Perception data processing method and device, storage medium and electronic equipment

Abstract

The invention provides a sensing data processing method and device, a storage medium and electronic equipment. The method comprises the following steps: acquiring acquired sensing data and a plurality of preset driving scene templates; according to the constraint condition of each driving scene template, a target driving scene template corresponding to each dynamic element is determined from the multiple driving scene templates, and track information of the dynamic elements in the sensing data conforms to the constraint condition corresponding to the target driving scene template; and generating a target scene file according to the sensing data and the target driving scene template corresponding to each dynamic element. The perception data can be automatically converted into the scene file according to the target scene template, the efficiency of converting the perception data into the scene file is improved, and then the training efficiency of a perception algorithm is improved.

Description

Technical Field

[0001] The present disclosure relates to the field of autonomous driving technology, and in particular to a method, device, storage medium and electronic device for processing perception data. Background Art

[0002] OSC (OpenScenario, an open standard) is a standard format for traffic simulation and vehicle motion data exchange. OSC scenario files that carry vehicle motion trajectory data can be used to support the development and testing of autonomous vehicles in a simulated environment. Using OSC scenario files to test autonomous vehicles can improve vehicle development and testing efficiency and promote the maturity of autonomous driving technology.

[0003] Currently, when converting the motion trajectory data of a vehicle into an OSC scene file, it is necessary to manually edit it according to the driving scene of the vehicle to restore the traffic scene, which leads to the problem of low efficiency in generating OSC scene files. Summary of the invention

[0004] In view of this, the present disclosure provides a method, device, storage medium and electronic device for processing perception data, which can improve the efficiency of converting perception data into scene files.

[0005] According to a first aspect of an embodiment of the present disclosure, a method for processing perception data is provided, the method comprising:

[0006] Acquire the collected perception data and multiple preset driving scene templates;

[0007] According to the constraint conditions of each driving scene template, determine a target driving scene template corresponding to each dynamic element from the multiple driving scene templates, wherein the trajectory information of the dynamic element in the perception data meets the constraint conditions corresponding to the target driving scene template;

[0008] A target scene file is generated according to a target driving scene template corresponding to the dynamic elements in the perception data.

[0009] In one embodiment, determining a target driving scene template corresponding to each dynamic element from the plurality of driving scene templates according to the constraint conditions of each driving scene template includes:

[0010] Determining categories and trajectory information of multiple dynamic elements in the perception data;

[0011] Determining, according to the categories of the multiple dynamic elements, at least one driving scene template of the same category corresponding to each dynamic element;

[0012] The trajectory information of each dynamic element is matched with the constraint conditions of at least one corresponding driving scene template of the same category to determine the target driving scene template.

[0013] In one embodiment, matching the trajectory information of each dynamic element with the constraint condition of at least one corresponding driving scene template of the same category to determine the target driving scene template includes:

[0014] Matching the trajectory points in the trajectory information of the dynamic element with the constraint conditions of at least one driving scene template of the same category respectively, and determining the matching rate between the dynamic element and each driving scene template of the same category;

[0015] Determining a highest matching rate among the matching rates between the dynamic element and at least one driving scene template of the same category;

[0016] If the highest matching rate reaches a preset threshold, the driving scene template of the same category corresponding to the highest matching rate is determined as the target driving scene template.

[0017] In one embodiment, the constraints include at least a start time constraint, an end time constraint and a process constraint;

[0018] The step of matching the trajectory points in the trajectory information of the dynamic element with the constraint conditions of at least one driving scene template of the same category respectively, and determining the matching rate between the dynamic element and each driving scene template of the same category, comprises:

[0019] In the case where there are a first trajectory point and a second trajectory point in the trajectory information of the dynamic element, determining a trajectory point set between the first trajectory point and the second trajectory point, wherein the first trajectory point is a trajectory point in the trajectory information of the dynamic element that meets the start time constraint of the same category driving scene template, the second trajectory point is a trajectory point in the trajectory information of the dynamic element that meets the end time constraint of the same category driving scene template, and the timestamp of the first trajectory point is earlier than the timestamp of the second trajectory point;

[0020] Determining the number of qualified trajectory points in the trajectory point set, wherein the qualified trajectory points are trajectory points in the trajectory point set that meet the process constraint conditions of the same category of driving scene templates;

[0021] The matching rate between the dynamic element and the driving scene template of the same category is determined according to the number of the qualified trajectory points and the total number of the trajectory point set.

