Data collection method, device and electronic equipment

Abstract

The present invention provides a data collection method, device and electronic device, including: obtaining driving data sent by a data collection device; analyzing the driving data using an automatic driving algorithm; judging whether the analysis result meets the preset data collection rules, wherein the preset data collection rules are rules set by the user for identifying valuable driving data; if the preset data collection rules are met, the target driving data within the preset time range when the analysis result is obtained is saved. The data collection method of the present invention can directly collect valuable driving data, and then use the valuable driving data to iteratively update the automatic driving algorithm, alleviating the technical problem that the existing data collection method cannot directly collect valuable data and increases the workload of subsequent manual screening of valuable data.

Description

Technical Field

[0001] The present invention relates to the technical field of data collection, and in particular to a data collection method, device and electronic equipment. Background Art

[0002] Level 2 autonomous driving has been mass-produced, and higher levels of autonomous driving are being promoted and verified. Currently, most autonomous driving systems use artificial intelligence algorithms based on deep learning. Collecting and obtaining valuable massive data is an effective way to solve the problem of deep learning-based artificial intelligence algorithms relying on massive data.

[0003] At present, when collecting relevant data for autonomous driving, multi-source sensors and data acquisition equipment are installed on the vehicle. In this way, when the vehicle is driving, the massive data collected by the multi-source sensors are sent to the data acquisition equipment, and then the data acquisition equipment saves or uploads the massive data to the cloud. When the autonomous driving algorithm needs to be iteratively updated, relevant personnel will filter out valuable data from the massive data collected above, and then use the valuable data to iteratively update the autonomous driving algorithm.

[0004] In summary, the data collection methods in the prior art cannot directly collect valuable data, which increases the workload of subsequent manual screening of valuable data. Summary of the invention

[0005] In view of this, the purpose of the present invention is to provide a data collection method, device and electronic device to alleviate the technical problem that the data collection method in the prior art cannot directly collect valuable data, which increases the workload of subsequent manual screening of valuable data.

[0006] In a first aspect, an embodiment of the present invention provides a data collection method, including:

[0007] Acquire driving data sent by a data acquisition device;

[0008] Analyzing the driving data using an autonomous driving algorithm;

[0009] Determine whether the analysis result complies with the preset data collection rules, wherein the preset data collection rules are rules set by the user for identifying valuable driving data;

[0010] If the preset data collection rule is met, the target driving data within the preset time range when the analysis result is obtained is saved.

[0011] Furthermore, the data acquisition device includes at least: an image acquisition device, a laser radar, a millimeter wave radar, an IMU, a GPS, a vehicle speed signal acquisition device, a steering wheel angle signal acquisition device, a brake pedal opening and closing degree acquisition device, and an accelerator pedal opening and closing degree acquisition device.

[0012] Furthermore, the driving data is analyzed using an autonomous driving algorithm, including:

[0013] Using a perception algorithm to perceive the driving data to obtain a perception result, wherein the perception result includes: a scene category corresponding to the driving data, location information, category information, and ID information of a target object in the driving data;

[0014] Based on the perception results, a planning algorithm and a positioning algorithm are used to obtain a vehicle planning result and a vehicle positioning result respectively, and based on the vehicle planning result and the vehicle positioning result, a control algorithm is used to obtain a vehicle control result.

[0015] Furthermore, the preset data collection rules include: perception fusion level data collection rules, regulation and control positioning level data collection rules and function level data collection rules;

[0016] The perception fusion level data collection rules include: the driving data is driving data of a predefined scene category, the position information and / or category information of the target object in the driving data is unstable, the ID information of the target object in the driving data is unstable, and the detection results of different target detection algorithms are inconsistent;

[0017] The regulatory positioning level data collection rules include: the vehicle planning result and / or the vehicle control result is greater than a preset threshold, the vehicle planning result and / or the vehicle control result does not meet the preset legal rules, and the vehicle planning result and / or the vehicle control result prompts an error or failure alarm message;

[0018] The functional-level data collection rules include: the vehicle control result is inconsistent with the driver's control operation, the automatic driving related function is activated, the automatic driving function is abnormal, the automatic driving function is degraded, the driver takes over or intervenes during the automatic driving process, and a preset event or preset working condition occurs.

