on the abnormal data points to obtain the transition data, and uses the transition data for vehicle computer testing to avoid test errors caused by data omissions.

[0068] In the specific implementation process, first, for each type of scene, the N groups of key scene navigation data under each type of scene, the scene travel time corresponding to each group of key scene navigation data is segmented to obtain P time segments for each group of key scene navigation data, where P ≥ 1 and is a positive integer. Specifically, the specific segmentation methods include but are not limited to average segmentation, random segmentation, etc. In order to avoid too many segments affecting the calculation speed, the time in each time segment is ≥ 10 minutes.

[0069] Secondly, in each time segment, the total data points in the time segment and the abnormal data points in the time segment are determined.

[0070] In the process of determining the total data points, the total data points are determined according to the formula S=Tf; wherein S represents the total data points, T represents the time length of the time segment, and f represents the data collection frequency.

[0071] In the process of determining abnormal data points, the abnormal data points in the time segment are determined based on the number of satellites and / or signal strength corresponding to each data point in the time segment. In a specific implementation process, for each data point in the time segment, if the number of satellites corresponding to the data point is less than the number threshold, the data point is determined as an abnormal data point; if the number of satellites corresponding to the data point is greater than or equal to the number threshold, and the number of satellites corresponding to the data point that are higher than the signal strength threshold is less than the number threshold, the data point is determined as an abnormal data point.

[0072] Specifically, each time segment contains several data points, and each data point has its own collection time, the number of satellites used, and the signal strength of each satellite. Generally speaking, at least 4 satellites are used when collecting data. Less than 4 satellites means that the positioning conditions are not met and abnormal navigation data will be generated. It can be seen that the number of satellites is one of the factors affecting data collection. Therefore, in this embodiment, it is determined whether the number of satellites corresponding to the data point is less than 4. If it is less than 4, the data point is determined as abnormal navigation data. Furthermore, since each satellite has its own signal strength, signal strength is also one of the factors affecting data collection. Therefore, in this embodiment, a signal strength threshold is set. Even if the number of satellites corresponding to the data point is greater than or equal to 4 satellites, but the number of satellites above the signal strength threshold is less than 4, the data point is also determined as abnormal navigation data.

[0073] Thirdly, the abnormal probability is calculated based on the total data points and the abnormal data points, and the abnormal data points are expanded according to the abnormal probability to obtain the abnormal data segment.

[0074] Specifically, in the process of calculating the anomaly probability, the number of anomaly data points is compared with the number of total data points to obtain the anomaly probability.

[0075] In the process of data expansion, it is determined whether the abnormal probability is less than the abnormal probability threshold, such as 40%. If the abnormal probability is less than the probability threshold, it means that the abnormal data point in the time segment is distributed data, so the following processing is performed for all abnormal data points in the time segment: the abnormal moment of the abnormal data point is used as the first reference time point for the first front and back expansion, and the first abnormal time period containing the abnormal moment is obtained, and the data segment corresponding to the first abnormal time period is determined as the abnormal data segment. Specifically, during the first front and back expansion, the time of the front and back expansion can be the same or different. For example, the abnormal moment t of the abnormal data point is used as the midpoint, and the length of the front and back expansion time is s, then the first abnormal time period is [ts, t+s], and the data segment in [ts, t+s] is the abnormal data segment. After the above data expansion is performed for each abnormal data point, each abnormal time point is an abnormal data segment.

[0076] If the abnormal probability is greater than or equal to the probability threshold, it means that the abnormal data points in the time segment are centralized data. The abnormal moments corresponding to the first and last abnormal data points in the time segment are used as the second reference time points for the second front-to-back expansion to obtain the second abnormal time period including all abnormal data points. The data segment corresponding to the second abnormal time period is determined as the abnormal data segment. Specifically, during the second front-to-back expansion, the time of the front-to-back expansion can be the same or different. In addition, the expansion time periods of the first front-to-back expansion and the second front-to-back expansion can also be the same. For example, the length of the front-to-back expansion time is s, and the abnormal moment corresponding to the first abnormal data point is t i , and the abnormal time corresponding to the abnormal data point at the end is t j , then the second abnormal time period is [t i -s,t j +s],[t i -s,t j +s] is the abnormal data segment. All abnormal data points in this embodiment correspond to the same abnormal data segment.