[0022] In one embodiment, the trajectory information of the dynamic element includes at least: trajectory points, timestamp information, orientation angle information, coordinate information of the dynamic element compared to the host vehicle, and lane information of the dynamic element.

[0023] In one embodiment, generating a target scene file according to the perception data and a target driving scene template corresponding to each dynamic element includes:

[0024] Determining the time sequence of the target driving scene template corresponding to each dynamic element according to the timestamp in the perception data;

[0025] Determining an arrangement of the target driving scene templates based on a time sequence of the plurality of target driving scene templates;

[0026] Generate an initial scene file based on the arrangement of driving scenes corresponding to the dynamic elements;

[0027] Insert high-precision map data into the initial scene file to generate the target scene file.

[0028] According to a second aspect of an embodiment of the present disclosure, a device for processing perception data is provided, the device comprising:

[0029] An acquisition unit, used to acquire the collected perception data and a plurality of preset driving scene templates;

[0030] A determination unit, configured to determine, according to the constraint conditions of each driving scene template, a target driving scene template corresponding to each dynamic element from the plurality of driving scene templates, wherein the trajectory information of the dynamic element in the perception data meets the constraint conditions corresponding to the target driving scene template;

[0031] The generating unit is used to generate a target scene file according to the target driving scene template corresponding to the dynamic element in the perception data.

[0032] In one embodiment, the determining unit is used to:

[0033] Determining categories and trajectory information of multiple dynamic elements in the perception data;

[0034] Determining, according to the categories of the multiple dynamic elements, at least one driving scene template of the same category corresponding to each dynamic element;

[0035] The trajectory information of each dynamic element is matched with the constraint conditions of at least one corresponding driving scene template of the same category to determine the target driving scene template.

[0036] In one embodiment, the determining unit is used to:

[0037] Matching the trajectory points in the trajectory information of the dynamic element with the constraint conditions of at least one driving scene template of the same category respectively, and determining the matching rate between the dynamic element and each driving scene template of the same category;

[0038] Determining a highest matching rate among the matching rates between the dynamic element and at least one driving scene template of the same category;

[0039] If the highest matching rate reaches a preset threshold, the driving scene template of the same category corresponding to the highest matching rate is determined as the target driving scene template.

[0040] In one embodiment, the determining unit is used to:

[0041] The step of matching the trajectory points in the trajectory information of the dynamic element with the constraint conditions of at least one driving scene template of the same category respectively, and determining the matching rate between the dynamic element and each driving scene template of the same category, comprises:

[0042] In the case where there are a first trajectory point and a second trajectory point in the trajectory information of the dynamic element, determining a trajectory point set between the first trajectory point and the second trajectory point, wherein the first trajectory point is a trajectory point in the trajectory information of the dynamic element that meets the start time constraint of the same category driving scene template, the second trajectory point is a trajectory point in the trajectory information of the dynamic element that meets the end time constraint of the same category driving scene template, and the timestamp of the first trajectory point is earlier than the timestamp of the second trajectory point;

[0043] Determining the number of qualified trajectory points in the trajectory point set, wherein the qualified trajectory points are trajectory points in the trajectory point set that meet the process constraint conditions of the same category of driving scene templates;

[0044] The matching rate between the dynamic element and the driving scene template of the same category is determined according to the number of the qualified trajectory points and the total number of the trajectory point set.

[0045] In one embodiment, the trajectory information of the dynamic element includes at least: trajectory points, timestamp information, orientation angle information, coordinate information of the dynamic element compared to the host vehicle, and lane information of the dynamic element.

[0046] In one embodiment, the generating unit is used to:

[0047] Determining the time sequence of the target driving scene template corresponding to each dynamic element according to the timestamp in the perception data;

[0048] Determining an arrangement of the target driving scene templates based on a time sequence of the plurality of target driving scene templates;

[0049] Generate an initial scene file based on the arrangement of driving scenes corresponding to the dynamic elements;

[0050] Insert high-precision map data into the initial scene file to generate the target scene file.