[0019] Furthermore, the target driving data within a preset time range when the analysis result is obtained is saved, including:

[0020] The target driving data is saved to a memory and / or uploaded to a cloud.

[0021] Furthermore, the method further comprises:

[0022] If the preset data collection rule is not met, the driving data corresponding to the analysis result is not saved.

[0023] Furthermore, the method further comprises:

[0024] Iteratively updating the perception algorithm using target driving data determined by the perception fusion level data acquisition rule;

[0025] Iteratively updating the planning algorithm and / or the control algorithm using the target driving data determined by the regulatory positioning level data acquisition rule;

[0026] The target driving data determined by the functional level data acquisition rule is used to iteratively update the algorithm to be updated, wherein the algorithm to be updated is determined after analyzing the perception algorithm, the planning algorithm, the positioning algorithm and the control algorithm.

[0027] In a second aspect, an embodiment of the present invention further provides a data collection device, including:

[0028] An acquisition unit, used for acquiring driving data sent by a data acquisition device;

[0029] An analysis unit, configured to analyze the driving data using an automatic driving algorithm;

[0030] A judgment unit, used to judge whether the analysis result complies with a preset data collection rule, wherein the preset data collection rule is a rule set by a user for identifying valuable driving data;

[0031] A saving unit is used to save the target driving data within a preset time range when the analysis result is obtained if the preset data collection rule is met.

[0032] In a third aspect, an embodiment of the present invention further provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of any one of the methods described in the first aspect when executing the computer program.

[0033] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores machine-executable instructions. When the machine-executable instructions are called and executed by a processor, the machine-executable instructions prompt the processor to execute any method described in the first aspect above.

[0034] In an embodiment of the present invention, a data collection method is provided, including: first obtaining driving data sent by a data collection device; then, using an autonomous driving algorithm to analyze the driving data; then, determining whether the analysis result meets the preset data collection rules set by the user for identifying valuable driving data; if it meets the preset data collection rules, the target driving data within the preset time range when the analysis result is obtained is saved. It can be seen from the above description that the data collection method of the present invention can directly collect valuable driving data, and then use the valuable driving data to iteratively update the autonomous driving algorithm, alleviating the technical problem that the existing data collection method cannot directly collect valuable data and increases the workload of subsequent manual screening of valuable data. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0036] Figure 1 A flow chart of a data collection method provided by an embodiment of the present invention;

[0037] Figure 2 A flow chart of a method for analyzing driving data using an autonomous driving algorithm provided in an embodiment of the present invention;

[0038] Figure 3 A schematic diagram of a data collection device provided by an embodiment of the present invention;

[0039] Figure 4 A schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0040] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] At present, when collecting relevant data for autonomous driving, massive amounts of data are directly collected. When it is necessary to iterate and update the autonomous driving algorithm, relevant personnel will filter out valuable data from the massive amounts of data collected above, and then use the valuable data to iterate and update the autonomous driving algorithm, which increases the workload of subsequent manual screening of valuable data.

[0042] Based on this, this embodiment provides a data collection method, which can directly collect valuable driving data, and then use the valuable driving data to iteratively update the autonomous driving algorithm.

[0043] To facilitate understanding of this embodiment, a data collection method disclosed in an embodiment of the present invention is first introduced in detail.

[0044] Embodiment 1:

[0045] According to an embodiment of the present invention, an embodiment of a data collection method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0046] Figure 1 is a flow chart of a data collection method according to an embodiment of the present invention. Figure 1 As shown, the method comprises the following steps:

[0047] Step S102, acquiring driving data sent by a data acquisition device;

[0048] In an embodiment of the present invention, the data collection method is applied to a collection device. The above-mentioned data collection device at least includes: an image collection device, a laser radar, a millimeter wave radar, an IMU, a GPS, a vehicle speed signal collection device, a steering wheel angle signal collection device, a brake pedal opening and closing degree collection device, and an accelerator pedal opening and closing degree collection device.

