Interaction behavior extraction method, system and equipment based on driving area map

Abstract

The embodiment of the invention provides an interactive behavior extraction method, which comprises the following steps: acquiring a traffic participant trajectory data set of a target road section, and acquiring a target trajectory data set from the traffic participant trajectory data set; searching a plurality of potential interaction objects of the target motor vehicle in a driving area map corresponding to the driving route of the target motor vehicle according to the trajectory data of the target motor vehicle and the traffic participant trajectory data set; according to the trajectory data of the target motor vehicle and the trajectory data of each potential interaction object, determining a numerical value of an interaction risk measurement index between the target motor vehicle and each potential interaction object; and determining the potential interaction object of which the numerical value of the interaction risk measurement index is within a target numerical value range as a target interaction object, taking the trajectory data of the target motor vehicle and the trajectory data of the corresponding target interaction object as an interaction behavior scene, and adding the interaction behavior scene into an interaction behavior scene set of the target motor vehicle. According to the invention, the comprehensiveness and accuracy of interactive behavior extraction are improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation technology, and in particular to a method, system, device and storage medium for extracting interactive behavior based on a driving area map. Background Art

[0002] Interaction scenarios between traffic participants are essential data for studying and analyzing road traffic safety, as well as developing and testing high-level autonomous driving. Currently, the extraction of these scenarios relies primarily on rules such as relative position and speed changes on lane area maps. This is more applicable to roads with clear lane markings. However, on unstructured roads, such as turning areas at intersections, narrow urban roads, and unpaved roads in rural areas, traffic participants often do not strictly follow lane markings or their own driving habits. Therefore, their interactions are difficult to categorize based on rules such as relative position and speed changes within lane areas. Therefore, relying solely on lane area map extraction rules is difficult to fully and accurately extract interaction scenarios on unstructured roads. Summary of the Invention

[0003] The embodiments of the present invention provide a method, system, device and storage medium for extracting interactive behaviors based on a driving area map, aiming to improve the accuracy and comprehensiveness of interactive behaviors extraction.

[0004] In a first aspect, an embodiment of the present invention provides a method for extracting interactive behaviors based on a driving area map, comprising:

[0005] Acquire a traffic participant trajectory dataset of a target road section, and acquire a target trajectory dataset from the traffic participant trajectory dataset, wherein the target trajectory dataset includes trajectory data of a plurality of target motor vehicles with clear travel routes;

[0006] searching, based on the target motor vehicle's trajectory data and the traffic participant trajectory dataset, for a plurality of potential interaction objects of the target motor vehicle in a driving area map corresponding to the target motor vehicle's driving route, wherein the driving area map is constructed based on the target trajectory dataset and is used to represent drivable areas of the plurality of target motor vehicles on the target road segment;

[0007] determining a value of an interaction risk measurement index between the target motor vehicle and each of the potential interaction objects based on the trajectory data of the target motor vehicle and the trajectory data of each of the potential interaction objects;

[0008] The potential interaction object whose value of the interaction risk measurement index is within the target value range is determined as the target interaction object of the target motor vehicle, and the trajectory data of the target motor vehicle and the corresponding trajectory data of the target interaction object are used as interaction behavior scenes and added to the interaction behavior scene set of the target motor vehicle.

[0009] In a second aspect, an embodiment of the present invention further provides a computer device, comprising a processor, a memory, a computer program stored on the memory and executable by the processor, and a data bus for realizing connection and communication between the processor and the memory, wherein when the computer program is executed by the processor, the interactive behavior extraction method described in the first aspect is realized.

