Vehicle driving data management method and device, storage medium and electronic equipment

Abstract

The invention discloses a vehicle driving data management method and device, a storage medium and electronic equipment, and the method comprises the steps: obtaining a to-be-associated target data set which comprises at least one piece of first driving event data and first driving event metadata corresponding to each piece of first driving event data, each piece of first driving event data comprises at least one frame of first driving data, each piece of first driving event metadata comprises at least one frame of first metadata information, and the first driving data corresponds to the first metadata information; according to the driving scene category of the to-be-associated target data set, the to-be-associated target data set is associated to a driving scene tree, the driving scene tree comprises at least one level of driving scene category, and each level of driving scene category comprises at least one driving scene category; and counting the quantity condition of the target data sets associated with each driving scene category of each level in the driving scene tree, and outputting and displaying the quantity condition.

Description

Technical Field

[0001] The present application relates to the field of intelligent driving technology, and more specifically, to a vehicle driving data management method, device, storage medium and electronic device. Background Art

[0002] AI (Artificial Intelligence) technology plays a key role in the field of intelligent driving, but the most widely used AI technology is still in the field of perceptual intelligence, while cognitive intelligence (including decision-making and planning) is mainly achieved by rule matching. However, the scenarios of vehicle driving are diverse and complex. This decision-making and planning method not only requires a lot of manpower and material resources to collect and set rules, but also has poor flexibility and often causes frustration during vehicle driving, seriously affecting the driving experience and driving safety. Therefore, how to use AI models to achieve safer, more comfortable and efficient driving effects has become one of the important research directions. The premise of training a high-quality decision-making and planning model is to collect a large amount of driving data, which can even reach hundreds of millions. Therefore, how to effectively manage large amounts of driving data in order to improve the training efficiency and accuracy of the decision-making and planning model is an urgent problem to be solved. Summary of the invention

[0003] The present application provides a vehicle driving data management method, device, storage medium and electronic device, which can realize effective management of driving data so as to improve the training efficiency and accuracy of decision-making planning models.

[0004] The specific technical solutions are as follows:

[0005] In a first aspect, an embodiment of the present application provides a vehicle driving data management method, the method comprising:

[0006] Acquire a target data set to be associated, wherein the target data set includes at least one first driving event data and first driving event metadata corresponding to each first driving event data, each first driving event data includes at least one frame of first driving data, each first driving event metadata includes at least one frame of first metadata information, the first driving data corresponds to the first metadata information, the first metadata information includes additional information for describing the first driving data, and the first driving event data includes first driving data recorded within a period of time;

[0007] Associating the target data set to be associated with a driving scenario tree according to the driving scenario category of the target data set to be associated, wherein the driving scenario tree includes at least one level of driving scenario categories, and each level of driving scenario categories includes at least one driving scenario category;

[0008] Count the quantity of the target data sets associated with each of the driving scenario categories at each level in the driving scenario tree, and output and display the quantity.

[0009] It can be seen from the above scheme that after obtaining the target data set, the embodiment of the present application associates the target data set to be associated to the driving scene tree according to the driving scene category of the target data set to be associated, and counts the number of target data sets associated with each driving scene category at each level in the driving scene tree, and outputs the display quantity, so that the technicians (hereinafter referred to as users) used to train the decision planning model can quickly and intuitively know the collection of the target data set associated with each driving scene category, so that the user can quickly know which driving scene category target data sets need to be collected to improve the model quality. It can be seen that the embodiment of the present application can achieve effective management of the target data set so as to improve the training efficiency and accuracy of the decision planning model.

[0010] In a first possible implementation manner of the first aspect, associating the target data set to be associated with the driving scene tree according to the driving scene category of the target data set to be associated includes:

[0011] For the target data set to be associated, output and display all the driving scene categories of the first level;

[0012] The driving scene category of the selected first level is received, and all the driving scene categories of the next level are output and displayed, until the driving scene category of the selected last level is received, and the driving scene category of the selected last level is associated with the target data set to be associated.

[0013] It can be seen from the above scheme that the embodiment of the present application can allow the user to select the driving scene category corresponding to the target data set to be associated one level at a time through human-computer interaction, thereby improving the accuracy of the target data set associated driving scene tree.

[0014] In a second possible implementation manner of the first aspect, when the first driving event data in the target data set to be associated includes the first driving data recorded within a preset time period before and after being taken over during the automatic driving process of the vehicle based on a decision planning model, associating the target data set to be associated with the driving scene tree according to the driving scene category of the target data set to be associated includes:

[0015] Determining the driving scene category according to the taken-over scene corresponding to the target data set to be associated;

[0016] According to the determined driving scene category, the target data set to be associated is associated with the driving scene tree.

[0017] It can be seen from the above scheme that in the embodiment of the present application, when the first driving event data in the target data set to be associated includes the first driving data recorded within a preset time period before and after being taken over during the automatic driving process of the vehicle based on the decision planning model, the driving scene category can be directly determined according to the taken-over scene corresponding to the target data set to be associated, and the target data set to be associated can be associated with the driving scene tree according to the determined driving scene category without the need for manual selection, thereby saving manpower and improving the efficiency of associating the target data set with the driving scene tree.

[0018] In a third possible implementation manner of the first aspect, obtaining the target data set to be associated includes:

[0019] Acquire an original driving data set from a preset storage space, wherein the original driving data set includes at least one second driving event data, and each second driving event data includes at least one frame of second driving data;

[0020] determining, according to a preset determination rule, second metadata information corresponding to each frame of the second driving data in the original driving data set, and writing the second metadata information corresponding to each frame of the second driving data into the preset storage space, wherein the second metadata information includes additional information for describing the second driving data;

[0021] Receiving a structured query language SQL statement for each driving scene category, and using the SQL statement to query the second metadata information in the preset storage space to obtain the first metadata information, and obtaining the first driving data from the preset storage space, wherein the first metadata information is the second metadata information that satisfies the SQL statement, and the SQL statement includes a statement for querying valid metadata information and / or a statement for querying invalid metadata information, the valid metadata information is the second metadata information that meets the training requirements of the decision planning model, the invalid metadata information is the second metadata information that does not meet the training requirements of the decision planning model, and the first driving data is the second driving data corresponding to the first metadata information;

[0022] A data set consisting of the first metadata information and the first driving data is determined as the target data set.