[0049] The above data acquisition device adopts multi-source input synchronization technology to ensure the synchronization of various driving data and realize the synchronous acquisition of multiple driving data. The above acquisition equipment (specifically, the acquisition hardware in mass production form) is deployed on the autonomous driving vehicle to obtain the driving data sent by the data acquisition device, and then, according to its own judgment logic, it determines the valuable driving data and saves it (the process is described in detail below).

[0050] Step S104, analyzing the driving data using an autonomous driving algorithm;

[0051] Specifically, the above-mentioned autonomous driving algorithm is a vehicle-side autonomous driving algorithm, which specifically includes: perception algorithm, planning algorithm, positioning algorithm and control algorithm.

[0052] Step S106, determining whether the analysis result complies with a preset data collection rule, wherein the preset data collection rule is a rule set by a user for identifying valuable driving data;

[0053] The above-mentioned valuable driving data refers to driving data in which the existing autonomous driving algorithms exhibit abnormal situations or cannot make correct predictions under specific scenarios and working conditions.

[0054] Step S108: If the preset data collection rules are met, the target driving data within the preset time range when the analysis result is obtained is saved.

[0055] The above-mentioned target driving data is valuable driving data. In this way, a large amount of valuable driving data can be obtained for subsequent data analysis and iterative updates of the autonomous driving algorithm.

[0056] The preset time range may be a preset time before and a preset time after the analysis result is obtained, and the embodiment of the present invention does not specifically limit the preset time range. The target driving data is the driving data obtained within the preset time range.

[0057] In an embodiment of the present invention, a data collection method is provided, including: first obtaining driving data sent by a data collection device; then, using an autonomous driving algorithm to analyze the driving data; then, determining whether the analysis result meets the preset data collection rules set by the user for identifying valuable driving data; if it meets the preset data collection rules, the target driving data within the preset time range when the analysis result is obtained is saved. It can be seen from the above description that the data collection method of the present invention can directly collect valuable driving data, and then use the valuable driving data to iteratively update the autonomous driving algorithm, alleviating the technical problem that the existing data collection method cannot directly collect valuable data and increases the workload of subsequent manual screening of valuable data.

[0058] In an alternative embodiment of the present invention, reference Figure 2 The above step S104, using the autonomous driving algorithm to analyze the driving data, specifically includes:

[0059] Step S201, using a perception algorithm to perceive the driving data to obtain a perception result, wherein the perception result includes: a scene category corresponding to the driving data, location information, category information and ID information of a target object in the driving data;

[0060] Step S202: Based on the perception results, a planning algorithm and a positioning algorithm are used to obtain a vehicle planning result and a vehicle positioning result respectively, and based on the vehicle planning result and the vehicle positioning result, a control algorithm is used to obtain a vehicle control result.

[0061] The above-mentioned autonomous driving algorithms include: perception algorithm, planning algorithm, control algorithm and positioning algorithm.

[0062] In an optional embodiment of the present invention, the above-mentioned preset data collection rules include: perception fusion level data collection rules, regulation and control positioning level data collection rules and function level data collection rules;

[0063] Perception fusion level data collection rules (target driving data determined by such rules are valuable data for iterative updates of perception algorithms) include: driving data of predefined scene categories, unstable location information and / or category information of target objects in driving data, unstable ID information of target objects in driving data, and inconsistent detection results of different target detection algorithms;

[0064] The data collection rules at the regulatory control and positioning level (the target driving data determined by such rules is valuable data for the iterative update of the planning algorithm and / or the control algorithm) include: the vehicle planning result and / or the vehicle control result is greater than a preset threshold, the vehicle planning result and / or the vehicle control result does not meet the preset legal rules, and the vehicle planning result and / or the vehicle control result prompts an error or failure alarm message;

[0065] Functional-level data collection rules (the target driving data determined by this type of rule is mainly aimed at the control level) include: vehicle control results are inconsistent with driver control operations, automatic driving related functions are activated, automatic driving functions are abnormal, automatic driving functions are degraded, the driver takes over or intervenes during automatic driving, and preset events or preset working conditions occur.