[0051] According to a third aspect of an embodiment of the present disclosure, there is provided a computer-readable storage medium on which a computer program is stored, and when the program is executed by a processor, the steps of any method described in the first aspect are implemented.

[0052] According to a fourth aspect of an embodiment of the present disclosure, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of any method described in the first aspect are implemented.

[0053] According to a fifth aspect of an embodiment of the present disclosure, a computer program product is provided, on which a computer program is stored. When the computer program product is executed by a processor, the steps of any method described in the first aspect are implemented.

[0054] The technical solution provided by the embodiments of the present disclosure may have the following beneficial effects:

[0055] By matching the perception data with motion trajectories collected by the vehicle with the preset driving scene template, determining the target scene template corresponding to the perception data, and automatically generating a scene file based on the target scene template, the efficiency of converting perception data into scene files is improved, thereby improving the training efficiency of the perception algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 is a flow chart of a method for processing perception data shown in an exemplary embodiment of the present disclosure;

[0057] Figure 2 is a flow chart showing a method of determining a target driving scene template according to an exemplary embodiment of the present disclosure;

[0058] Figure 3 is a schematic diagram of dynamic element trajectory information shown in an exemplary embodiment of the present disclosure;

[0059] Figure 4 is a flowchart of determining a target driving scene template from driving scene templates of the same category shown in an exemplary embodiment of the present disclosure;

[0060] Figure 5 is a flow chart showing a method of generating a target scene file according to an exemplary embodiment of the present disclosure;

[0061] Figure 6is a block diagram of a device for processing perception data shown in an exemplary embodiment of the present disclosure;

[0062] Figure 7 It is a hardware structure diagram of an electronic device shown in an exemplary embodiment of the present disclosure. DETAILED DESCRIPTION

[0063] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0064] The terms used in this disclosure are for the purpose of describing specific embodiments only and are not intended to limit the disclosure. The singular forms of "a", "said" and "the" used in this disclosure and the appended claims are also intended to include plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0065] It should be understood that although the terms first, second, third, etc. may be used in the present disclosure to describe various information, such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present disclosure, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0066] OpenScenario provides a standardized way to describe dynamic elements in a simulation environment, such as the behavior of traffic participants and how these participants interact with other participants and the environment, thereby ensuring that different simulation platforms can use the same scenario definition for testing. By converting the perception data collected by the vehicle into an OpenScenario2.0 (a version of OpenScenario, hereinafter referred to as OSC2.0) file, complex and realistic traffic scenarios can be generated. These scenarios can help perception algorithms train under a variety of conditions, thereby improving the robustness and accuracy of the algorithms, thereby ensuring the reliability of perception algorithms in the real world and promoting the development of autonomous driving technology.

[0067] However, at present, when converting the motion trajectory of dynamic elements in the perception data into an OSC2.0 scene file, it is necessary to manually edit it according to the vehicle's driving scene and restore the traffic scenario, which leads to the problem of low efficiency in generating OSC2.0 scene files.

[0068] Based on this, the present disclosure provides a method for processing perception data, which can be applied to electronic devices that perform simulation tests on vehicle driving scenarios. Figure 1 As shown in the flow chart, the method comprises the following steps:

[0069] S101, acquiring collected perception data and a plurality of preset driving scene templates.

[0070] Perception data can be road data collected by physical sensors on public roads that have been mass-produced, such as image data, video data, or GPS (Global Positioning System) data. Driving scenario templates are pre-designed scenario templates based on various driving scenarios that a vehicle may encounter during driving, such as vehicle entry scenario templates, vehicle exit scenario templates, vehicle oncoming driving scenario templates, and pedestrian crossing scenario templates. Driving scenario templates are used to describe the changes in dynamic elements on the road in different scenarios. For example, the vehicle entry scenario is other vehicles cutting into the same lane as the main vehicle from another lane, and the main vehicle is the vehicle that collects perception data.

[0071] S102: determining a target driving scene template corresponding to each dynamic element from the plurality of driving scene templates according to the constraint conditions of each driving scene template.

