A method for triggering data collection, a method for data collection and a device

Abstract

The embodiments of the present application provide a method for triggering data collection, a method and device for data collection. Specifically, when executing the technical solution provided in the embodiments of the present application, the server can obtain a first data set, and the first data set includes multiple data, and the multiple data are data generated when the autonomous driving system fails. Then, the processor can determine the target collection scenario of the autonomous driving system based on the first data set. The target collection scenario is a scenario in which the autonomous driving system fails. After determining the target collection scenario, the server can send a target data collection instruction to a vehicle equipped with an autonomous driving system to instruct the vehicle to collect data under the target collection scenario. In this way, the vehicle is controlled to collect data in a targeted manner to reduce the amount of data uploaded by the vehicle.

Description

Technical Field

[0001] The present application relates to the field of autonomous driving, and in particular to a method for triggering data collection, a method and device for data collection. Background Art

[0002] With the development of computer technology, autonomous driving technology has been widely used. Autonomous driving technology refers to the technology that simulates the driver's behavior through the autonomous driving system on the vehicle to control the normal driving of the vehicle. Through autonomous driving technology, the driver's operations when driving the vehicle can be reduced, which improves the driver's driving experience.

[0003] The more realistic the simulation of the driver by the autonomous driving system, the better the effect of autonomous driving. Therefore, in order to improve the effect of autonomous driving, the operation data of the driver when driving the vehicle can be obtained, and the autonomous driving system can be updated and iterated according to the operation data of the driver when driving the vehicle, so as to improve the realistic degree of simulation of the driver's behavior by the autonomous driving system. Specifically, one or more sensors can be deployed on the vehicle to collect data when the vehicle is driving through the sensors. The vehicle can upload this data to the server through the network so that the server can update the autonomous driving system according to the collected data.

[0004] However, a vehicle generates a lot of data during driving. If all the data is uploaded to the server, the server's computing and processing resources will be wasted. Summary of the invention

[0005] In view of this, the embodiments of the present application provide a method for triggering data collection, a method and device for data collection, which aim to control the vehicle to collect data in a targeted manner and reduce the amount of data uploaded by the vehicle.

[0006] In a first aspect, an embodiment of the present application provides a method for triggering data collection, the method being applied to a server, the method comprising:

[0007] The server acquires a first data set, wherein the first data set includes data generated by the autonomous driving system when the system fails;

[0008] The server determines, according to the first data set, a target acquisition scenario of the autonomous driving system, where the target acquisition scenario is a scenario in which the autonomous driving system fails;

[0009] The server sends a target data collection instruction to the vehicle, where the target data collection instruction is used to instruct the vehicle to collect data in the target collection scenario, and the vehicle is equipped with the automatic driving system.

[0010] Optionally, the server determines, according to the first data set, a target acquisition scenario of the autonomous driving system, including:

[0011] The server determines, according to the first data set, a correspondence between a plurality of failure scenarios and failure causes, the failure cause being a cause causing the failure of the autonomous driving system, and the failure scenario indicating an environment in which the autonomous driving system is located when the autonomous driving system fails;

[0012] The server determines a target collection scenario from the multiple failure scenarios based on the correspondence between the multiple failure scenarios and the failure causes, and the probability that the autonomous driving system fails due to the failure causes corresponding to the target collection scenario is higher than the probability that the autonomous driving system fails due to the failure causes corresponding to the non-target collection scenario.

[0013] Optionally, the method further comprises:

[0014] The server obtains data collected by the vehicle in the target collection scenario;

[0015] The server updates the autonomous driving system based on the data.

[0016] In a second aspect, an embodiment of the present application provides a method for data collection, the method being applied to a vehicle equipped with an automatic driving system, the method comprising:

[0017] The vehicle acquires a target data collection instruction sent by a server, wherein the target data collection instruction is used to instruct the vehicle to collect data under a target collection scenario, and the target collection scenario is a cause of failure of the automatic driving system;

[0018] In response to the vehicle satisfying the target collection scenario, the vehicle collects the data according to the target data collection instruction;

[0019] The vehicle sends the data to the server.