[0010] In a third aspect, an embodiment of the present invention further provides an interactive behavior extraction system based on a driving area map, comprising: a computer device, a height maintenance device, and an image acquisition device, wherein:

[0011] The height maintaining device is deployed on the target road section and is used to carry the image acquisition device, so that the image acquisition device can acquire a plurality of original bird's-eye view image sequences of the traffic flow of the target road section from a bird's-eye view perspective;

[0012] The image acquisition device is mounted on the altitude maintaining device and is used to acquire a plurality of original bird's-eye view image sequences of the traffic flow of the target road section from a bird's-eye view perspective;

[0013] The computer device is configured to obtain the plurality of original bird's-eye view image sequences and obtain a traffic participant trajectory dataset of the target road section based on the plurality of original bird's-eye view image sequences;

[0014] The computer device is further configured to obtain a target trajectory dataset from the traffic participant trajectory dataset, wherein the target trajectory dataset includes trajectory data of a plurality of target motor vehicles with clear driving routes;

[0015] The computer device is further configured to construct a driving area map on the target road segment based on the target trajectory dataset, wherein the driving area map is configured to represent the driving areas of the multiple target motor vehicles on the target road segment;

[0016] The computer device is further configured to search for a plurality of potential interaction objects of the target motor vehicle in the driving area map corresponding to the driving route of the target motor vehicle based on the trajectory data of the target motor vehicle and the traffic participant trajectory dataset;

[0017] The computer device is further configured to determine a value of an interaction risk measurement indicator between the target motor vehicle and each of the potential interaction objects based on the trajectory data of the target motor vehicle and the trajectory data of each of the potential interaction objects;

[0018] The computer device is also used to determine the potential interactive object whose value of the interaction risk measurement index is within the target value range as the target interactive object of the target motor vehicle, and add the trajectory data of the target motor vehicle and the corresponding trajectory data of the target interactive object as interactive behavior scenes to the interactive behavior scene set of the target motor vehicle.

[0019] In a fourth aspect, an embodiment of the present invention further provides a storage medium for computer-readable storage, wherein the storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the interactive behavior extraction method as described in the first aspect.

[0020] The embodiment of the present invention provides an interactive behavior extraction method, system, device and storage medium based on a driving area map. The driving area map in the embodiment of the present invention is constructed based on a target trajectory data set, so that no matter the target road section is a structured road section with clear lane markings and good traffic order, or an unstructured road section with blurred lane markings or even no lane markings, or an unstructured road section where motor vehicles do not strictly follow lane markings, the driving area map can accurately represent the drivable area under different routes of the target road section, thereby ensuring the accuracy of the drivable area. When searching for potential interactive objects, there is no need to search on the entire target road section, but only to search in the driving area map, thereby reducing the search range of potential interactive objects, which not only improves the search efficiency, but also ensures the accuracy of the potential interactive objects searched. Then, the target motor vehicle is further screened out based on the value of the interaction risk measurement index between the target motor vehicle and each potential interactive object. The interactive objects are marked, and the trajectory data of the target motor vehicle and the corresponding trajectory data of the target interactive object are extracted as the interactive behavior scene, thereby improving the comprehensiveness and accuracy of the interactive behavior extraction. Therefore, whether it is a structured road section with clear lane markings and good traffic order, or an unstructured road section with blurred lane markings or even no lane markings, or an unstructured road section where motor vehicles do not strictly follow the lane markings, the interactive behavior scene can be fully and accurately extracted using the driving area map, which improves the universality, comprehensiveness and accuracy of the interactive behavior scene extraction. The driving area map can not only represent the drivable areas under different routes, but also map each trajectory in the trajectory dataset used to construct the driving area map to the corresponding route, so that the routes of the motor vehicle participants in the interactive behavior scene extracted based on the driving area map and the trajectory data of the traffic participants are known and the purpose is clear, thereby improving the interpretability of the interactive behavior scene. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0022] Figure 1 A flowchart of a method for extracting interactive behaviors based on a driving area map provided by an embodiment of the present invention;

[0023] Figure 2 yes Figure 1 Schematic diagram of the sub-step flow of the interactive behavior extraction method in [1].