[0023] Through the above scheme, it can be seen that the embodiment of the present application can first obtain an original driving data set including at least one second driving event data, determine the second metadata information corresponding to each frame of the second driving data in the original driving data set according to the preset determination rule, and write the second metadata information corresponding to each frame of the second driving data into the preset storage space, and then query the required first metadata information in the preset storage space by receiving the SQL (Structured Query Language) statement for each driving scene category, and finally determine the data set consisting of the queried first metadata information and the first driving data corresponding to the first metadata information as the target data set, so that when the SQL statement is a statement for querying valid metadata information, the obtained target data set can be used as a valid data set to participate in the training of the decision planning model, and when the SQL statement is a statement for querying invalid metadata information, the obtained target data set is used as an invalid data set to avoid affecting the quality of the decision planning model due to participating in the decision planning model. It can be seen from this that the embodiment of the present application only needs to manually write the SQL statement for each driving scene category, and the other steps can be automatically realized, and metadata information is added to assist model training on the basis of the original driving data, and invalid data sets can be queried to avoid affecting the quality of model training, so that a large number of high-quality training samples of decision planning models can be quickly obtained.

[0024] In a fourth possible implementation manner of the first aspect, the driving scene category at the first level includes a highway and / or an urban road;

[0025] When the driving scenario category of the first level is a highway, the driving scenario category of the second level includes at least one of navigation lane change, lane change suppression, overtaking lane change, speed control, lane centering, and lane avoidance;

[0026] When the driving scenario category of the first level is urban road, the driving scenario category of the second level includes at least one of navigation lane change, suppressed lane change, detour lane change, overtaking lane change, intersection passage, vehicle speed control, lane avoidance, lane centering, and avoidance passage.

[0027] In a fifth possible implementation of the first aspect, the first metadata information includes the curvature of the current road, diverging flow information, the lateral distance of the vehicle deviating from the current lane, the vehicle's intention to change lanes, relevant information of the vehicle in front of the left lane, relevant information of the vehicle in front of the right lane, relevant information of the vehicle behind the left lane, relevant information of the vehicle behind the right lane, whether there is a vehicle in front of the current lane, relevant information of the vehicle in front of the current lane corresponding to the vehicle in front of the current lane when the vehicle in front of the current lane exists, road speed limit information, at least one of the collision time TTC between the vehicle and the other vehicle during the lane change process, and the TTC between the vehicle and the other vehicle during the lane change process.

[0028] In a second aspect, an embodiment of the present application provides a vehicle driving data management device, the device comprising:

[0029] an acquisition unit, configured to acquire a target data set to be associated, wherein the target data set includes at least one first driving event data and first driving event metadata corresponding to each first driving event data, each first driving event data includes at least one frame of first driving data, each first driving event metadata includes at least one frame of first metadata information, the first driving data corresponds to the first metadata information, the first metadata information includes additional information for describing the first driving data, and the first driving event data includes first driving data recorded within a period of time;

[0030] an associating unit, configured to associate the target data set to be associated with a driving scenario tree according to the driving scenario category of the target data set to be associated, wherein the driving scenario tree includes at least one level of driving scenario categories, and each level of driving scenario categories includes at least one driving scenario category;

[0031] A statistical unit, used to count the number of the target data sets associated with each driving scenario category at each level in the driving scenario tree;

[0032] The display unit is used to output and display the quantity situation.

[0033] In a first possible implementation manner of the second aspect, the associating unit includes:

[0034] An output display module, configured to output and display all the driving scene categories of the first level for the target data set to be associated;

[0035] A first receiving module, configured to receive a selected first-level driving scenario category;

[0036] The output display module is used to output and display all the driving scene categories of the next level;

[0037] The first associating module is used to associate the selected driving scene category of the last level with the target data set to be associated after the first receiving module receives the selected driving scene category of the last level.

[0038] In a second possible implementation manner of the second aspect, the associating unit includes:

[0039] A first determination module is configured to determine the driving scene category according to the taken-over scene corresponding to the target data set to be associated when the first driving event data in the target data set to be associated includes the first driving data recorded within a preset time period before and after the take-over during the autonomous driving process of the vehicle based on the decision planning model;

[0040] The second associating module is used to associate the target data set to be associated with the driving scene tree according to the determined driving scene category.

[0041] In a third possible implementation manner of the second aspect, the acquiring unit includes:

[0042] An acquisition module, configured to acquire an original driving data set from a preset storage space, wherein the original driving data set includes at least one second driving event data, and each second driving event data includes at least one frame of second driving data;

[0043] A second determination module, configured to determine, according to a preset determination rule, second metadata information corresponding to each frame of the second driving data in the original driving data set;

[0044] a writing module, configured to write the second metadata information corresponding to each frame of the second driving data into the preset storage space, wherein the second metadata information includes additional information for describing the second driving data;

[0045] A second receiving module is used to receive a structured query language SQL statement for each driving scenario category;

[0046] A query module, used to query the second metadata information in the preset storage space using the SQL statement to obtain the first metadata information, wherein the first metadata information is the second metadata information that satisfies the SQL statement, and the SQL statement includes a statement for querying valid metadata information and / or a statement for querying invalid metadata information, the valid metadata information is the second metadata information that meets the training requirements of the decision planning model, and the invalid metadata information is the second metadata information that does not meet the training requirements of the decision planning model;

[0047] The acquisition module is further used to acquire the first driving data from the preset storage space, wherein the first driving data is the second driving data corresponding to the first metadata information;

[0048] A generating module is used to determine a data set consisting of the first metadata information and the first driving data as the target data set.