[0066] The following is a detailed description of each data collection rule:

[0067] Driving data is driving data of predefined scene categories: During implementation, the perception algorithm classifies the scene corresponding to the driving data (driving data after data fusion collected by the data acquisition device). If the scene corresponding to the driving data is a predefined scene category, it complies with the preset data acquisition rules. For example, the predefined scene categories are ramp scene categories, bridge hole scene categories, and tunnel scene categories. After the perception algorithm performs scene recognition on the image data in the driving data, it determines that the corresponding scene is a bridge hole scene category, which complies with the preset data acquisition rules. It should be noted that the above-mentioned predefined scene categories are scene categories that are prone to collision accidents or data acquisition device failures. The embodiments of the present invention do not specifically limit the above-mentioned predefined scene categories.

[0068] The position information and / or category information of the target object in the driving data is unstable: During implementation, for example, when the perception algorithm perceives the captured image, if the position information and / or category information of the target object is unstable (i.e., the detection box of the target object jumps or the category of the target object jumps), it means that the perception algorithm cannot correctly perceive the image (the above image is valuable driving data).

[0069] The ID information of the target object in the driving data is unstable: When a target object is detected for a long time, a unique ID information will be generated for the target object. If the ID information of the target object changes during the perception process (for example, the ID of the target object originally detected is 1, and then jumps to 3, it means that the perception algorithm is unstable).

[0070] The detection results of different target detection algorithms (i.e., perception algorithms) are inconsistent: for images captured by the image acquisition device, the detection result obtained by the corresponding target detection algorithm is that the target object is a car; while for point cloud data captured by the lidar, the detection result obtained by the corresponding target detection algorithm is that the target object is a person, that is, the detection results of different target detection algorithms are inconsistent.

[0071] The vehicle planning result and / or vehicle control result is greater than the preset threshold: In the current scenario, if the vehicle planning result planned by the autonomous driving algorithm is that the steering wheel turns 50 degrees to the left, and the preset threshold of the steering wheel angle is 30 degrees, that is, the vehicle planning result is greater than the preset threshold, it means that the preset data collection rules are met.

[0072] The vehicle planning result and / or vehicle control result do not comply with the preset legal rules: In the current scenario, if the vehicle planning result does not comply with the traffic rules (for example, the vehicle planning result considers driving forward when the traffic light is red), or there are roads that cannot be driven on the path to the destination planned by the vehicle, the embodiment of the present invention does not impose specific restrictions on the above-mentioned preset legal rules. The above situation indicates that the vehicle planning result and / or vehicle control result do not comply with the preset legal rules.

[0073] The vehicle planning results and / or vehicle control results prompt error or failure alarm information, for example, error signals appear on the vehicle body, planning failure in the algorithm, steering wheel stuck, positioning failure, etc., which can be set as needed.

[0074] The vehicle control result is inconsistent with the driver's control operation: When the vehicle is fully operated by the driver or some of the vehicle's functions are performed by the assisted driving system, the automatic driving controller obtains the real operating environment data (also called driving data) through the data acquisition device and runs the automatic driving algorithm in the background, but does not perform execution control on the vehicle's actuators. When the vehicle control result output by the automatic driving algorithm is different from the actual control operation of the driver, it meets the preset data collection rules.

[0075] Automatic driving related functions are activated: Automatic driving related functions may include: emergency braking function, adaptive cruise function, etc. The embodiments of the present invention do not specifically limit the above automatic driving related functions.

[0076] When the autonomous driving function is abnormal, that is, when a prompt signal of the autonomous driving function abnormality appears, it complies with the preset data collection rules.