[0020] Optionally, the target data collection instruction further includes a target data collection rule, and the target data collection rule is used to indicate a method in which the vehicle collects the data. The vehicle collects the data according to the target data collection instruction, including:

[0021] The vehicle collects the data according to the target data collection rule.

[0022] In a third aspect, an embodiment of the present application provides a device for triggering data collection, the device being applied to a server, and the device comprising:

[0023] An acquisition unit, configured to acquire a first data set, wherein the first data set includes data generated when the autonomous driving system fails;

[0024] a processing unit, configured to determine a target acquisition scenario of the autonomous driving system according to the first data set, wherein the target acquisition scenario is a scenario in which the autonomous driving system fails;

[0025] A sending unit is used to send a target data collection instruction to a vehicle, wherein the target data collection instruction is used to instruct the vehicle to collect data in the target collection scenario, and the vehicle is equipped with the automatic driving system.

[0026] Optionally, the processing unit is specifically used to determine a correspondence between multiple failure scenarios and failure causes based on the first data set, the failure cause being the reason that causes the failure of the autonomous driving system, and the failure scenario indicating the environment in which the autonomous driving system fails; based on the correspondence between the multiple failure scenarios and the failure causes, determine a target collection scenario from the multiple failure scenarios, the probability that the autonomous driving system fails due to the failure cause corresponding to the target collection scenario is higher than the probability that the autonomous driving system fails due to the failure cause corresponding to the non-target collection scenario.

[0027] Optionally, the acquisition unit is further used to acquire data collected by the vehicle in the target collection scenario; the processing unit is further used to update the automatic driving system according to the data.

[0028] In a fourth aspect, an embodiment of the present application provides a data collection device, which is applied to a vehicle equipped with an automatic driving system, and the device includes:

[0029] an acquisition unit, configured to acquire a target data acquisition instruction sent by a server, wherein the target data acquisition instruction is used to instruct the vehicle to collect data under a target acquisition scenario, wherein the target acquisition scenario is a cause of failure of the automatic driving system;

[0030] A collection unit, configured to collect the data according to the target data collection instruction in response to the vehicle satisfying the target collection scenario;

[0031] A sending unit is used to send the data to the server.

[0032] Optionally, the target data collection instruction further includes a target data collection rule, wherein the target data collection rule is used to indicate a method in which the vehicle collects the data, and the collection unit is further used to collect the data according to the target data collection rule.

[0033] In a fifth aspect, an embodiment of the present application provides a server, comprising a memory and a processor, wherein the memory is used to store instructions or codes, and the processor is used to execute the instructions or codes so that the device executes the method for triggering data collection as described in any one of the first aspects above.

[0034] In a sixth aspect, an embodiment of the present application provides a vehicle, comprising a data acquisition device, a memory and a processor, wherein the data acquisition device is used to collect data, the memory is used to store instructions or codes, and the processor is used to execute the instructions or codes so that the device executes the data acquisition method described in any one of the aforementioned second aspects.

[0035] In the seventh aspect, an embodiment of the present application provides a computer storage medium, in which a code is stored. When the code is executed, the device executing the code implements the method for triggering data collection as described in any one of the first aspect, or implements the method for data collection as described in any one of the second aspect.