[0049] In a fourth possible implementation manner of the second aspect, the driving scene category at the first level includes a highway and / or an urban road;

[0050] When the driving scenario category of the first level is a highway, the driving scenario category of the second level includes at least one of navigation lane change, lane change suppression, overtaking lane change, speed control, lane centering, and lane avoidance;

[0051] When the driving scenario category of the first level is urban road, the driving scenario category of the second level includes at least one of navigation lane change, suppressed lane change, detour lane change, overtaking lane change, intersection passage, vehicle speed control, lane avoidance, lane centering, and avoidance passage.

[0052] In a fifth possible implementation of the second aspect, the first metadata information includes the curvature of the current road, diverging flow information, the lateral distance of the vehicle deviating from the current lane, the vehicle's intention to change lanes, relevant information of the vehicle in front of the left lane, relevant information of the vehicle in front of the right lane, relevant information of the vehicle behind the left lane, relevant information of the vehicle behind the right lane, whether there is a vehicle in front of the current lane, relevant information of the vehicle in front of the current lane corresponding to the vehicle in front of the current lane when there is the vehicle in front of the current lane, road speed limit information, collision time TTC between the vehicle and the other vehicle during the lane change process, and at least one of the TTC between the vehicle and the other vehicle during the lane change process.

[0053] Through the above scheme, it can be known that after obtaining the target data set, the embodiment of the present application associates the target data set to be associated with the driving scene tree according to the driving scene category of the target data set to be associated, and counts the number of target data sets associated with each driving scene category at each level in the driving scene tree, and outputs and displays the number, so that the user can quickly and intuitively know the collection of the target data set associated with each driving scene category, so that the user can quickly know which driving scene category target data sets need to be collected to improve the quality of the model. It can be seen that the embodiment of the present application can achieve effective management of the target data set so as to improve the training efficiency and accuracy of the decision planning model.

[0054] In a third aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in any possible implementation manner of the first aspect.

[0055] In a fourth aspect, an embodiment of the present application provides an electronic device, the electronic device comprising:

[0056] one or more processors;

[0057] The processor is coupled to a storage device, and the storage device is used to store one or more programs;

[0058] When one or more programs are executed by one or more processors, the electronic device implements the method described in any possible implementation manner of the first aspect.

[0059] In a fifth aspect, an embodiment of the present application provides a computer program product, which includes instructions. When the instructions are executed on a computer or a processor, the computer or the processor executes the method described in any possible implementation of the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art description are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings without creative work.

[0061] Figure 1 A flow chart of a vehicle driving data management method provided in an embodiment of the present application;

[0062] Figure 2 An example diagram of a driving scenario tree provided in an embodiment of the present application;

[0063] Figure 3 A schematic diagram of a method for obtaining a target data set and managing a target data set provided in an embodiment of the present application;

[0064] Figure 4 A block diagram of a vehicle driving data management device provided in an embodiment of the present application;

[0065] Figure 5 A schematic diagram of the structure of an electronic device or computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0066] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0067] It should be noted that, in the absence of conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. The terms "including" and "having" in the embodiments of the present application and the accompanying drawings and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device comprising a series of steps or units is not limited to the listed steps or units, but optionally also includes steps or units that are not listed, or optionally also includes other steps or units inherent to these processes, methods, products or devices.

[0068] Figure 1 The present invention is a flow chart of a vehicle driving data management method, which can be applied to electronic equipment or computer equipment, and can be applied to a terminal or a server. Specifically, the method may include the following steps:

[0069] S110: Acquire the target data set to be associated.

[0070] The target data set includes at least one first driving event data and first driving event metadata corresponding to each first driving event data, each first driving event data includes at least one frame of first driving data, each first driving event metadata includes at least one frame of first metadata information, the first driving data corresponds to the first metadata information, the first metadata information includes additional information for describing the first driving data, and the first driving event data includes the first driving data recorded within a period of time. Here, "one frame" can refer to a video frame in the traditional sense, or a driving event data can be divided into multiple data segments according to a preset time interval, and each data segment is called a frame of driving data. The content of the first metadata information corresponding to each frame of first driving data can include one metadata information or multiple metadata information.

[0071] The first driving data is objective data obtained by processing the raw data collected by the data collection device on the vehicle, and the data collection device includes an image collector, a laser radar, a positioning system, etc. The first driving data includes at least one of the self-vehicle information, the other vehicle information, the static landmark information, and the traffic light information. The self-vehicle information includes the real-time global position of the self-vehicle, the real-time global orientation angle, the real-time vehicle speed, the real-time following gear position, the target vehicle speed of the self-vehicle, etc. The other vehicle information includes the real-time global position of the other vehicles around the self-vehicle, the real-time global orientation angle, the real-time vehicle speed, the real-time following gear position, etc. The static landmark information includes the static landmark information on the road surface within a preset distance range around the self-vehicle, such as lane lines, road boundary lines, static obstacles, etc. The traffic light information includes the type and state of the traffic light. The types of traffic lights include motor vehicle traffic lights, non-motor vehicle traffic lights, lane lights, flashing warning lights, pedestrian crossing lights, direction indicators, road and railway level crossing lights, etc. The state of the traffic light includes the on and off state of the traffic light. The target speed of the self-vehicle is the expected speed set by the user.