[0024] Figure 3is a schematic diagram of a bird's-eye view image of a target road section acquired by a first unmanned aerial vehicle in an embodiment of the present invention;

[0025] Figure 4 is a schematic diagram of a bird's-eye view image of a target road section acquired by a second unmanned aerial vehicle in an embodiment of the present invention;

[0026] Figure 5 is an example diagram of parameters included in the static data of traffic participants in an embodiment of the present invention;

[0027] Figure 6 1 is a schematic diagram of a process for correcting the position and shape of a detection frame to be corrected according to an embodiment of the present invention;

[0028] Figure 7 is a schematic diagram of routes of multiple target clusters corresponding to a target road segment in an embodiment of the present invention;

[0029] Figure 8 is another schematic diagram of routes of multiple target clusters corresponding to a target road segment in an embodiment of the present invention;

[0030] Figure 9 This is a schematic diagram of a process for constructing a driving area map based on a target trajectory dataset in an embodiment of the present invention;

[0031] Figure 10 is a schematic diagram of a driving area map in an embodiment of the present invention;

[0032] Figure 11 is another schematic diagram of a driving area map in an embodiment of the present invention;

[0033] Figure 12 1 is a schematic diagram comparing a driving area map and a standard lane map in an embodiment of the present invention;

[0034] Figure 13 This is a schematic diagram of a scenario of searching for potential interactive objects in a driving area map according to an embodiment of the present invention;

[0035] Figure 14 This is a schematic diagram of a scenario of searching for potential interactive objects in a standard lane area according to an embodiment of the present invention;

[0036] Figure 15 is a classification diagram for classifying interactive behavior scenarios in an embodiment of the present invention;

[0037] Figure 16 is a schematic diagram of the classification results of the interactive behavior scenarios in an embodiment of the present invention;

[0038] Figure 17 This is a schematic diagram of a process for constructing an interactive behavior scenario library in an embodiment of the present invention;

[0039] Figure 18 This is a schematic block diagram of the structure of a computer device provided by an embodiment of the present invention;

[0040] Figure 19 This is a schematic block diagram of the structure of an interactive behavior extraction system based on a driving area map provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0041] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0042] The flowcharts shown in the accompanying drawings are for illustrative purposes only and do not necessarily include all contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may be decomposed, combined, or partially merged, so the actual execution order may vary depending on the actual situation.

[0043] It should be understood that the terms used in this specification are only for the purpose of describing particular embodiments and are not intended to limit the present invention. As used in the specification and appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0044] Interaction scenarios between traffic participants are essential data for studying and analyzing road safety, as well as for developing and testing high-level autonomous driving. Currently, lane area maps are primarily used to extract these scenarios. However, these maps are only suitable for extracting interactions between traffic participants on structured roads with clear lane markings and well-ordered traffic. They are not suitable for extracting interactions between traffic participants on unstructured roads with blurred or even absent lane markings, or on unstructured roads where vehicles do not strictly follow lane markings. Therefore, lane area maps cannot accurately extract interactions between traffic participants on unstructured roads.

[0045] For example, the lane areas in the lane area map are divided based on lane markings. However, the sections with mixed traffic have irregular traffic characteristics and are characterized by narrow and crowded roads. At the same time, there are dense vendors, serious illegal parking, and poor supervision. As a result, most motor vehicles do not strictly follow the lane markings on the sections with mixed traffic. There are a large number of driving behaviors such as driving on the lanes and occupying opposite lanes, which makes the actual driving area of ​​the motor vehicle deviate from the lane area corresponding to the lane markings. If the lane area map is used to extract the interactive behavior scenarios between traffic participants on the sections with mixed traffic, the accuracy of the extracted interactive behavior scenarios cannot be guaranteed.