[0036] The embodiment of the present application provides a method for triggering data collection, a method for data collection, and a device. Among them, the method for triggering data collection can be applied to a server, and the method for data collection can be applied to a vehicle equipped with an automatic driving system. Specifically, when executing the technical solution provided in the embodiment of the present application, the server can obtain a first data set, and the first data set includes multiple data, and the multiple data are data generated when the automatic driving system fails. Then, the processor can determine the target collection scenario of the automatic driving system according to the first data set. The target collection scenario is a scenario in which the automatic driving system fails. After determining the target collection scenario, the server can send a target data collection instruction to the vehicle equipped with the automatic driving system to instruct the vehicle to collect data under the target collection scenario. According to the first data set, the scenario when the automatic driving system is prone to failure is determined. Then send a target data collection instruction to the vehicle so that the vehicle collects data when the automatic driving system is prone to failure. In this way, the vehicle does not collect data when the automatic driving system is not prone to failure, and collects data when the automatic driving system is prone to failure. In this way, the vehicle is controlled to collect data in a targeted manner to reduce the amount of data uploaded by the vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in this embodiment or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0038] Figure 1 A signaling interaction diagram of a method for triggering data collection and a method for data collection provided in an embodiment of the present application;

[0039] Figure 2 A schematic diagram of a structure of a device for triggering data collection provided in an embodiment of the present application;

[0040] Figure 3 A schematic diagram of the structure of a data acquisition device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0041] The autonomous driving system, also known as the assisted driving system, can be used to simulate the driver's behavior when driving the vehicle to automatically control the vehicle, thereby reducing the driver's operation of driving the vehicle and improving the driving experience. At present, with the development of autonomous driving technology, the update of the autonomous driving system is increasingly dependent on the data collected when the vehicle is driving. Based on the data collected when the vehicle is driving, the autonomous driving system can be optimized so that the decision of the autonomous driving system is closer and closer to the actual operation of the driver.

[0042] For example, in order to collect valuable data, sensors such as cameras, throttle sensors, and radars can be deployed on the vehicle. When the driver is driving the vehicle, the vehicle's controller, such as the vehicle's electronic control unit (ECU) or on-board computer, can record information about the driver's operations to control the vehicle, as well as information about the environment in which the vehicle is located when the driver performs the operation. Among them, environmental information may include, for example, information about the environment in which the vehicle is located and the distance between the vehicle and surrounding obstacles. The vehicle can send environmental information and operation information to the server through the network. After receiving the data sent by the vehicle, the server can retrain the autonomous driving system based on the environmental information and operation information. In this way, by continuously training the autonomous driving system, the autonomous driving system is more closely matched with the driver's actual operations.

[0043] However, the autonomous driving system involves access to multi-source data, and the amount of data collected is huge. Specifically, the amount of data collected by a car during a day's driving may be at the GB level or higher. Obviously, such a huge amount of data consumes a lot of resources such as data transmission, storage, computing, and human participation.

[0044] In order to reduce the amount of data collected and uploaded by the vehicle, an embodiment of the present application provides a method for triggering data collection and a method for data collection. The method for triggering data collection is applied to a server, and the method for data collection is applied to a vehicle equipped with an automatic driving system. The server can send instructions to the vehicle to instruct the vehicle to collect data. The vehicle returns the data collected by the vehicle to the server via the Internet or a wired connection.

[0045] Specifically, the server may be a server or server cluster for training or updating an autonomous driving model. The data collection method may be executed by an onboard computer or controller of a vehicle equipped with an autonomous driving system. The vehicle may also be equipped with one or more components for collecting data, such as cameras, throttle sensors, radars, and other devices.

[0046] Optionally, the technical solution provided in the embodiment of the present application is applied to a controller, which may be an electronic control unit (ECU) on a vehicle, or other device on the vehicle with data processing capabilities.

[0047] The following is from the perspective of the server and the vehicle respectively. The method for triggering data collection and the method for data collection provided in the embodiments of the present application are described. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0048] See also Figure 1 , Figure 1 An information interaction diagram of a method for triggering data collection and a method for data collection provided in an embodiment of the present application includes:

[0049] S101: The server obtains a first data set.

[0050] In order to control the vehicle to collect data in a suitable scenario, the server may first obtain a first data set. The first data set includes data of the autonomous driving system when a failure event occurs. The failure event of the autonomous driving system is also called the failure of the autonomous driving system, which means that the operation determined by the autonomous driving system is different from the actual operation of the driver. Optionally, the data in the first data set may be collected when the driver is driving the vehicle.