[0072] The first metadata information includes the curvature of the current road, diverging flow information, the lateral distance of the vehicle deviating from the current lane, the vehicle's intention to change lanes, relevant information of the vehicle in front of the left lane, relevant information of the vehicle in front of the right lane, relevant information of the vehicle behind the left lane, relevant information of the vehicle behind the right lane, whether there is a vehicle in front of the current lane, relevant information of the vehicle in front of the current lane when there is a vehicle in front of the current lane, road speed limit information, TTC (Time to Collision) between the vehicle and the other vehicle during the vehicle's lane change, and at least one of the TTC between the vehicle and the other vehicle during the other vehicle's lane change.

[0073] The front vehicle in the left lane refers to the vehicle that is closest to the own vehicle and relatively located in front of the own vehicle in the left lane adjacent to the current lane where the own vehicle is located; the front vehicle in the right lane refers to the vehicle that is closest to the own vehicle and relatively located in front of the own vehicle in the right lane adjacent to the current lane where the own vehicle is located; the rear vehicle in the left lane refers to the vehicle that is closest to the own vehicle and relatively located behind the own vehicle in the left lane adjacent to the current lane where the own vehicle is located; the rear vehicle in the right lane refers to the vehicle that is closest to the own vehicle and relatively located behind the own vehicle in the right lane adjacent to the current lane where the own vehicle is located; the front vehicle in the current lane refers to the vehicle that is closest to the own vehicle in the current lane where the own vehicle is located.

[0074] The relevant information of the front vehicle in the left lane includes the position of the front vehicle in the left lane relative to the vehicle, the orientation angle of the front vehicle in the left lane relative to the vehicle, the speed of the front vehicle in the left lane relative to the vehicle, etc. The relevant information of the front vehicle in the right lane, the relevant information of the rear vehicle in the left lane, and the relevant information of the rear vehicle in the right lane are the same as the relevant information of the front vehicle in the left lane, and will not be repeated here.

[0075] S120: Associating the target data set to be associated with the driving scenario tree according to the driving scenario category of the target data set to be associated with.

[0076] The driving scenario tree includes at least one level of driving scenario categories, and each level of driving scenario categories includes at least one driving scenario category.

[0077] The first-level driving scenario categories include expressways and / or urban roads; when the first-level driving scenario categories are expressways, the second-level driving scenario categories include at least one of navigation lane change, suppressed lane change, overtaking lane change, vehicle speed control, lane centering, and lane avoidance; when the first-level driving scenario categories are urban roads, the second-level driving scenario categories include at least one of navigation lane change, suppressed lane change, detour lane change, overtaking lane change, intersection passing, vehicle speed control, lane avoidance, lane centering, and avoidance passing.

[0078] In practical applications, the third and higher level driving scenario categories can also be included, such as Figure 2 In the part of the driving scenario tree shown, the next level of lane avoidance may include CUT IN avoidance, V-type RB (Road Boundary) lane avoidance, lane avoidance for vehicles on the line, lane avoidance for large curve RB lanes, lane avoidance for large vehicles, and lane avoidance for entry and exit ramps RB lanes. In the figure, HNP_FST means that the first-level driving scenario category includes highways. CUT IN avoidance refers to avoidance when another vehicle changes lanes and drives in front of the vehicle. The next level of large vehicle lane avoidance may include large vehicle-LB (Lane Boundary) avoidance and large vehicle-RB avoidance.

[0079] In this embodiment, the implementation methods of associating the target data set to be associated with the driving scene tree include but are not limited to the following two:

[0080] Method 1: Human-computer interaction

[0081] For the target data set to be associated, all driving scene categories of the first level are output and displayed; the selected driving scene category of the first level is received, and all driving scene categories of the next level are output and displayed, until the selected driving scene category of the last level is received, and the selected driving scene category of the last level is associated with the target data set to be associated.

[0082] In this method, for the target data set to be associated, all driving scene categories of the first level are output and displayed for the user to select. After the user clicks to select a driving scene category of the first level, all driving scene categories of the second level can be output and displayed for the user to select. After the user clicks to select a driving scene category of the second level, all driving scene categories of the next level continue to be output until the user selects a driving scene category of the last level, and the selected driving scene category of the last level is associated with the target data set to be associated.

[0083] The embodiment of the present application can allow the user to select the driving scene category corresponding to the target data set to be associated one level at a time through human-computer interaction, thereby improving the accuracy of the target data set associated driving scene tree.

[0084] Method 2: Fully automatic

[0085] When the first driving event data in the target data set to be associated includes the first driving data recorded within a preset time period before and after being taken over during the automatic driving process of the vehicle based on the decision planning model, the driving scene category is determined according to the taken-over scene corresponding to the target data set to be associated; and the target data set to be associated is associated with the driving scene tree according to the determined driving scene category.

[0086] After a portion of the target data set is obtained and the target data set is involved in model training to obtain a version of the decision planning model, the version of the decision planning model can be used for autonomous driving. During the autonomous driving process, if a driving risk is about to occur in a certain scenario and the driver takes over, the original driving data information within a preset time period before and after the takeover can be recorded, and the original driving data information can be processed to obtain the first driving data, and the takeover scenario of the first driving data can be recorded. When it is necessary to subsequently associate the target data set including the first driving data with the driving scenario tree, the takeover scenario corresponding to the first driving data can be determined as a driving scenario category, and the target data set to be associated can be associated with the driving scenario tree according to the determined driving scenario category.

[0087] In the embodiment of the present application, when the first driving event data in the target data set to be associated includes the first driving data recorded within a preset time period before and after being taken over during the automatic driving process of the vehicle based on the decision planning model, the driving scene category can be directly determined according to the taken-over scene corresponding to the target data set to be associated, and the target data set to be associated can be associated with the driving scene tree according to the determined driving scene category without the need for manual selection, thereby saving manpower and improving the efficiency of associating the target data set with the driving scene tree.