[0072] The constraints of each driving scene template are pre-set. For example, for a vehicle cut-in scenario, the constraints may be that before the vehicle cuts in, the target vehicle's driving lane is different from the host vehicle's driving lane; after the vehicle cuts in, the target vehicle's driving lane is the same as the host vehicle's driving lane. Dynamic elements refer to movable target objects on the road, such as other cars, pedestrians, bicycles, etc. on the road except the host vehicle. The target driving scene template refers to a driving scene template in which the trajectory information of the dynamic element in the perception data meets its constraints. For example, if another vehicle cuts in to the host vehicle's lane, the trajectory information of the vehicle will meet the constraints of the vehicle cut-in scenario.

[0073] For each dynamic element, the trajectory information of the dynamic element is matched with the constraints of multiple driving scene templates respectively, and the target scene template corresponding to the trajectory information of the dynamic element is determined in the multiple driving scene templates according to the matching results. It should be noted that the trajectory information of the dynamic element in the perception data meets the constraints corresponding to the target driving scene template.

[0074] In one embodiment, the process of determining the target driving scene template is as follows: Figure 2 As shown, Figure 2 FIG. 1 is a flowchart of determining a target driving scene template according to an exemplary embodiment of the present disclosure, and the flowchart includes the following steps:

[0075] S201, determining categories and trajectory information of multiple dynamic elements in the perception data.

[0076] S202: Determine, according to the categories of the multiple dynamic elements, at least one driving scene template of the same category corresponding to each dynamic element.

[0077] S203: Match the trajectory information of each dynamic element with the constraint condition of at least one corresponding driving scene template of the same category to determine the target driving scene template.

[0078] When determining the target driving scene, first classify the dynamic elements, for example, the dynamic elements are vehicles, pedestrians or bicycles, etc. Then determine the trajectory information of the dynamic elements in the main vehicle coordinate system, and the trajectory information includes at least: the trajectory point of the dynamic element, timestamp information, orientation angle information, the coordinate information of the dynamic element compared to the main vehicle, and the lane information. The timestamp information is the timestamp recorded when the vehicle collects perception data. The timestamp corresponds to the trajectory point in the motion trajectory. Each trajectory point carries a timestamp information. The orientation angle information is the angle between the orientation of the dynamic element and the x-axis of the main vehicle coordinate. The coordinate information of the dynamic element compared to the main vehicle can be represented by the xyz coordinate, and the lane information can be represented by the lane ID. Please refer to Figure 3 , where the lane ID of the main vehicle is 0, the main vehicle coordinates and other vehicle coordinates can be expressed based on the xyz axis of the main vehicle coordinate system, the lane ID of other vehicles at the beginning is 1, and the heading angle relative to the x-axis of the main vehicle is h. The vehicle drives according to this heading angle, and its driving trajectory is as follows Figure 3 As shown by the black dot in the middle, the vehicle eventually drives in the same lane as the main vehicle, achieving vehicle cut-in.

[0079] Then, according to the category of the dynamic element, the driving scene template of the same category corresponding to each dynamic element is determined. For example, if the dynamic element category is vehicle, the driving scene template corresponding to the vehicle may include vehicle cut-in, vehicle cut-out, vehicle driving in the opposite direction, etc. For another example, if the dynamic element category is pedestrian, the driving scene corresponding to pedestrian may include pedestrian crossing, etc.

[0080] After determining the driving scene template of the same category corresponding to each dynamic element according to the category of the dynamic element, the trajectory information of each dynamic element is matched with the driving scene template of the same category respectively, and according to the matching result, the preferred item is determined in the driving scene template of the same category, and it is determined as the target driving scene template corresponding to the dynamic element.

[0081] According to the category information of the dynamic element, the corresponding driving scene template of the same category is determined, and the trajectory information of the dynamic element is matched with the driving scene template of the same category respectively, and the target driving scene template is determined in the driving scene template of the same category, thereby improving the efficiency of matching the dynamic element trajectory information and the driving scene template.

[0082] In one embodiment, the process of determining the target driving scene template from the same category of driving scene templates can be found in Figure 4 The flowchart shown includes the following steps:

[0083] S401 : Match the trajectory points in the trajectory information of the dynamic element with the constraint conditions of at least one driving scene template of the same category respectively, and determine the matching rate between the dynamic element and each driving scene template of the same category.

[0084] The constraints include at least a start time constraint, an end time constraint and a process constraint, wherein the start time constraint is used to determine the state of the dynamic element at the start time of the driving scene, the end time constraint is used to determine the state of the dynamic element at the end time of the driving scene, and the process constraint is used to determine the state of the dynamic element between the start time and the end time in the driving scene.