[0051] That is to say, when the driver is driving a vehicle equipped with an automatic driving system, the on-board computer (or other controllers on the vehicle) can collect data on the driver's operation of the vehicle, as well as data output by the automatic driving system for operating the vehicle. Then, the data on the driver's operation of the vehicle and the data on the automatic driving system operating the vehicle can be compared to see if they are the same. If the two are different, it means that the decision of the automatic driving system is different from the driver's decision, and a failure event occurs in the automatic driving system. Then, the vehicle can collect data through sensors to obtain the data of the automatic driving system in the first data set when a failure event occurs.

[0052] It is understandable that the first data set may also include data of the autonomous driving system when it is not failed. For example, the first data set may be data used in the test phase of the autonomous driving system. When testing the autonomous driving system, in order to ensure the comprehensiveness of the test, the vehicle may be controlled to continuously collect data. Accordingly, the collected first data set includes both data when the autonomous driving system is failed and data generated when the autonomous driving system is not failed.

[0053] S102: The server determines a target acquisition scene according to the first data set.

[0054] After acquiring the first data set, the server may determine a target acquisition scenario. The target acquisition scenario is the scenario in which the vehicle is located when a failure event occurs in the autonomous driving system. The target acquisition scenario may include the environment in which the vehicle is located, and may also include the operating condition information of the vehicle. For example, if the autonomous driving system fails when the vehicle merges to the right, then the target acquisition scenario may include "merging to the right". Alternatively, if the autonomous driving system fails when there is an obstacle 50 meters in front of the vehicle, then the target acquisition scenario may include "an obstacle 50 meters in front of the vehicle". It is understandable that the target acquisition scenario may also be represented in other ways.

[0055] The following describes how the server determines the target collection scene.

[0056] After acquiring the first data set, the server can analyze the causes of the failure event of the autonomous driving system based on the first data set, and then obtain the target collection scenario. Specifically, the server can perform attribution analysis on the first data set, classify different failure scenarios into failure causes, and obtain a failure knowledge base. The failure knowledge base includes the association between failure scenarios and failure causes, indicating the causes of the failure of the autonomous driving system under the failure scenario.

[0057] Next, the server can count the number of failures corresponding to each failure cause, determine the important failure causes, and determine the failure scenarios corresponding to the important failure causes as target acquisition scenarios. Among them, the important failure causes refer to failure causes that cause the number and / or frequency of failures of the autonomous driving system to be high. In other words, the probability that the failure causes corresponding to the target acquisition scenarios cause the failure of the autonomous driving system is higher than the probability that the failure causes corresponding to the non-target acquisition scenarios cause the failure of the autonomous driving system. For example, the server can count the number of failures corresponding to each failure cause, and determine the failure causes whose failure times exceed the failure times threshold as important failure causes, and then obtain the corresponding target acquisition scenarios.

[0058] Specifically, in some possible implementations, the failure causes and failure scenarios can be clustered to obtain a visualized failure heat map. The failure heat map indicates the frequency of failure causes corresponding to multiple failure scenarios. In this way, the failure heat map can clearly reflect the frequency of scenarios and causes that cause the failure of the autonomous driving system. In this way, technicians can determine the main causes and main scenarios that cause the failure of the autonomous driving system based on the failure heat map, which is convenient for technicians to update the autonomous driving system.

[0059] In some possible implementations, the server can also be used to analyze the severity of failure events and determine the failure scenarios corresponding to failure events with high severity as target acquisition scenarios. That is, the severity of failure events corresponding to target acquisition scenarios is higher than the severity of failure events corresponding to non-target acquisition scenarios. For example, for failure event A "accelerator and brake judgment error" and failure event B "lighting control error", failure event A incorrectly controls the accelerator and brake of the vehicle, resulting in more serious consequences. Therefore, the server can determine the failure scenario corresponding to failure event A as the target acquisition scenario.