[0088] S130: Counting the number of target data sets associated with each driving scenario category at each level in the driving scenario tree, and outputting and displaying the number.

[0089] After associating each target data set to be associated with the driving scenario tree, the number of target data sets associated in the driving scenario tree can be counted regularly or irregularly, and output for display, so that the user can intuitively know the target data set collection status of various driving scenario categories. The statistical methods are not limited to the following: (1) Count the number of target data sets associated with each driving scenario category at each level at each time point, such as the form of a chart, the horizontal axis of the chart represents time, and the vertical axis represents the number of target data sets. (2) Count the cumulative number of target data sets associated with each driving scenario category at each level, and display it in the form of a chart, the horizontal axis of the chart represents each driving scenario category at the same level, and the vertical axis represents the cumulative number of the corresponding driving scenario category, so that it can be viewed from a global perspective which driving scenario category has fewer target data sets and needs to be collected more. (3) When a target data set participates in model training, the target data set can be marked so that the amount of data participating in model training for the target data set associated with each driving scenario category at each level can be counted separately and displayed in a chart. The horizontal axis of the chart represents the driving scenario categories at the same level, and the vertical axis represents the cumulative number of the corresponding driving scenario categories and the amount of data participating in model training.

[0090] After counting the number of target data sets associated with each driving scenario category at each level in the driving scenario tree, a data conflict check can be performed on each target data set associated in the driving scenario tree. If a conflict is detected between two target data sets, one of the target data sets can be deleted. The method for checking data conflicts includes: matching the data identifiers in the two target data sets to be checked. If there are target numbers of data identifier pairs that are the same, it is determined that there is a data conflict between the two target data sets. Among them, the data identifier includes but is not limited to a timestamp, a number that uniquely identifies a piece of data, etc. The target number refers to a number greater than a preset number threshold, and the preset number threshold can be determined based on actual conditions.

[0091] The vehicle driving data management method provided by the embodiment of the present application, after obtaining the target data set, associates the target data set to be associated to the driving scene tree according to the driving scene category of the target data set to be associated, and counts the number of target data sets associated with each driving scene category at each level in the driving scene tree, and outputs and displays the number, so that the technicians (hereinafter referred to as users) used to train the decision planning model can quickly and intuitively know the collection of the target data set associated with each driving scene category, so that the user can quickly know which driving scene category target data sets need to be collected to improve the quality of the model. It can be seen from this that the embodiment of the present application can achieve effective management of the target data set so as to improve the training efficiency and accuracy of the decision planning model.

[0092] In one implementation, the method for obtaining a target data set to be associated may include steps A1-A4:

[0093] (A1) Acquire an original driving data set from a preset storage space, wherein the original driving data set includes at least one second driving event data, and each second driving event data includes at least one frame of second driving data.

[0094] The preset storage space includes a data lake, a database, etc. The second driving event data includes second driving data recorded within a period of time.

[0095] (A2) Determine second metadata information corresponding to each frame of second driving data in the original driving data set according to a preset determination rule, and write the second metadata information corresponding to each frame of second driving data into a preset storage space, wherein the second metadata information includes additional information for describing the second driving data.

[0096] The preset determination rules can be manually customized or automatically generated. The automatic generation method includes: determining the preset determination rules according to the decision planning rules based on autonomous driving. Making decisions and planning on the driving behavior of the vehicle in advance is one of the key technologies in the field of intelligent driving. In related technologies, decision planning can be carried out by matching decision planning rules based on autonomous driving. "Decision planning rules" refer to traditional planning algorithms based on optimization, which are rules pre-set one by one based on driving experience. During autonomous driving, decision planning can be carried out by matching decision planning rules. The pre-set decision planning rules themselves will not change with the automatic learning of different actual driving scenarios. Therefore, the decision planning rules are different from the planning algorithms based on deep learning, that is, they are different from the decision planning models trained based on driving data in the embodiments of the present application.

[0097] The embodiment of the present application can determine a preset determination rule based on the decision planning rule, such as extracting key information from the decision planning rule, and using the calculation method of the key information as a preset determination rule. The key information includes some customized data in addition to the objective data in the second driving data. For example, the second driving data only contains information about the vehicle itself and other vehicles, but no information related to the vehicle itself and other vehicles, such as the concepts of the front vehicle in the left lane and the rear vehicle in the left lane. Therefore, when such information exists in the decision planning rule, the key information can be extracted from the decision planning rule, and the calculation method of the key information can be used as a preset determination rule. The calculation method of the key information can be referred to the prior art and will not be repeated here.

[0098] (A3) receiving an SQL statement for each driving scenario category, and using the SQL statement to query the second metadata information in the preset storage space to obtain the first metadata information, and obtaining the first driving data from the preset storage space, wherein the first metadata information is the second metadata information that satisfies the SQL statement, and the SQL statement includes a statement for querying valid metadata information and / or a statement for querying invalid metadata information, the valid metadata information is the second metadata information that meets the training requirements of the decision planning model, the invalid metadata information is the second metadata information that does not meet the training requirements of the decision planning model, and the first driving data is the second driving data corresponding to the first metadata information.

[0099] Since the first metadata information is the second metadata information that satisfies the SQL statement, and the first driving data is the second driving data corresponding to the first metadata information, the first metadata information and the first driving data are actually data filtered out from the set of second metadata information and the set of second driving data, respectively. Therefore, the second metadata information and the first metadata information have the same meaning, and the second driving data and the first driving data have the same meaning. The specific contents of the second metadata information and the second driving data can be found in the detailed explanation of step S110 and will not be repeated here.