[0085] Taking the vehicle entry scene as an example, the dynamic elements are other vehicles. The starting time constraint of the vehicle entry scene template is shown in formula (1):

[0086]

[0087] in, represents the constraint condition at the start time, x represents the lateral coordinate of other vehicles compared to the coordinate of the main vehicle, that is, the front and rear distance between other vehicles and the main vehicle, y represents the longitudinal coordinate of other vehicles compared to the coordinate of the main vehicle, that is, the left and right distance between other vehicles and the main vehicle, W represents the width of the lane where the main vehicle is located, and h npcrepresents the direction angle of other vehicles, h npc Indicates the orientation angle of the main vehicle, laneID npc Indicates the lane ID of other vehicles, laneID ego Indicates the lane ID of the host vehicle.

[0088] The end time constraint of the vehicle entering the scene template is shown in formula (2):

[0089]

[0090] in, represents the end time constraint, and l represents the body length of the main vehicle.

[0091] The process constraints of the vehicle entering the scene template are shown in formula (3):

[0092]

[0093] in, Indicates the longitudinal coordinates of other vehicles compared to the main vehicle's coordinates at the start time.

[0094] After determining the constraint conditions of at least one driving scene template of the same category, determine whether there are a first trajectory point and a second trajectory point in the trajectory information of the dynamic element, wherein the first trajectory point is a trajectory point in the trajectory information of the dynamic element that meets the start time constraint conditions of the driving scene template of the same category, and the second trajectory point is a trajectory point in the trajectory information of the dynamic element that meets the end time constraint conditions of the driving scene template of the same category.

[0095] If the first trajectory point and the second trajectory point exist in the trajectory information of the dynamic element, it means that the dynamic element may meet the driving conditions specified by the driving scene template, then the matching rate between the dynamic element and the driving scene template is calculated based on the trajectory point set between the first trajectory point corresponding to the start time and the second trajectory point corresponding to the end time. Specifically, the trajectory points in the trajectory point set that meet the process constraints of the driving scene template can be determined as qualified trajectory points by matching the trajectory points in the trajectory point set with the process constraints. The trajectory points in the trajectory point set that do not meet the process constraints of the driving scene template are determined as unqualified trajectory points. The matching rate is the ratio of the number of qualified trajectory points to the total number of trajectory points in the trajectory point set.

[0096] S402: Determine the highest matching rate among the matching rates between the dynamic element and at least one driving scene template of the same category.

[0097] It should be noted that different driving scene templates may have some of the same constraints. For example, the constraint of the start time of vehicle entry is the same as the constraint of the end time of vehicle exit. Therefore, for the same dynamic element, its motion trajectory may meet the constraints of multiple driving scene templates of the same category, so it is necessary to determine a preferred option among multiple driving scene templates of the same category. In the disclosed embodiment, the preferred option can be determined by comparing the matching rates. Specifically, a highest matching rate is determined among the matching rates of multiple driving scene templates of the same category, and it is determined whether the highest matching rate reaches a preset threshold.

[0098] S403: If the highest matching rate reaches a preset threshold, a driving scene template of the same category corresponding to the highest matching rate is determined as a target driving scene template.

[0099] If the highest matching rate is greater than or equal to the preset threshold, the same category driving scene template corresponding to the highest matching rate is determined as the target driving scene template; if the highest matching rate is less than the preset threshold, it means that the dynamic element fails to match the driving scene template.

[0100] S103: Generate a target scene file according to the target driving scene template corresponding to the dynamic elements in the perception data.

[0101] In some embodiments, the target scene file can be an OSC2.0 scene file. After determining the target driving scene corresponding to each dynamic element, multiple target driving scene templates are sorted and combined according to the timestamp information in the trajectory information of the dynamic element to generate a target scene file. It should be noted that when a vehicle is driving on the road, the driving conditions may be complex and changeable. Therefore, there may be multiple dynamic elements in the perception data, and the driving scene of each dynamic element may be completely different. For example, when a vehicle is driving, it may encounter multiple scenes such as other vehicles cutting in, other vehicles cutting out, and pedestrians crossing within a period of time. A complex simulated traffic scene can be generated through a target scene file (i.e., an OSC2.0 file), so the driving scenes of multiple dynamic elements in the perception data can exist in the target scene file. In order to improve the accuracy of generating simulated traffic scenes according to the target scene file, multiple target driving scene templates can be sorted according to the chronological order.