[0046] In order to solve the above problems, a method, system, device and storage medium for extracting interactive behaviors based on a driving area map are provided. The driving area map in the embodiment of the present invention is constructed based on a target trajectory data set, so that no matter the target road section is a structured road section with clear lane markings and good traffic order, or an unstructured road section with blurred lane markings or even no lane markings, or an unstructured road section where motor vehicles do not strictly follow lane markings, the driving area map can accurately represent the drivable area under different routes of the target road section, thereby ensuring the accuracy of the drivable area. In addition, when searching for potential interactive objects, it is not necessary to search on the entire target road section, but only to search in the driving area map, thereby reducing the search range of potential interactive objects, which not only improves the search efficiency, but also ensures the accuracy of the potential interactive objects searched, and then further screens them according to the value of the interaction risk measurement index between the target motor vehicle and each potential interactive object. The target interactive object is selected, and the trajectory data of the target motor vehicle and the corresponding trajectory data of the target interactive object are extracted as the interactive behavior scene, thereby improving the accuracy of the interactive behavior extraction. Therefore, whether it is a structured road section with clear lane markings and good traffic order, or an unstructured road section with blurred lane markings or even no lane markings, or an unstructured road section where motor vehicles do not strictly follow the lane markings, the interactive behavior scene can be fully and accurately extracted using the driving area map, thereby improving the universality, comprehensiveness and accuracy of the interactive behavior scene extraction. Moreover, the driving area map can not only represent the drivable areas under different routes, but also map each trajectory in the trajectory dataset used to construct the driving area map to the corresponding route, so that the routes of the motor vehicle participants in the interactive behavior scene extracted based on the driving area map and the trajectory data of the traffic participants are known and the purpose is clear, thereby improving the interpretability of the interactive behavior scene.

[0047] In some embodiments, the driving area map is used to represent the drivable areas of multiple target motor vehicles on the target road section, specifically refers to the safe area where the target motor vehicles can pass on the target road section. The driving area map can be obtained by manual drawing or constructed based on clustering of traffic participant trajectory data sets. Compared with the lane area in the lane area map in the standard format, the drivable area represented by the driving area map is not limited to lane boundaries and lane markings. Even when there are no lane boundaries and lane markings, the drivable area represented by the driving area map can accurately represent the safe area where the target motor vehicle can pass on the target road section.

[0048] In some embodiments, traffic participants refer to the common users of the target road section in traffic. Traffic participants can include two major categories: motor vehicles and non-motor vehicles, and can also be further divided into cars, trucks, vans, pedestrians, electric vehicles, etc. Each category has different driving characteristics. Interaction objects refer to traffic participants that have an impact on their own decisions, including motor vehicle participants (target motor vehicles) and other traffic participants (non-target motor vehicles). The interactive behavior scenario defined in the embodiment of the present invention limits two interaction objects, namely, one motor vehicle participant and one other traffic participant. The types of motor vehicle participants can include three categories: cars, vans, and buses. Other traffic participants can include seven categories: cars, vans, buses, pedestrians, cyclists, electric motorcycles, and tricycles. Interaction behavior scenarios refer to behavioral data before and after the influence of mutual decision-making between traffic participants. Interaction behavior scenarios are important basic data in the fields of human driver behavior research, autonomous driving system anthropomorphic decision research, etc.

[0049] It is understandable that the types of motor vehicle participants are not limited to the three categories of cars, trucks, and buses, and the types of other traffic participants are not limited to the seven categories of cars, trucks, buses, pedestrians, cyclists, electric motorcycles, and tricycles, and can be set based on actual conditions. For example, the types of motor vehicle participants can also include the four categories of cars, RVs, trucks, and buses, and the types of other traffic participants can include the eight categories of cars, RVs, trucks, buses, pedestrians, cyclists, electric motorcycles, and tricycles. For another example, the types of motor vehicle participants can also include the five categories of cars, sport utility vehicles (SUVs), RVs, trucks, and buses, and the types of other traffic participants can include the nine categories of cars, SUVs, RVs, trucks, buses, pedestrians, cyclists, electric motorcycles, and tricycles.

[0050] In an embodiment of the present invention, the interactive behavior extraction method based on the driving area map can be applied to a computer device, which may include a terminal device or a server. The terminal device may include a smart phone, a tablet computer, a laptop computer, a desktop computer, a personal digital assistant and a wearable device, etc. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0051] The following embodiments of the present invention are described in detail with reference to the accompanying drawings. In the absence of conflict, the following embodiments and features therein may be combined with each other.

[0052] See also Figure 1 , Figure 1 This is a flow chart of a method for extracting interactive behaviors based on a driving area map provided by an embodiment of the present invention.