[0100] In order to improve the training efficiency and quality of the decision planning model, the corresponding decision planning models can be trained for different driving scenario categories, rather than training only one comprehensive decision planning model that includes all driving scenario categories. In this case, it is necessary to first filter out the data in each driving scenario category from all the second driving data and second metadata information. Therefore, an SQL statement can be written in the SQL operation interface for each driving scenario category to query, first obtain the first metadata information of each driving scenario category, and then obtain the first driving data corresponding to the first metadata information.

[0101] (A4) A data set consisting of the first metadata information and the first driving data is determined as a target data set.

[0102] After obtaining the first metadata information and the first driving data corresponding to the first metadata information, the driving event data to which the first driving data corresponding to the first metadata information belongs can be called the first driving event data, and the first driving event data and the first metadata information corresponding to each first driving data in the first driving event data can be combined into a target data set.

[0103] The embodiment of the present application can first obtain an original driving data set including at least one second driving event data, determine the second metadata information corresponding to each frame of the second driving data in the original driving data set according to a preset determination rule, and write the second metadata information corresponding to each frame of the second driving data into a preset storage space, and then query the required first metadata information in the preset storage space by receiving an SQL statement for each driving scene category, and finally determine the data set consisting of the queried first metadata information and the first driving data corresponding to the first metadata information as the target data set, so that when the SQL statement is a statement for querying valid metadata information, the obtained target data set can be used as a valid data set to participate in the training of the decision planning model, and when the SQL statement is a statement for querying invalid metadata information, the obtained target data set is used as an invalid data set to avoid affecting the quality of the decision planning model due to participating in the decision planning model. It can be seen from this that the embodiment of the present application only needs to manually write an SQL statement for each driving scene category, and the other steps can be automatically realized, and metadata information is added to assist model training on the basis of the original driving data, and invalid data sets can be queried to avoid affecting the quality of model training, so that a large number of high-quality training samples of decision planning models can be quickly obtained.

[0104] In one embodiment, Figure 3 As shown, the embodiment of the present application can obtain the target data set based on the original driving data set through the data funnel, then associate the target data set with the driving scenario tree, perform data conflict checks on multiple target data sets in the driving scenario tree, delete conflicting redundant target data sets, and finally count the number of target data sets associated with each driving scenario category at each level in the driving scenario tree, and display them in a chart. Among them, the data funnel refers to the method described in steps A1-A4.

[0105] Corresponding to the above method embodiment, another embodiment of the present application provides a vehicle driving data management device, such as Figure 4 As shown, the device comprises:

[0106] an acquiring unit 210, configured to acquire a target data set to be associated, wherein the target data set includes at least one first driving event data and first driving event metadata corresponding to each first driving event data, each first driving event data includes at least one frame of first driving data, each first driving event metadata includes at least one frame of first metadata information, the first driving data corresponds to the first metadata information, the first metadata information includes additional information for describing the first driving data, and the first driving event data includes first driving data recorded within a period of time;

[0107] an associating unit 220, configured to associate the target data set to be associated with a driving scenario tree according to the driving scenario category of the target data set to be associated, wherein the driving scenario tree includes at least one level of driving scenario categories, and each level of driving scenario categories includes at least one driving scenario category;

[0108] A statistics unit 230, configured to count the number of the target data sets associated with each of the driving scenario categories at each level in the driving scenario tree;

[0109] The display unit 240 is used to output and display the quantity situation.

[0110] In a possible implementation, the associating unit 220 includes:

[0111] An output display module, configured to output and display all the driving scene categories of the first level for the target data set to be associated;

[0112] A first receiving module, configured to receive a selected first-level driving scenario category;

[0113] The output display module is used to output and display all the driving scene categories of the next level;

[0114] The first associating module is used to associate the selected driving scene category of the last level with the target data set to be associated after the first receiving module receives the selected driving scene category of the last level.

[0115] In a possible implementation, the associating unit 220 includes:

[0116] A first determination module is configured to determine the driving scene category according to the taken-over scene corresponding to the target data set to be associated when the first driving event data in the target data set to be associated includes the first driving data recorded within a preset time period before and after the take-over during the autonomous driving process of the vehicle based on the decision planning model;

[0117] The second associating module is used to associate the target data set to be associated with the driving scene tree according to the determined driving scene category.

[0118] In a possible implementation, the acquiring unit 210 includes:

[0119] An acquisition module, configured to acquire an original driving data set from a preset storage space, wherein the original driving data set includes at least one second driving event data, and each second driving event data includes at least one frame of second driving data;

[0120] A second determination module, configured to determine, according to a preset determination rule, second metadata information corresponding to each frame of the second driving data in the original driving data set;

[0121] a writing module, configured to write the second metadata information corresponding to each frame of the second driving data into the preset storage space, wherein the second metadata information includes additional information for describing the second driving data;

[0122] A second receiving module is used to receive a structured query language SQL statement for each driving scenario category;

[0123] A query module, used to query the second metadata information in the preset storage space using the SQL statement to obtain the first metadata information, wherein the first metadata information is the second metadata information that satisfies the SQL statement, and the SQL statement includes a statement for querying valid metadata information and / or a statement for querying invalid metadata information, the valid metadata information is the second metadata information that meets the training requirements of the decision planning model, and the invalid metadata information is the second metadata information that does not meet the training requirements of the decision planning model;

[0124] The acquisition module is further used to acquire the first driving data from the preset storage space, wherein the first driving data is the second driving data corresponding to the first metadata information;

[0125] A generating module is used to determine a data set consisting of the first metadata information and the first driving data as the target data set.

[0126] In a possible implementation, the driving scene categories at the first level include expressways and / or urban roads;

[0127] When the driving scenario category of the first level is a highway, the driving scenario category of the second level includes at least one of navigation lane change, lane change suppression, overtaking lane change, speed control, lane centering, and lane avoidance;

[0128] When the driving scenario category of the first level is urban road, the driving scenario category of the second level includes at least one of navigation lane change, suppressed lane change, detour lane change, overtaking lane change, intersection passage, vehicle speed control, lane avoidance, lane centering, and avoidance passage.