[0102] In one embodiment, the process of generating the target scene file is as follows: Figure 5 The flowchart shown includes the following steps:

[0103] S501, determining the time sequence of the target driving scene template corresponding to each dynamic element according to the timestamp information of the dynamic elements in the perception data.

[0104] S502: Determine an arrangement of the target driving scene templates based on a time sequence of the plurality of target driving scene templates.

[0105] S503: Generate an initial scene file based on the arrangement of the driving scenes corresponding to the dynamic elements.

[0106] S504: Insert high-precision map data into the initial scene file to generate the target scene file.

[0107] First, based on the timestamp information of each dynamic element, the time sequence of the target driving scene template corresponding to each dynamic element can be determined. Specifically, the time sequence can be determined by sorting the start time or end time in the constraint conditions of each target driving scene template. For example, the first vehicle corresponds to the vehicle entry scene template, and the second vehicle corresponds to the vehicle cut-out scene template. According to the start time constraint of the vehicle entry scene template, the start time of the first vehicle's vehicle cut-in can be determined, and according to the start time constraint of the vehicle cut-out scene template, the start time of the second vehicle's vehicle cut-out can be determined. In this way, the time sequence of the vehicle entry scene template and the vehicle cut-out scene template is determined.

[0108] After determining the time sequence of multiple target driving scene templates, the arrangement method and order of the multiple target driving scene templates in the scene file can be determined according to the time sequence. For example, the time sequence of the vehicle entry scene template and the vehicle cut-out scene template is parallel, that is, the vehicle cut-in of the first vehicle and the vehicle cut-out of the second vehicle are performed at the same time, then the vehicle entry scene template and the vehicle cut-out scene template are arranged in a parallel arrangement. Alternatively, the time sequence of the vehicle entry scene template and the pedestrian crossing scene template is that the vehicle cuts in first and then the pedestrian crosses, then the vehicle entry scene template and the pedestrian crossing scene template can be arranged in a serial arrangement, with the vehicle entry scene template in front and the pedestrian crossing scene template in the back.

[0109] After all target driving templates are arranged, the initial scene file can be obtained. In order to further improve the information in the initial scene file, high-precision map information can be inserted into the initial scene file. High-precision maps are used to provide detailed road information, traffic signs, lane structure and other data, which can increase the realism and accuracy of the simulation scene in the scene file. In addition, weather information, environmental information and other information can also be added to the initial scene file to simulate factors that affect vehicle driving safety, such as extreme weather (such as heavy rain, blizzards, haze, etc.) and different light environments. After generating the target scene file, the target scene file can be imported into the simulation platform to generate various types of data for training perception algorithms, including images, annotated true value data, etc. This improves the accuracy and adaptability of the algorithm and ensures the safety and reliability of the autonomous driving system in a real-world environment.

[0110] In the disclosed embodiment, the perception data with motion tracks collected by the vehicle is matched with a preset driving scene template, the target scene template corresponding to the perception data is determined, and a scene file is automatically generated based on the target scene template, thereby improving the efficiency of converting the perception data into a scene file and thereby improving the training efficiency of the perception algorithm.

[0111] Corresponding to the embodiments of the aforementioned method for processing perception data, the present disclosure also provides embodiments of an apparatus for processing perception data.

[0112] Please refer to Figure 6 , Figure 6 is a block diagram of a device for processing perception data shown in an exemplary embodiment of the present disclosure, the device comprising:

[0113] An acquisition unit 601 is used to acquire the collected perception data and a plurality of preset driving scene templates;

[0114] A determination unit 602 is used to determine a target driving scene template corresponding to each dynamic element from the multiple driving scene templates according to the constraint conditions of each driving scene template, and the trajectory information of the dynamic element in the perception data meets the constraint conditions corresponding to the target driving scene template;

[0115] The generating unit 603 is used to generate a target scene file according to the target driving scene template corresponding to the dynamic element in the perception data.