[0077] Downgrade of autonomous driving function: For example, the previous autonomous driving was lateral and longitudinal. Horizontal and longitudinal mean that it could change lanes and go straight, but later the autonomous driving could only go straight. This is a downgrade of autonomous driving function. Of course, other autonomous driving functions can also be downgraded.

[0078] During the autonomous driving process, driver takeover or intervention means that the driver takes over or intervenes in the driving during the original autonomous driving process, which means that an emergency situation or improper autonomous driving control is encountered during the driving process. The driving data in this case is also valuable driving data.

[0079] Preset events or preset operating conditions occur: The above-mentioned preset events or preset operating conditions may include: when the driver controls the vehicle or the system controls the vehicle, emergency braking, emergency lane change, road water accumulation, road icing and other events or conditions occur. The embodiments of the present invention do not impose specific restrictions on the above-mentioned preset events and preset operating conditions, and can also be other events or operating conditions.

[0080] The above-mentioned preset data collection rules of the present invention can collect target driving data for different autonomous driving algorithms (perception algorithm, planning algorithm, positioning algorithm and control algorithm). Subsequently, the corresponding target driving data can be directly used to iteratively update the corresponding autonomous driving algorithm, which is more convenient and quick.

[0081] In addition, the above-mentioned preset data collection rules of the present invention are multi-mechanism data collection rules designed by the inventor based on different levels of the autonomous driving algorithm. The collected target driving data is more comprehensive, and the algorithms at each level of the autonomous driving algorithm can be subsequently iterated and updated in a targeted manner, which can reduce the workload of iteratively updating algorithms at all levels of the autonomous driving algorithm. Moreover, the comprehensive and rich target driving data can make the iteratively updated autonomous driving algorithm of the present invention more reliable and stable.

[0082] In an optional embodiment of the present invention, the above step S108, saving the target driving data within the preset time range when the analysis result is obtained, specifically includes: saving the target driving data to a memory and / or uploading it to the cloud. Afterwards, the target driving data is analyzed, annotated, marked, tested for false positives and false negatives, and the algorithm is iteratively updated in a better manner.

[0083] In an optional embodiment of the present invention, the method further comprises:

[0084] If the analysis result does not meet the preset data collection rules, the driving data corresponding to the analysis result will not be saved.

[0085] In an optional embodiment of the present invention, the method further comprises:

[0086] (1) Iteratively update the perception algorithm using target driving data determined by the perception fusion level data collection rules;

[0087] Specifically, the target driving data determined by the perception fusion level data collection rule is valuable data for iterative updating of the perception algorithm. After obtaining the target driving data collected by this rule, this part of the target driving data can be used to iteratively update the perception algorithm.

[0088] (2) Iteratively updating the planning algorithm and / or control algorithm using target driving data determined by the regulatory positioning level data acquisition rules;

[0089] Specifically, the target driving data determined by the regulatory positioning-level data collection rules are valuable data for iterative updating of the planning algorithm and / or the control algorithm. After obtaining the target driving data collected by such rules, this part of the target driving data can be used to iteratively update the planning algorithm and / or the control algorithm.

[0090] (3) The target driving data determined by the functional level data collection rules are used to iteratively update the algorithm to be updated, wherein the algorithm to be updated is determined after analyzing the perception algorithm, planning algorithm, positioning algorithm and control algorithm.

[0091] Specifically, the target driving data determined by the functional-level data collection rules is mainly aimed at the control level. When the autonomous driving algorithm is subsequently iteratively updated, it is necessary to analyze (derive) the algorithms that need to be iteratively updated in the perception, planning, positioning and decision-making levels of the autonomous driving algorithm, and then use the target driving data collected by this rule to iteratively update the determined algorithms that need to be iteratively updated.

[0092] The present invention directly uses the corresponding target driving data to iteratively update the corresponding autonomous driving algorithm, which is more convenient and quick, and reduces the workload of iteratively updating all levels of the autonomous driving algorithm. Moreover, the comprehensive and rich target driving data can make the iteratively updated autonomous driving algorithm of the present invention more reliable and stable.