[0060] S103: The server sends a target data collection instruction to the vehicle.

[0061] After determining the target collection scenario, the server sends a target data collection instruction to the vehicle. The target data collection instruction is used to instruct the vehicle to collect data in the target collection scenario.

[0062] Optionally, in addition to the target collection scenario, the target data collection instruction may also include a target data collection rule. Accordingly, the target data collection instruction is also used to instruct the vehicle to collect data according to the target data collection rule.

[0063] Specifically, the target data collection rule is a rule for collecting data. For example, if the vehicle includes a camera, and the rated data collection rule of the camera is 30 frames per second, the camera will collect 30 frames of images per second. In order to reduce the amount of data collected by the vehicle, the target data collection rule can be used to instruct the camera to collect 15 frames of images per second. Optionally, the target data collection rule can also be used to indicate the time period for the vehicle to collect data. For example, the target data collection rule can be used to indicate the length of time the camera collects data. In this way, the strategy for collecting vehicle data is adjusted through the target data collection rule, so that the data collected by the vehicle is the key data for updating the autonomous driving system on the basis of reducing the amount of data collected and uploaded by the vehicle.

[0064] In the embodiment of the present application, the target data collection instruction is used to instruct the vehicle to collect data in the target collection scenario, and does not limit the vehicle to collect data immediately after receiving the target data collection instruction. Optionally, if the vehicle is not in the target collection scenario after receiving the target data collection instruction, the vehicle may not collect data until the vehicle is determined to be in the target collection scenario and then collects data. Alternatively, the server may also issue the target data collection instruction after determining that the vehicle is in the target collection scenario, and the embodiment of the present application does not limit this.

[0065] S104: The vehicle collects data according to the target data collection instruction.

[0066] After receiving the target data collection instruction, the vehicle can collect data according to the target data collection instruction. Specifically, after determining that the scene in which the vehicle is located matches the target collection scene, the vehicle can collect data through the sensor. If the target data collection instruction also includes a target data collection rule, the vehicle can collect data according to the target data collection rule.

[0067] In an embodiment of the present application, the data collected by the vehicle includes the driver's operating data and the vehicle's environmental data.

[0068] In some possible implementations, the vehicle may also collect data according to target data collection rules after determining that the driver's actual operation is different from the operation determined by the autonomous driving system, which will not be elaborated here.

[0069] S105: The vehicle sends the collected data to the server.

[0070] After collecting the data, the vehicle can send the data collected by the vehicle to the server via the Internet. Optionally, the vehicle can send the collected data to the server in real time after collecting the data, or the vehicle can also send the collected data to the server periodically (e.g., daily).

[0071] After receiving the data collected by the vehicle, the server can update the autonomous driving model according to the data collected by the vehicle, so that the decision of the autonomous driving model is closer to that of humans. In addition, the server can also update the aforementioned failure database and failure heat map according to the data collected by the vehicle. In other words, the server can add the data collected by the vehicle to the first data set, and determine the target collection scenario according to the new first data set.

[0072] The embodiment of the present application provides a method for triggering data collection and a method for data collection. Among them, the method for triggering data collection can be applied to a server, and the method for data collection can be applied to a vehicle equipped with an automatic driving system. Specifically, when executing the technical solution provided in the embodiment of the present application, the server can obtain a first data set, and the first data set includes multiple data, and the multiple data are data generated when the automatic driving system fails. Then, the processor can determine the target collection scenario of the automatic driving system according to the first data set. The target collection scenario is a scenario in which the automatic driving system fails. After determining the target collection scenario, the server can send a target data collection instruction to the vehicle equipped with the automatic driving system to instruct the vehicle to collect data under the target collection scenario. According to the first data set, the scenario when the automatic driving system is prone to failure is determined. Then send a target data collection instruction to the vehicle so that the vehicle collects data again when the automatic driving system is prone to failure. In this way, the vehicle does not collect data when the automatic driving system is not prone to failure, and collects data when the automatic driving system is prone to failure. In this way, the vehicle is controlled to collect data in a targeted manner to reduce the amount of data uploaded by the vehicle.