[0116] In one embodiment, the determining unit 602 is configured to:

[0117] Determining categories and trajectory information of multiple dynamic elements in the perception data;

[0118] Determining, according to the categories of the multiple dynamic elements, at least one driving scene template of the same category corresponding to each dynamic element;

[0119] The trajectory information of each dynamic element is matched with the constraint conditions of at least one corresponding driving scene template of the same category to determine the target driving scene template.

[0120] In one embodiment, the determining unit 602 is configured to:

[0121] Matching the trajectory points in the trajectory information of the dynamic element with the constraint conditions of at least one driving scene template of the same category respectively, and determining the matching rate between the dynamic element and each driving scene template of the same category;

[0122] Determining a highest matching rate among the matching rates between the dynamic element and at least one driving scene template of the same category;

[0123] If the highest matching rate reaches a preset threshold, the driving scene template of the same category corresponding to the highest matching rate is determined as the target driving scene template.

[0124] In one embodiment, the determining unit 602 is configured to:

[0125] The step of matching the trajectory points in the trajectory information of the dynamic element with the constraint conditions of at least one driving scene template of the same category respectively, and determining the matching rate between the dynamic element and each driving scene template of the same category, comprises:

[0126] In the case where there are a first trajectory point and a second trajectory point in the trajectory information of the dynamic element, determining a trajectory point set between the first trajectory point and the second trajectory point, wherein the first trajectory point is a trajectory point in the trajectory information of the dynamic element that meets the start time constraint of the same category driving scene template, the second trajectory point is a trajectory point in the trajectory information of the dynamic element that meets the end time constraint of the same category driving scene template, and the timestamp of the first trajectory point is earlier than the timestamp of the second trajectory point;

[0127] Determining the number of qualified trajectory points in the trajectory point set, wherein the qualified trajectory points are trajectory points in the trajectory point set that meet the process constraint conditions of the same category of driving scene templates;

[0128] The matching rate between the dynamic element and the driving scene template of the same category is determined according to the number of the qualified trajectory points and the total number of the trajectory point set.

[0129] In one embodiment, the trajectory information of the dynamic element includes at least: trajectory points, timestamp information, orientation angle information, coordinate information of the dynamic element compared to the host vehicle, and lane information of the dynamic element.

[0130] In one embodiment, the generating unit 603 is used to:

[0131] Determining the time sequence of the target driving scene template corresponding to each dynamic element according to the timestamp in the perception data;

[0132] Determining an arrangement of the target driving scene templates based on a time sequence of the plurality of target driving scene templates;

[0133] Generate an initial scene file based on the arrangement of driving scenes corresponding to the dynamic elements;

[0134] Insert high-precision map data into the initial scene file to generate the target scene file.

[0135] The implementation process of the functions and effects of each unit in the above-mentioned device is specifically described in the implementation process of the corresponding steps in the above-mentioned method, and will not be repeated here.

[0136] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can refer to the partial description of the method embodiments. The device embodiments described above are only schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the disclosed solution. A person of ordinary skill in the art can understand and implement it without paying any creative work.

[0137] The embodiments of the processing device for perceptual data disclosed in the present disclosure can be applied to electronic devices. The device embodiments can be implemented by software, hardware, or a combination of software and hardware. Taking software implementation as an example, as a device in a logical sense, it is formed by the processor of the electronic device in which it is located reading the corresponding computer program instructions in the non-volatile memory into the memory and running them. From the hardware level, if Figure 7 As shown, it is a hardware structure diagram of an electronic device where the processing device of the perception data disclosed in the present invention is located, except Figure 7 In addition to the processor, memory, network interface, and non-volatile memory shown, the electronic device in which the device in the embodiment is located may also include other hardware according to the actual function of the electronic device, which will not be described in detail.

[0138] Based on the same inventive concept, the present disclosure also provides a computer-readable storage medium on which a computer program is stored, and when the program is executed by a processor, the steps of the above-mentioned method for processing perception data are implemented.

[0139] Based on the same inventive concept, the present disclosure also provides a computer program product on which a computer program is stored. When the computer program product is executed by a processor, the steps of the above-mentioned method for processing perception data are implemented.

[0140] Thus, specific embodiments of the subject matter have been described. Other embodiments are within the scope of the appended claims. In some cases, the actions recited in the claims can be performed in a different order and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the particular order or sequential order shown to achieve the desired results. In some implementations, multitasking and parallel processing may be advantageous.