[0093] Embodiment 2:

[0094] The embodiment of the present invention further provides a data collection device, which is mainly used to execute the data collection method provided in the first embodiment of the present invention. The data collection device provided in the embodiment of the present invention is specifically introduced below.

[0095] Figure 3 is a schematic diagram of a data collection device according to an embodiment of the present invention. Figure 3 As shown, the device mainly includes: an acquisition unit 10, an analysis unit 20, a judgment unit 30 and a storage unit 40, wherein:

[0096] An acquisition unit, used for acquiring driving data sent by a data acquisition device;

[0097] An analysis unit for analyzing driving data using an autonomous driving algorithm;

[0098] A judgment unit, used to judge whether the analysis result complies with a preset data collection rule, wherein the preset data collection rule is a rule set by a user for identifying valuable driving data;

[0099] The storage unit is used to store the target driving data within a preset time range when the analysis result is obtained if the preset data collection rules are met.

[0100] In an embodiment of the present invention, a data collection device is provided, including: first obtaining driving data sent by a data collection device; then, using an autonomous driving algorithm to analyze the driving data; then, determining whether the analysis result meets the preset data collection rules set by the user for identifying valuable driving data; if it meets the preset data collection rules, the target driving data within the preset time range when the analysis result is obtained is saved. It can be seen from the above description that the data collection device of the present invention can directly collect valuable driving data, and then use the valuable driving data to iteratively update the autonomous driving algorithm, alleviating the technical problem that the existing data collection method cannot directly collect valuable data and increases the workload of subsequent manual screening of valuable data.

[0101] Optionally, the data acquisition device includes at least: an image acquisition device, a laser radar, a millimeter wave radar, an IMU, a GPS, a vehicle speed signal acquisition device, a steering wheel angle signal acquisition device, a brake pedal opening and closing degree acquisition device, and an accelerator pedal opening and closing degree acquisition device.

[0102] Optionally, the analysis unit is also used to: use a perception algorithm to perceive the driving data to obtain a perception result, wherein the perception result includes: a scene category corresponding to the driving data, location information, category information and ID information of the target object in the driving data; based on the perception result, use a planning algorithm and a positioning algorithm to respectively obtain a vehicle planning result and a vehicle positioning result, and based on the vehicle planning result and the vehicle positioning result, use a control algorithm to obtain a vehicle control result.

[0103] Optionally, the preset data collection rules include: perception fusion level data collection rules, regulation and control positioning level data collection rules and function level data collection rules; the perception fusion level data collection rules include: driving data is driving data of a predefined scene category, the position information and / or category information of the target object in the driving data is unstable, the ID information of the target object in the driving data is unstable, and the detection results of different target detection algorithms are inconsistent; the regulation and control positioning level data collection rules include: the vehicle planning results and / or vehicle control results are greater than a preset threshold, the vehicle planning results and / or vehicle control results do not comply with the preset legal rules, and the vehicle planning results and / or vehicle control results prompt an error or failure alarm message; the function level data collection rules include: the vehicle control results are inconsistent with the driver's control operation, the automatic driving related functions are activated, the automatic driving functions are abnormal, the automatic driving functions are downgraded, the driver takes over or intervenes during the automatic driving process, and a preset event or preset working condition occurs.

[0104] Optionally, the storage unit is further used to: save the target driving data to a memory and / or upload it to the cloud.

[0105] Optionally, the device is also used to: if a preset data collection rule is not met, then the driving data corresponding to the analysis result is not saved.

[0106] Optionally, the device is also used to: iteratively update the perception algorithm using the target driving data determined by the perception fusion level data collection rules; iteratively update the planning algorithm and / or control algorithm using the target driving data determined by the planning and positioning level data collection rules; iteratively update the algorithm to be updated using the target driving data determined by the functional level data collection rules, wherein the algorithm to be updated is determined after analyzing the perception algorithm, planning algorithm, positioning algorithm and control algorithm.