[0073] The above is a method for triggering data collection and some specific implementations of the data collection method provided in the embodiment of the present application. Based on this, the present application also provides a corresponding device. The device provided in the embodiment of the present application will be introduced from the perspective of functional modularization.

[0074] See also Figure 2 The structure diagram of the device 200 for triggering data collection is shown, and the device 200 includes an acquisition unit 210, a processing unit 220 and a sending unit 230.

[0075] The acquisition unit 210 is used to acquire a first data set, where the first data set includes data generated by the autonomous driving system when it fails.

[0076] The processing unit 220 is used to determine a target acquisition scenario of the autonomous driving system based on the first data set, where the target acquisition scenario is a scenario in which the autonomous driving system fails.

[0077] The sending unit 230 is used to send a target data collection instruction to the vehicle, wherein the target data collection instruction is used to instruct the vehicle to collect data in the target collection scenario, and the vehicle is equipped with the automatic driving system.

[0078] See also Figure 3 The structure diagram of the data acquisition device 300 shown in FIG. 3 includes an acquisition unit 310 , a collection unit 320 and a sending unit 330 .

[0079] The acquisition unit 310 is used to acquire a target data collection instruction sent by a server, wherein the target data collection instruction is used to instruct the vehicle to collect data under a target collection scenario, and the target collection scenario is the cause of the failure of the automatic driving system.

[0080] The collection unit 320 is used to collect the data according to the target data collection instruction in response to the vehicle satisfying the target collection scenario.

[0081] The sending unit 330 is configured to send the data to the server.

[0082] The embodiment of the present application provides a device for triggering data collection and a device for data collection. Among them, the device for triggering data collection can be applied to a server, and the device for data collection can be applied to a vehicle equipped with an automatic driving system. Specifically, when executing the technical solution provided in the embodiment of the present application, the server can obtain a first data set, and the first data set includes multiple data, and the multiple data are data generated when the automatic driving system fails. Then, the processor can determine the target collection scenario of the automatic driving system according to the first data set. The target collection scenario is a scenario in which the automatic driving system fails. After determining the target collection scenario, the server can send a target data collection instruction to the vehicle equipped with the automatic driving system to instruct the vehicle to collect data under the target collection scenario. According to the first data set, the scenario when the automatic driving system is prone to failure is determined. Then send a target data collection instruction to the vehicle so that the vehicle collects data when the automatic driving system is prone to failure. In this way, the vehicle does not collect data when the automatic driving system is not prone to failure, and collects data when the automatic driving system is prone to failure. In this way, the vehicle is controlled to collect data in a targeted manner to reduce the amount of data uploaded by the vehicle. Optionally, in some possible implementations, the processing unit 220 is further used to determine that the engine does not have any one or more of an open circuit fault, a power short circuit fault, and a ground short circuit fault.

[0083] Optionally, in some possible implementations, the processing unit 220 is specifically used to determine the correspondence between multiple failure scenarios and failure causes based on the first data set, the failure cause is the reason that causes the failure of the autonomous driving system, and the failure scenario indicates the environment in which the autonomous driving system fails; based on the correspondence between the multiple failure scenarios and the failure causes, determine a target collection scenario from the multiple failure scenarios, the probability that the autonomous driving system fails due to the failure cause corresponding to the target collection scenario is higher than the probability that the autonomous driving system fails due to the failure cause corresponding to the non-target collection scenario.

[0084] Optionally, in some possible implementations, the acquisition unit 210 is further configured to acquire data collected by the vehicle in the target collection scenario. The processing unit 220 is further configured to update the automatic driving system according to the data.