[0129] In a possible implementation, the first metadata information includes the curvature of the current road, diverging flow information, the lateral distance of the vehicle deviating from the current lane, the vehicle's intention to change lanes, relevant information of the vehicle in front of the left lane, relevant information of the vehicle in front of the right lane, relevant information of the vehicle behind the left lane, relevant information of the vehicle behind the right lane, whether there is a vehicle in front of the current lane, relevant information of the vehicle in front of the current lane corresponding to the vehicle in front of the current lane when there is the vehicle in front of the current lane, road speed limit information, collision time TTC between the vehicle and the other vehicle during the lane change process, and at least one of the TTC between the vehicle and the other vehicle during the lane change process.

[0130] The vehicle driving data management device provided by the embodiment of the present application, after obtaining the target data set, associates the target data set to be associated to the driving scene tree according to the driving scene category of the target data set to be associated, and counts the number of target data sets associated with each driving scene category at each level in the driving scene tree, and outputs and displays the number, so that the technicians (hereinafter referred to as users) used to train the decision planning model can quickly and intuitively know the collection of the target data set associated with each driving scene category, so that the user can quickly know which driving scene category target data sets need to be collected to improve the quality of the model. It can be seen from this that the embodiment of the present application can achieve effective management of the target data set so as to improve the training efficiency and accuracy of the decision planning model.

[0131] Based on the above method embodiments, another embodiment of the present application provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the method described in any of the above embodiments is implemented.

[0132] Based on the above method embodiment, another embodiment of the present application provides an electronic device or a computer device, such as Figure 5 As shown, including:

[0133] One or more processors 310;

[0134] The processor 310 is coupled to a storage device 320, and the storage device 320 is used to store one or more programs;

[0135] When the one or more programs are executed by the one or more processors 310, the electronic device or computer device implements the method described in any of the above embodiments.

[0136] Based on the above method embodiments, another embodiment of the present application provides a vehicle, which includes the device as described in any of the above embodiments, or includes the electronic device as described above.

[0137] Based on the above embodiments, another embodiment of the present application provides a computer program product, which includes instructions. When the instructions are executed on a computer or a processor, the computer or the processor executes the method described in any of the above embodiments.

[0138] The above device embodiment corresponds to the method embodiment and has the same technical effect as the method embodiment. For specific description, please refer to the method embodiment. The device embodiment is obtained based on the method embodiment. For specific description, please refer to the method embodiment part, which will not be repeated here. It can be understood by those skilled in the art that the accompanying drawings are only schematic diagrams of one embodiment, and the modules or processes in the accompanying drawings are not necessarily required to implement the present application.

[0139] Those skilled in the art can understand that the modules in the device in the embodiment can be distributed in the device in the embodiment according to the description of the embodiment, or can be changed accordingly and located in one or more devices different from the embodiment. The modules in the above embodiment can be combined into one module, or can be further divided into multiple sub-modules.

[0140] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A vehicle driving data management method, characterized in that: The method comprises: Acquire a target data set to be associated, wherein the target data set includes at least one first driving event data and first driving event metadata corresponding to each first driving event data, each first driving event data includes at least one frame of first driving data, each first driving event metadata includes at least one frame of first metadata information, the first driving data corresponds to the first metadata information, the first metadata information includes additional information for describing the first driving data, and the first driving event data includes first driving data recorded within a period of time; Associating the target data set to be associated with a driving scenario tree according to the driving scenario category of the target data set to be associated, wherein the driving scenario tree includes at least one level of driving scenario categories, and each level of driving scenario categories includes at least one driving scenario category; Count the quantity of the target data sets associated with each of the driving scenario categories at each level in the driving scenario tree, and output and display the quantity.

2. The method according to claim 1, characterized in that The step of associating the target data set to be associated with the driving scene tree according to the driving scene category of the target data set to be associated includes: For the target data set to be associated, output and display all the driving scene categories of the first level; The driving scene category of the selected first level is received, and all the driving scene categories of the next level are output and displayed, until the driving scene category of the selected last level is received, and the driving scene category of the selected last level is associated with the target data set to be associated.

3. The method according to claim 1, characterized in that When the first driving event data in the target data set to be associated includes the first driving data recorded within a preset time period before and after being taken over during the automatic driving process of the vehicle based on the decision planning model, the target data set to be associated is associated with the driving scene tree according to the driving scene category of the target data set to be associated, including: Determining the driving scene category according to the taken-over scene corresponding to the target data set to be associated; According to the determined driving scene category, the target data set to be associated is associated with the driving scene tree.

4. The method according to claim 1, characterized in that The step of obtaining the target data set to be associated includes: Acquire an original driving data set from a preset storage space, wherein the original driving data set includes at least one second driving event data, and each second driving event data includes at least one frame of second driving data; determining, according to a preset determination rule, second metadata information corresponding to each frame of the second driving data in the original driving data set, and writing the second metadata information corresponding to each frame of the second driving data into the preset storage space, wherein the second metadata information includes additional information for describing the second driving data; Receiving a structured query language SQL statement for each driving scene category, and using the SQL statement to query the second metadata information in the preset storage space to obtain the first metadata information, and obtaining the first driving data from the preset storage space, wherein the first metadata information is the second metadata information that satisfies the SQL statement, and the SQL statement includes a statement for querying valid metadata information and / or a statement for querying invalid metadata information, the valid metadata information is the second metadata information that meets the training requirements of the decision planning model, the invalid metadata information is the second metadata information that does not meet the training requirements of the decision planning model, and the first driving data is the second driving data corresponding to the first metadata information; A data set consisting of the first metadata information and the first driving data is determined as the target data set.