[0077] After obtaining the abnormal data segments of each time segment, the abnormal data segments of each group of key scene navigation data in P time segments can be obtained. Further, according to the abnormal data segments of each group of key scene navigation data in P time segments, the abnormal data segments corresponding to each group of key scene navigation data are determined. Specifically, for each group of key scene navigation data, the sum of the abnormal data segments of each time segment is used as the abnormal data segments of each group of key scene navigation data. For example, the group of key scene navigation data has 3 time segments, and the abnormal data segments in each time segment are 10 segments, 1 segment, and 1 segment respectively. Then the group of key scene navigation data has 12 abnormal data segments. Further, based on the abnormal data segments corresponding to each group of key scene navigation data, the abnormal navigation data in N groups of key scene navigation data is obtained. Specifically, the sum of the abnormal data segments corresponding to each group of key scene navigation data is used as the abnormal navigation data in N groups of key scene navigation data. For example, there are 3 groups of key scene navigation data in this scene, and the abnormal data segments of each group of key scene navigation data are 12, 25, and 30 respectively. There are 67 abnormal data segments in this scenario.

[0078] In this embodiment, by expanding the abnormal data points into abnormal data segments, transitional data can also be included in the test, thereby avoiding test errors caused by data omissions.

[0079] Based on the same inventive concept, see Figure 2 , in the following embodiment, a test system is disclosed, comprising:

[0080] The data acquisition device 201 is used to obtain video navigation data; wherein the video navigation data is the actual road data during navigation;

[0081] The data processing device 202 is used to divide the video navigation data according to key scenes to obtain M types of key scene navigation data; M≥1 and is a positive integer; and extract respective abnormal navigation data from the M types of key scene navigation data for playback of vehicle computer testing.

[0082] As an optional embodiment, the test system further includes: a navigation simulator, for transmitting abnormal navigation data;

[0083] The vehicle computer is used to receive abnormal navigation data for navigation testing.

[0084] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present application.

[0085] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A scenario-based abnormal navigation data extraction method, characterized in that: The method comprises: Acquire video navigation data; wherein the video navigation data is actual road data captured during navigation; The video navigation data is divided according to key scenes to obtain M types of key scene navigation data; M ≥ 1 and is a positive integer; the M types of key scene navigation data include N groups of key scene navigation data of the same scene, each group of key scene navigation data has its own scene transit time, N ≥ 1 and is a positive integer; Extracting respective abnormal navigation data from the M types of key scene navigation data for vehicle computer testing, specifically comprising: for N groups of key scene navigation data under each type of scene, segmenting the scene passage time corresponding to each group of key scene navigation data to obtain P time segments for each group of key scene navigation data, where P≥1 and is a positive integer; in each time segment, determining the total data points in the time segment and the abnormal data points in the time segment, and calculating the abnormal probability based on the total data points and the abnormal data points, performing data expansion on the abnormal data points based on the abnormal probability to obtain abnormal data segments; determining the abnormal data segments corresponding to each group of key scene navigation data based on the abnormal data segments in the P time segments of each group of key scene navigation data; and obtaining the abnormal navigation data in the N groups of key scene navigation data based on the abnormal data segments corresponding to each group of key scene navigation data.

2. The method according to claim 1, characterized in that The dividing the video navigation data according to key scenes to obtain M types of key scene navigation data specifically includes: The video navigation data is divided according to the scene transit time corresponding to each of the M types of scenes to obtain the M types of key scene navigation data.