[0085] Optionally, in some possible implementations, the target data collection instruction also includes a target data collection rule, and the target data collection rule is used to indicate the manner in which the vehicle collects the data. The collection unit 320 is further used to collect the data according to the target data collection rule.

[0086] The embodiments of the present application also provide corresponding servers, vehicles and computer storage media for implementing the solutions provided in the embodiments of the present application.

[0087] The server includes a memory and a processor, the memory is used to store instructions or codes, and the processor is used to execute the instructions or codes so that the device executes the method for triggering data collection described in any embodiment of the present application.

[0088] The vehicle includes a data acquisition device, a memory and a processor, the data acquisition device is used to collect data, the memory is used to store instructions or codes, and the processor is used to execute the instructions or codes so that the device executes the data acquisition method described in any embodiment of the present application.

[0089] The computer storage medium stores codes, and when the codes are executed, the device executing the codes implements the method for triggering data collection or the method for data collection described in any embodiment of the present application.

[0090] The "first" and "second" in the names such as "first" and "second" (if any) mentioned in the embodiments of the present application are only used as name identifiers and do not represent the first or second in order.

[0091] Through the description of the above implementation methods, it can be known that those skilled in the art can clearly understand that all or part of the steps in the above-mentioned embodiment method can be implemented by means of software plus a general hardware platform. Based on such an understanding, the technical solution of the present application can be embodied in the form of a software product, which can be stored in a storage medium, such as a read-only memory (ROM) / RAM, a magnetic disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network communication device such as a router) to execute the methods described in each embodiment of the present application or some parts of the embodiments.

[0092] Each embodiment in this specification is described in a progressive manner, and the same and similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative work.

[0093] The above description is merely an exemplary embodiment of the present application and is not intended to limit the protection scope of the present application.

Claims

1. A method for triggering data collection, characterized in that: The method is applied to a server, and the method comprises: The server acquires a first data set, the first data set including data generated by the autonomous driving system when a failure event occurs, the failure event being that data used by the driver to operate the vehicle is different from data used by the autonomous driving system to operate the vehicle; The server determines, according to the first data set, a target acquisition scenario of the autonomous driving system, where the target acquisition scenario is a scenario in which the autonomous driving system fails; The server sends a target data collection instruction to the vehicle, wherein the target data collection instruction is used to instruct the vehicle to collect data in the target collection scenario, and the vehicle is equipped with the automatic driving system; The server determines, based on the first data set, a target acquisition scenario for the autonomous driving system, including: the server determines, based on the first data set, a correspondence between multiple failure scenarios and failure causes, the failure cause being a cause causing the failure of the autonomous driving system, and the failure scenario indicating an environment in which the autonomous driving system fails; the server determines, based on the correspondence between the multiple failure scenarios and the failure causes, a target acquisition scenario from the multiple failure scenarios, the probability that the autonomous driving system fails due to the failure cause corresponding to the target acquisition scenario is higher than the probability that the autonomous driving system fails due to the failure cause corresponding to the non-target acquisition scenario.

2. The method according to claim 1, characterized in that: The method further comprises: The server obtains data collected by the vehicle in the target collection scenario; The server updates the autonomous driving system based on the data.