[0107] The device provided in the embodiment of the present invention has the same implementation principle and technical effects as those of the aforementioned method embodiment. For the sake of brief description, for matters not mentioned in the device embodiment, reference may be made to the corresponding contents in the aforementioned method embodiment.

[0108] like Figure 4As shown, an electronic device 600 provided in an embodiment of the present application includes: a processor 601, a memory 602 and a bus, wherein the memory 602 stores machine-readable instructions executable by the processor 601. When the electronic device is running, the processor 601 communicates with the memory 602 through the bus, and the processor 601 executes the machine-readable instructions to perform the steps of the data collection method as described above.

[0109] Specifically, the above-mentioned memory 602 and processor 601 can be general-purpose memories and processors, which are not specifically limited here. When the processor 601 runs the computer program stored in the memory 602, the above-mentioned data collection method can be executed.

[0110] The processor 601 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit or software instructions in the processor 601. The above processor 601 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiments of the present application can be directly embodied as a hardware decoding processor for execution, or a combination of hardware and software modules in the decoding processor for execution. The software module may be located in a storage medium mature in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 602, and the processor 601 reads the information in the memory 602 and completes the steps of the above method in combination with its hardware.

[0111] Corresponding to the above-mentioned data collection method, an embodiment of the present application also provides a computer-readable storage medium, which stores machine-executable instructions. When the computer-executable instructions are called and executed by a processor, the computer-executable instructions prompt the processor to execute the steps of the above-mentioned data collection method.

[0112] The data collection device provided in the embodiment of the present application can be specific hardware on the device or software or firmware installed on the device. The device provided in the embodiment of the present application, its implementation principle and the technical effect produced are the same as those in the aforementioned method embodiment. For the sake of brief description, the parts not mentioned in the device embodiment can refer to the corresponding contents in the aforementioned method embodiment. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, devices and units described above can all refer to the corresponding processes in the aforementioned method embodiment, and will not be repeated here.

[0113] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0114] For another example, the flowchart and block diagram in the accompanying drawings show the possible architecture, function and operation of the device, method and computer program product according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of the code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or the flowchart, and the combination of the boxes in the block diagram and / or the flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0115] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0116] In addition, each functional unit in the embodiments provided in the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0117] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can essentially or partly contribute to the prior art or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a number of instructions for an electronic device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the vehicle marking method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0118] It should be noted that similar numbers and letters represent similar items in the following figures. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. In addition, the terms "first", "second", "third", etc. are only used to distinguish the description and are not to be understood as indicating or implying relative importance.

[0119] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present application, which are used to illustrate the technical solution of the present application, rather than to limit it. The protection scope of the present application is not limited thereto. Although the present application is described in detail with reference to the aforementioned embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solution recorded in the aforementioned embodiments within the technical scope disclosed in the present application, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solution deviate from the scope of the technical solution of the embodiment of the present application. They should all be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A data collection method, characterized in that: include: Acquire driving data sent by a data acquisition device; Analyzing the driving data using an autonomous driving algorithm; Determine whether the analysis result complies with the preset data collection rules, wherein the preset data collection rules are rules set by the user for identifying valuable driving data; The valuable driving data refers to driving data in which the autonomous driving algorithm has abnormal conditions or cannot make correct predictions under specific scenarios and working conditions; the analysis results include perception results, vehicle planning results, vehicle positioning results, and vehicle control results; If the preset data collection rule is met, the target driving data within the preset time range when the analysis result is obtained is saved; The preset data collection rules include: perception fusion level data collection rules, regulation and control positioning level data collection rules and function level data collection rules; The perception fusion level data collection rules include: the driving data is driving data of a predefined scene category, the position information and / or category information of the target object in the driving data is unstable, the ID information of the target object in the driving data is unstable, and the detection results of different target detection algorithms are inconsistent; the predefined scene category is a ramp scene category, a bridge hole scene category, and a tunnel scene category; The regulatory positioning level data collection rules include: the vehicle planning result and / or the vehicle control result is greater than a preset threshold, the vehicle planning result and / or the vehicle control result does not meet the preset legal rules, and the vehicle planning result and / or the vehicle control result prompts an error or failure alarm message; The functional-level data collection rules include: the vehicle control result is inconsistent with the driver's control operation, the automatic driving related function is activated, the automatic driving function is abnormal, the automatic driving function is degraded, the driver takes over or intervenes during the automatic driving process, and a preset event or preset working condition occurs.