[0053] like Figure 1 As shown, the interaction behavior extraction method includes steps S101 to S105.

[0054] Step S101: Acquire a traffic participant trajectory dataset of a target road section, and acquire a target trajectory dataset from the traffic participant trajectory dataset. The target trajectory dataset includes trajectory data of a plurality of target motor vehicles with clear driving routes.

[0055] In this embodiment, the target road section may include a structured road section with clear lane markings and good traffic order, for example, a highway, expressway, or intersection. The target road section may also include an unstructured road section with blurred or even no lane markings, such as a snow-covered road section, a road with severely worn lane markings, or a country road. The target road section may also include a road section where motor vehicles do not strictly follow lane markings, such as an urban road section with mixed pedestrian and vehicle traffic, or a road section where the lane area is occupied by illegally parked vehicles or obstacles.

[0056] In some embodiments, the traffic participant trajectory data set includes a plurality of traffic participant trajectory data, where the traffic participant trajectory data refers to the trajectory data of the traffic participant, and the trajectory data of the traffic participant may include the location information, length and width, orientation angle and speed information of the traffic participant at each collection moment, and the location information includes the location coordinates of the center point of the traffic participant and the location coordinates of the four corner points of the directional detection frame of the traffic participant. Among them, traffic participants may include cars, trucks, buses, pedestrians, cyclists, electric motorcycles and tricycles. Alternatively, traffic participants may also include cars, trucks, buses, tricycles (including electric tricycles or gasoline tricycles), motorcycles (including electric motorcycles or gasoline motorcycles), cyclists or pedestrians, etc.

[0057] In some embodiments, as Figure 2 As shown, step S101 may include sub-steps S1011 to S1012.

[0058] Sub-step S1011: Acquire a target bird's-eye view image sequence corresponding to the traffic flow of the target road section.

[0059] In this embodiment, the target bird's-eye view image sequence is obtained by capturing the traffic flow of the target road section from a bird's-eye view using an image acquisition device mounted on an altitude maintenance device deployed on the target road section. In response to the angle between the shooting optical axis of the image acquisition device and the road surface of the target road section being within a preset angle range, the image acquisition perspective of the image acquisition device is determined to be a bird's-eye view. The preset angle range can be set by the user or by default. For example, the preset angle range includes [60°, 90°]. The altitude maintenance device can include an unmanned aerial vehicle, a bracket, or a streetlight, and the image acquisition device can include a digital camera, a single-lens reflex camera, an infrared camera, or a depth camera.

[0060] In some embodiments, obtaining a target bird's-eye view image sequence corresponding to the traffic flow of a target road section may include: obtaining an original bird's-eye view image sequence obtained by an image acquisition device capturing the traffic flow of the target road section from a bird's-eye view perspective; and performing image stabilization processing on the original bird's-eye view image sequence to obtain the target bird's-eye view image sequence. This embodiment, by performing image stabilization processing on the original bird's-eye view image sequence captured by the image acquisition device, eliminates video jitter caused by vibration of the image acquisition device itself and environmental influences, thereby improving the accuracy of the subsequent acquisition of the traffic participant trajectory dataset.

[0061] In some embodiments, obtaining a target bird's-eye view image sequence corresponding to the traffic flow of a target road section may include: obtaining multiple original bird's-eye view image sequences obtained by an image acquisition device capturing the traffic flow of the target road section from a bird's-eye view perspective; performing image stabilization processing on each of the multiple original bird's-eye view image sequences, and aligning the multiple original bird's-eye view image sequences after image stabilization processing to the same image coordinate system to obtain multiple target bird's-eye view image sequences. This embodiment eliminates video jitter caused by the vibration of the image acquisition device itself and environmental influences by performing image stabilization processing on the original bird's-eye view image sequences, thereby improving the accuracy of subsequent acquisition of traffic participant trajectory data sets. By aligning the multiple original bird's-eye view image sequences after image stabilization processing to the same image coordinate system, it is possible to eliminate image field deviations caused by position deviations of multiple task acquisitions, so that all trajectory data subsequently acquired from the same point can share the same coordinate system, and only one bird's-eye view image at a point needs to be mapped to meet the needs of all trajectory data for map use, thereby significantly improving the processing efficiency of map construction.