3. The method according to claim 1 or 2, characterized in that The M types of key scene navigation data include N groups of key scene navigation data of the same scene, each group of key scene navigation data has its own scene transit time, N ≥ 1 and is a positive integer; The extracting respective abnormal navigation data from the M types of key scene navigation data specifically includes: For the N groups of key scene navigation data in each type of scene, abnormal navigation data in the N groups of key scene navigation data are determined based on the number of satellites and / or signal strength corresponding to the key scene navigation data during the scene transit time.

4. The method according to claim 3, characterized in that For the N groups of key scene navigation data under each type of scene, determining abnormal navigation data in the N groups of key scene navigation data based on the number of satellites and / or signal strength corresponding to the key scene navigation data during the scene transit time, includes: For each type of scenario, if the number of satellites corresponding to a data point in the key scene navigation data is less than a quantity threshold, the data point is determined as the abnormal navigation data; If the number of satellites corresponding to the data point is greater than or equal to the number threshold, and the number of satellites corresponding to the data point that are above the signal strength threshold is less than the number threshold, the data point is determined as the abnormal navigation data.

5. The method according to claim 1, characterized in that The determining of the total data points in the time segment and the abnormal data points in the time segment specifically includes: Determine the total data points according to the formula S=Tf; wherein S represents the total data points, T represents the time length of the time segment, and f represents the data collection frequency; Based on the number of satellites and / or signal strength corresponding to each data point in the time segment, the abnormal data point in the time segment is determined.

6. The method according to claim 5, characterized in that The determining of abnormal data points in the time segment based on the number of satellites and / or signal strength corresponding to each data point in the time segment specifically includes: For each data point in the time segment, if the number of satellites corresponding to the data point is less than a quantity threshold, the data point is determined as the abnormal data point; If the number of satellites corresponding to the data point is greater than or equal to the number threshold, and the number of satellites corresponding to the data point that are above the signal strength threshold is less than the number threshold, the data point is determined as the abnormal data point.

7. The method according to claim 1, characterized in that The data expansion of the abnormal data points according to the abnormal probability to obtain the abnormal data segment specifically includes: If the abnormal probability is less than the probability threshold, the abnormal moment of the abnormal data point is used as the first reference time point for a first forward and backward expansion to obtain a first abnormal time period including the abnormal moment, and the data segment corresponding to the first abnormal time period is determined as the abnormal data segment; If the abnormal probability is greater than or equal to the probability threshold, the abnormal moments corresponding to the first and last abnormal data points in the time segment are used as the second reference time points for a second forward and backward expansion to obtain a second abnormal time period including all abnormal data points, and the data segment corresponding to the second abnormal time period is determined as the abnormal data segment.

8. The method according to claim 7, characterized in that The expansion time periods of the first pre-post expansion and the second pre-post expansion are the same.

9. A testing system, characterized in that: include: A data acquisition device, used to obtain video navigation data; wherein the video navigation data is actual road data during navigation; The data processing device is used to divide the video navigation data according to key scenes to obtain M types of key scene navigation data; M ≥ 1 and is a positive integer; the M types of key scene navigation data contain N groups of key scene navigation data of the same scene, each group of key scene navigation data has its own scene transit time, N ≥ 1 and is a positive integer; extract respective abnormal navigation data from the M types of key scene navigation data for vehicle computer testing, specifically including: for the N groups of key scene navigation data under each type of scene, the scene transit time corresponding to each group of key scene navigation data is segmented to obtain respective abnormal navigation data of each group of key scene navigation data P time segments, P ≥ 1 and is a positive integer; in each time segment, the total data points in the time segment and the abnormal data points in the time segment are determined, and the abnormal probability is calculated according to the total data points and the abnormal data points, and the abnormal data points are expanded according to the abnormal probability to obtain abnormal data segments; according to the abnormal data segments of each group of key scene navigation data in the P time segments, the abnormal data segments corresponding to each group of key scene navigation data are determined; based on the abnormal data segments corresponding to each group of key scene navigation data, the abnormal navigation data in the N groups of key scene navigation data are obtained.

Images