3. A method for data collection, characterized in that: The method is applied to a vehicle equipped with an automatic driving system, and the method comprises: The vehicle acquires a target data collection instruction sent by a server, wherein the target data collection instruction is used to instruct the vehicle to collect data under a target collection scenario, and the target collection scenario is a cause of failure of the automatic driving system; In response to the vehicle satisfying the target collection scenario, the vehicle collects the data according to the target data collection instruction; The vehicle sends the data to the server; The target collection scenario is determined by the server according to a first data set, wherein the first data set includes data generated by the autonomous driving system when a failure event occurs, wherein the failure event is that data used by the driver to operate the vehicle is different from data used by the autonomous driving system to operate the vehicle; The server determines the target acquisition scenario of the autonomous driving system based on the first data set, including the server determining the correspondence between multiple failure scenarios and failure causes based on the first data set, the failure cause is the cause that causes the failure of the autonomous driving system, and the failure scenario indicates the environment in which the autonomous driving system fails; the server determines the target acquisition scenario from the multiple failure scenarios based on the correspondence between the multiple failure scenarios and the failure causes, the probability that the autonomous driving system fails due to the failure cause corresponding to the target acquisition scenario is higher than the probability that the autonomous driving system fails due to the failure cause corresponding to the non-target acquisition scenario.

4. The method according to claim 3, characterized in that The target data collection instruction further includes a target data collection rule, wherein the target data collection rule is used to indicate a method in which the vehicle collects the data. The vehicle collects the data according to the target data collection instruction, including: The vehicle collects the data according to the target data collection rule.

5. A device for triggering data collection, characterized in that: The device is applied to a server, and comprises: an acquisition unit, configured to acquire a first data set, wherein the first data set includes data generated by the automatic driving system when a failure event occurs, wherein the failure event is that data used by the driver to operate the vehicle is different from data used by the automatic driving system to operate the vehicle; a processing unit, configured to determine a target acquisition scenario of the autonomous driving system according to the first data set, wherein the target acquisition scenario is a scenario in which the autonomous driving system fails; a sending unit, configured to send a target data collection instruction to a vehicle, wherein the target data collection instruction is used to instruct the vehicle to collect data in the target collection scenario, and the vehicle is equipped with the automatic driving system; The processing unit is specifically used to determine the correspondence between multiple failure scenarios and failure causes based on the first data set, the failure cause is the reason that causes the failure of the autonomous driving system, and the failure scenario indicates the environment in which the autonomous driving system fails; based on the correspondence between the multiple failure scenarios and the failure causes, determine a target collection scenario from the multiple failure scenarios, the probability that the autonomous driving system fails due to the failure cause corresponding to the target collection scenario is higher than the probability that the autonomous driving system fails due to the failure cause corresponding to the non-target collection scenario.

6. The device according to claim 5, characterized in that The acquisition unit is further used to acquire the data collected by the vehicle in the target collection scenario; The processing unit is also used to update the automatic driving system according to the data.

7. A data collection device, characterized in that: The device is applied to a vehicle equipped with an automatic driving system, and the device comprises: an acquisition unit, configured to acquire a target data acquisition instruction sent by a server, wherein the target data acquisition instruction is used to instruct the vehicle to collect data under a target acquisition scenario, wherein the target acquisition scenario is a cause of failure of the automatic driving system; A collection unit, configured to collect the data according to the target data collection instruction in response to the vehicle satisfying the target collection scenario; A sending unit, configured to send the data to the server; The target collection scenario is determined by the server according to a first data set, wherein the first data set includes data generated by the autonomous driving system when a failure event occurs, wherein the failure event is that data used by the driver to operate the vehicle is different from data used by the autonomous driving system to operate the vehicle; The server determines the target acquisition scenario of the autonomous driving system based on the first data set, including the server determining the correspondence between multiple failure scenarios and failure causes based on the first data set, the failure cause is the cause that causes the failure of the autonomous driving system, and the failure scenario indicates the environment in which the autonomous driving system fails; the server determines the target acquisition scenario from the multiple failure scenarios based on the correspondence between the multiple failure scenarios and the failure causes, the probability that the autonomous driving system fails due to the failure cause corresponding to the target acquisition scenario is higher than the probability that the autonomous driving system fails due to the failure cause corresponding to the non-target acquisition scenario.

8. The device according to claim 7, characterized in that The target data collection instruction also includes a target data collection rule, and the target data collection rule is used to indicate the way in which the vehicle collects the data. The collection unit is further used to collect the data according to the target data collection rule.

Images