[0062] In some embodiments, multiple original bird's-eye view image sequences may include original bird's-eye view image sequences collected by the same image acquisition device in different time periods and / or original bird's-eye view image sequences collected by different image acquisition devices in the same or different time periods. For example, during the morning and evening rush hours, the first unmanned aerial vehicle is controlled to hover above the target road section at a set altitude (for example, 80 meters), and the image acquisition device carried on it is controlled to collect data on the traffic flow from a bird's-eye view perspective (the angle between the shooting optical axis of the image acquisition device and the road surface where the target road section is located is 90°), and the five original bird's-eye view image sequences shown in Table 1 are obtained, including the original bird's-eye view image sequence collected by the image acquisition device carried by the first unmanned aerial vehicle from 8:50 to 9:50 on July 19, 2024, ... The original bird's-eye view image sequence is collected by the device between 3:00 PM and 4:30 PM on July 19, 2024; the original bird's-eye view image sequence is collected by the image acquisition device on the first unmanned aerial vehicle between 10:20 PM and 10:50 PM on July 19, 2024; the original bird's-eye view image sequence is collected by the image acquisition device on the first unmanned aerial vehicle between 9:30 AM and 10:00 AM on September 19, 2024; and the original bird's-eye view image sequence is collected by the image acquisition device on the first unmanned aerial vehicle between 3:00 PM and 4:40 PM on September 19, 2024. Stabilization is performed on each of these five original bird's-eye view image sequences, and the five stabilized original bird's-eye view image sequences are aligned to the same coordinate system to obtain five target bird's-eye view image sequences.

[0063] Table 1

[0064]

[0065] For another example, during the morning and evening rush hours, the first unmanned aerial vehicle is controlled to hover at a set altitude (for example, 80 meters) above the first position of the target road section, and the second unmanned aerial vehicle is controlled to hover at a set altitude (for example, 80 meters) above the second position of the target road section. Then, the image acquisition device carried by the first unmanned aerial vehicle is controlled to collect traffic flow data from a bird's-eye view (the angle between the shooting optical axis of the image acquisition device and the road surface where the target road section is located is 90°). At the same time, the image acquisition device carried by the second unmanned aerial vehicle is controlled to collect traffic flow data from a bird's-eye view (the angle between the shooting optical axis of the image acquisition device and the road surface where the target road section is located is 90°). Five original bird's-eye view image sequences are obtained as shown in Table 1, and a total of 10 original bird's-eye view image sequences are obtained. The bird's-eye view image sequences collected by the image acquisition device carried by the first unmanned aerial vehicle above the first position of the target road section include: Figure 3 The bird's-eye view image sequence shown in FIG. 1 includes the following: Figure 4 The bird's-eye view image shown in FIG. 10 is a stabilization process performed on each of the 10 original bird's-eye view image sequences, and the 10 stabilized original bird's-eye view image sequences are aligned to the same coordinate system to obtain 10 target bird's-eye view image sequences.

[0066] The above 10 original bird's-eye view image sequences include the original bird's-eye view image sequences collected by the image acquisition device carried by the first unmanned aerial vehicle and the image acquisition device carried by the second unmanned aerial vehicle respectively from 8:50 to 9:50 on July 19, 2024, the original bird's-eye view image sequences collected by the image acquisition device carried by the first unmanned aerial vehicle and the image acquisition device carried by the second unmanned aerial vehicle respectively from 15:00 to 16:30 on July 19, 2024, and the original bird's-eye view image sequences collected by the image acquisition device carried by the first unmanned aerial vehicle and the image acquisition device carried by the second unmanned aerial vehicle respectively from 15:00 to 16:30 on July 19, 2024. The original bird's-eye image sequence was collected by the image acquisition device carried by the first unmanned aerial vehicle from 22:20 to 22:50 on July 19, 2024, the original bird's-eye image sequence was collected by the image acquisition device carried by the first unmanned aerial vehicle and the image acquisition device carried by the second unmanned aerial vehicle from 9:30 to 10:00 on September 19, 2024, and the original bird's-eye image sequence was collected by the image acquisition device carried by the first unmanned aerial vehicle and the image acquisition device carried by the second unmanned aerial vehicle from 15:00 to 16:40 on September 19, 2024.