[0141] The above description is only a preferred embodiment of the present disclosure and is not intended to limit the present disclosure. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present disclosure shall be included in the scope of protection of the present disclosure.

Claims

1. A method for processing perception data, characterized in that: The method comprises: Acquire the collected perception data and multiple preset driving scene templates; According to the constraint conditions of each driving scene template, determine a target driving scene template corresponding to each dynamic element from the multiple driving scene templates, wherein the trajectory information of the dynamic element in the perception data meets the constraint conditions corresponding to the target driving scene template; A target scene file is generated according to a target driving scene template corresponding to the dynamic elements in the perception data.

2. The method according to claim 1, characterized in that The step of determining a target driving scene template corresponding to each dynamic element from the plurality of driving scene templates according to the constraint conditions of each driving scene template comprises: Determining categories and trajectory information of multiple dynamic elements in the perception data; Determining, according to the categories of the multiple dynamic elements, at least one driving scene template of the same category corresponding to each dynamic element; The trajectory information of each dynamic element is matched with the constraint conditions of at least one corresponding driving scene template of the same category to determine the target driving scene template.

3. The method according to claim 2, characterized in that The step of matching the trajectory information of each dynamic element with the constraint condition of at least one corresponding driving scene template of the same category to determine the target driving scene template includes: Matching the trajectory points in the trajectory information of the dynamic element with the constraint conditions of at least one driving scene template of the same category respectively, and determining the matching rate between the dynamic element and each driving scene template of the same category; Determining a highest matching rate among the matching rates between the dynamic element and at least one driving scene template of the same category; If the highest matching rate reaches a preset threshold, the driving scene template of the same category corresponding to the highest matching rate is determined as the target driving scene template.

4. The method according to claim 3, characterized in that: The constraints at least include a start time constraint, an end time constraint and a process constraint; The step of matching the trajectory points in the trajectory information of the dynamic element with the constraint conditions of at least one driving scene template of the same category respectively, and determining the matching rate between the dynamic element and each driving scene template of the same category, comprises: In the case where there are a first trajectory point and a second trajectory point in the trajectory information of the dynamic element, determining a trajectory point set between the first trajectory point and the second trajectory point, wherein the first trajectory point is a trajectory point in the trajectory information of the dynamic element that meets the start time constraint of the same category driving scene template, the second trajectory point is a trajectory point in the trajectory information of the dynamic element that meets the end time constraint of the same category driving scene template, and the timestamp of the first trajectory point is earlier than the timestamp of the second trajectory point; Determining the number of qualified trajectory points in the trajectory point set, wherein the qualified trajectory points are trajectory points in the trajectory point set that meet the process constraint conditions of the same category of driving scene templates; The matching rate between the dynamic element and the driving scene template of the same category is determined according to the number of the qualified trajectory points and the total number of the trajectory point set.

5. The method according to claim 1, characterized in that The trajectory information of the dynamic element includes at least: the trajectory point of the dynamic element, timestamp information, orientation angle information, coordinate information of the dynamic element compared to the host vehicle, and lane information.

6. The method according to claim 1, characterized in that The generating a target scene file according to the target driving scene template corresponding to the dynamic element in the perception data includes: Determining the time sequence of the target driving scene template corresponding to each dynamic element according to the timestamp information of the dynamic elements in the perception data; Determining an arrangement of the target driving scene templates based on a time sequence of the plurality of target driving scene templates; Generate an initial scene file based on the arrangement of driving scenes corresponding to the dynamic elements; Insert high-precision map data into the initial scene file to generate the target scene file.

7. A sensory data processing device, characterized in that: The device comprises: An acquisition unit, used to acquire the collected perception data and a plurality of preset driving scene templates; A determination unit, configured to determine, according to the constraint conditions of each driving scene template, a target driving scene template corresponding to each dynamic element from the plurality of driving scene templates, wherein the trajectory information of the dynamic element in the perception data meets the constraint conditions corresponding to the target driving scene template; The generating unit is used to generate a target scene file according to the target driving scene template corresponding to the dynamic element in the perception data.

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

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

10. A computer program product having a computer program stored thereon, characterized in that: When the computer program product is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images