2. The method according to claim 1, characterized in that: The data acquisition device includes at least: an image acquisition device, a laser radar, a millimeter wave radar, an IMU, a GPS, a vehicle speed signal acquisition device, a steering wheel angle signal acquisition device, a brake pedal opening and closing degree acquisition device, and an accelerator pedal opening and closing degree acquisition device.

3. The method according to claim 1, characterized in that The driving data is analyzed using an autonomous driving algorithm, including: Using a perception algorithm to perceive the driving data to obtain a perception result, wherein the perception result includes: a scene category corresponding to the driving data, location information, category information, and ID information of a target object in the driving data; Based on the perception results, a planning algorithm and a positioning algorithm are used to obtain a vehicle planning result and a vehicle positioning result respectively, and based on the vehicle planning result and the vehicle positioning result, a control algorithm is used to obtain a vehicle control result.

4. The method according to claim 1, characterized in that Saving the target driving data within a preset time range when the analysis result is obtained includes: The target driving data is saved to a memory and / or uploaded to a cloud.

5. The method according to claim 1, characterized in that The method further comprises: If the preset data collection rule is not met, the driving data corresponding to the analysis result is not saved.

6. The method according to claim 3, characterized in that The method further comprises: Iteratively updating the perception algorithm using target driving data determined by the perception fusion level data acquisition rule; Iteratively updating the planning algorithm and / or the control algorithm using the target driving data determined by the regulatory positioning level data acquisition rule; The target driving data determined by the functional level data acquisition rule is used to iteratively update the algorithm to be updated, wherein the algorithm to be updated is determined after analyzing the perception algorithm, the planning algorithm, the positioning algorithm and the control algorithm.

7. A data collection device, characterized in that: include: An acquisition unit, used for acquiring driving data sent by a data acquisition device; An analysis unit, configured to analyze the driving data using an automatic driving algorithm; A judgment unit, used to judge whether the analysis result complies with a preset data collection rule, wherein the preset data collection rule is a rule set by a user for identifying valuable driving data; the valuable driving data refers to driving data in which the autonomous driving algorithm has an abnormal situation or cannot be correctly predicted under specific scenarios and working conditions; the analysis result includes a perception result, a vehicle planning result, a vehicle positioning result, and a vehicle control result; A storage unit, configured to store the target driving data within a preset time range when the analysis result is obtained if the preset data collection rule is met; The preset data collection rules include: perception fusion level data collection rules, regulation and control positioning level data collection rules and function level data collection rules; The perception fusion level data collection rules include: the driving data is driving data of a predefined scene category, the position information and / or category information of the target object in the driving data is unstable, the ID information of the target object in the driving data is unstable, and the detection results of different target detection algorithms are inconsistent; the predefined scene category is a ramp scene category, a bridge hole scene category, and a tunnel scene category; The regulatory positioning level data collection rules include: the vehicle planning result and / or the vehicle control result is greater than a preset threshold, the vehicle planning result and / or the vehicle control result does not meet the preset legal rules, and the vehicle planning result and / or the vehicle control result prompts an error or failure alarm message; The functional-level data collection rules include: the vehicle control result is inconsistent with the driver's control operation, the automatic driving related function is activated, the automatic driving function is abnormal, the automatic driving function is degraded, the driver takes over or intervenes during the automatic driving process, and a preset event or preset working condition occurs.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores machine-executable instructions. When the machine-executable instructions are called and executed by a processor, the machine-executable instructions cause the processor to execute the method according to any one of claims 1 to 6.

Images