5. The method according to claim 1, characterized in that The driving scenario categories of the first level include highways and / or urban roads; When the driving scenario category of the first level is a highway, the driving scenario category of the second level includes at least one of navigation lane change, lane change suppression, overtaking lane change, speed control, lane centering, and lane avoidance; When the driving scenario category of the first level is urban road, the driving scenario category of the second level includes at least one of navigation lane change, suppressed lane change, detour lane change, overtaking lane change, intersection passage, vehicle speed control, lane avoidance, lane centering, and avoidance passage.

6. The method according to any one of claims 1 to 5, characterized in that The first metadata information includes the curvature of the current road, diverging flow information, the lateral distance of the vehicle deviating from the current lane, the vehicle's intention to change lanes, relevant information of the vehicle in front of the left lane, relevant information of the vehicle in front of the right lane, relevant information of the vehicle behind the left lane, relevant information of the vehicle behind the right lane, whether there is a vehicle in front of the current lane, relevant information of the vehicle in front of the current lane corresponding to the vehicle in front of the current lane when there is the vehicle in front of the current lane, road speed limit information, collision time TTC between the vehicle and the other vehicle during the vehicle's lane change, and at least one of the TTC between the vehicle and the other vehicle during the other vehicle's lane change.

7. A vehicle driving data management device, characterized in that: The device comprises: an acquisition unit, configured to acquire a target data set to be associated, wherein the target data set includes at least one first driving event data and first driving event metadata corresponding to each first driving event data, each first driving event data includes at least one frame of first driving data, each first driving event metadata includes at least one frame of first metadata information, the first driving data corresponds to the first metadata information, the first metadata information includes additional information for describing the first driving data, and the first driving event data includes first driving data recorded within a period of time; an associating unit, configured to associate the target data set to be associated with a driving scenario tree according to the driving scenario category of the target data set to be associated, wherein the driving scenario tree includes at least one level of driving scenario categories, and each level of driving scenario categories includes at least one driving scenario category; A statistical unit, used to count the number of the target data sets associated with each driving scenario category at each level in the driving scenario tree; The display unit is used to output and display the quantity situation.

8. The device according to claim 7, characterized in that The association unit comprises: An output display module, configured to output and display all the driving scene categories of the first level for the target data set to be associated; A first receiving module, configured to receive a selected first-level driving scenario category; The output display module is used to output and display all the driving scene categories of the next level; The first associating module is used to associate the selected driving scene category of the last level with the target data set to be associated after the first receiving module receives the selected driving scene category of the last level.

9. The device according to claim 7, characterized in that The association unit comprises: A first determination module is configured to determine the driving scene category according to the taken-over scene corresponding to the target data set to be associated when the first driving event data in the target data set to be associated includes the first driving data recorded within a preset time period before and after the take-over during the autonomous driving process of the vehicle based on the decision planning model; The second associating module is used to associate the target data set to be associated with the driving scene tree according to the determined driving scene category.

10. The device according to claim 7, characterized in that The acquisition unit comprises: An acquisition module, configured to acquire an original driving data set from a preset storage space, wherein the original driving data set includes at least one second driving event data, and each second driving event data includes at least one frame of second driving data; A second determination module, configured to determine, according to a preset determination rule, second metadata information corresponding to each frame of the second driving data in the original driving data set; a writing module, configured to write the second metadata information corresponding to each frame of the second driving data into the preset storage space, wherein the second metadata information includes additional information for describing the second driving data; A second receiving module is used to receive a structured query language SQL statement for each driving scenario category; A query module, used to query the second metadata information in the preset storage space using the SQL statement to obtain the first metadata information, wherein the first metadata information is the second metadata information that satisfies the SQL statement, and the SQL statement includes a statement for querying valid metadata information and / or a statement for querying invalid metadata information, the valid metadata information is the second metadata information that meets the training requirements of the decision planning model, and the invalid metadata information is the second metadata information that does not meet the training requirements of the decision planning model; The acquisition module is further used to acquire the first driving data from the preset storage space, wherein the first driving data is the second driving data corresponding to the first metadata information; A generating module is used to determine a data set consisting of the first metadata information and the first driving data as the target data set.

11. The device according to claim 7, characterized in that The driving scenario categories of the first level include highways and / or urban roads; When the driving scenario category of the first level is a highway, the driving scenario category of the second level includes at least one of navigation lane change, lane change suppression, overtaking lane change, speed control, lane centering, and lane avoidance; When the driving scenario category of the first level is urban road, the driving scenario category of the second level includes at least one of navigation lane change, suppressed lane change, detour lane change, overtaking lane change, intersection passage, vehicle speed control, lane avoidance, lane centering, and avoidance passage.

12. The device according to any one of claims 7 to 11, characterized in that The first metadata information includes the curvature of the current road, diverging flow information, the lateral distance of the vehicle deviating from the current lane, the vehicle's intention to change lanes, relevant information of the vehicle in front of the left lane, relevant information of the vehicle in front of the right lane, relevant information of the vehicle behind the left lane, relevant information of the vehicle behind the right lane, whether there is a vehicle in front of the current lane, relevant information of the vehicle in front of the current lane corresponding to the vehicle in front of the current lane when there is the vehicle in front of the current lane, road speed limit information, collision time TTC between the vehicle and the other vehicle during the vehicle's lane change, and at least one of the TTC between the vehicle and the other vehicle during the other vehicle's lane change.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

14. An electronic device, characterized in that: The electronic device comprises: one or more processors; The processor is coupled to a storage device, and the storage device is used to store one or more programs; When the one or more programs are executed by the one or more processors, the electronic device implements the method according to any one of claims 1 to 6.

Images