[0067] When an unmanned aerial vehicle (UAV) collects bird's-eye view image sequences, the image sequences can experience jitter due to environmental factors such as UAV vibration and strong winds. This can cause rotation and translation in the preceding and following image frames, leading to positional errors in the subsequent trajectory acquisition based on the image coordinate system. Therefore, by performing image stabilization on each of the 10 original bird's-eye view image sequences, this jitter caused by environmental factors such as UAV vibration and strong winds is eliminated, thereby improving the accuracy of the subsequent traffic participant trajectory dataset. Due to the limitations of the UAV's battery life, during the execution of the bird's-eye view image sequence collection task, the UAV needs to be controlled to land and replace the battery. After replacing the battery, it returns to the sky above the target road section to continue collecting the bird's-eye view image sequence. Under the influence of reciprocating takeoff and landing, the points collected by the UAV each time cannot be consistent, resulting in field of view deviations between different original bird's-eye view image sequences. The field of view deviations between different original bird's-eye view image sequences are mainly caused by changes in collection height, direction and position, as well as perspective deviations of surrounding buildings. Therefore, by aligning the above 10 original bird's-eye view image sequences to a unified image coordinate system, the field of view deviations between different original bird's-eye view image sequences can be eliminated, thereby significantly improving the processing efficiency of subsequent map construction and trajectory alignment.

[0068] Sub-step S1012: acquiring a traffic participant trajectory dataset of the target road section according to the target bird's-eye view image sequence.

[0069] In this embodiment, the traffic participant trajectory data includes a target static data sequence and a speed data sequence of the traffic participant. The speed data sequence of the traffic participant may include a speed curve of the traffic participant. The target static data sequence of the traffic participant includes the position information, length, width and orientation angle of each bird's-eye view image of the traffic participant in the target bird's-eye view image sequence at the time of acquisition. The position information includes the position coordinates of the center point of the traffic participant and the position coordinates of the four corner points of the directional detection frame of the traffic participant. For example, Figure 5 As shown, the i-th target static data in the target static data sequence of the traffic participant 110 includes the length L, width W, orientation angle α of the traffic participant 110, and position coordinates (x i, y i ), the position coordinates (x ) of the four corner points of the directional detection frame 120 of the traffic participant 110 1i, y 1i ), (x 2i, y 2i ), (x 3i, y 3i ), (x 4i, y4i ). i is an integer greater than or equal to 1.

[0070] In some embodiments, a traffic participant trajectory dataset of a target road section is obtained based on a target bird's-eye image sequence, including: calling a preset traffic participant detection model to perform traffic participant detection on each bird's-eye image in the target bird's-eye image sequence to obtain a traffic participant detection result for each bird's-eye image; determining static data of each traffic participant in each bird's-eye image based on the traffic participant detection result of each bird's-eye image; performing multi-target tracking on the traffic participants in each bird's-eye image based on the static data of each traffic participant in each bird's-eye image to assign a tracking ID to each successfully tracked traffic participant; correlating the static data of each traffic participant in each bird's-eye image based on the tracking ID of each traffic participant to obtain a static data sequence corresponding to each tracking ID; correcting the static data in the static data sequence corresponding to each tracking ID to obtain a target static data sequence corresponding to each tracking ID; determining a speed data sequence of the traffic participant corresponding to each tracking ID based on the target static data sequence corresponding to each tracking ID; using the target static data sequence and speed data sequence corresponding to the same tracking ID as the trajectory data of the corresponding traffic participant, and aggregating the trajectory data of each traffic participant to obtain a traffic participant trajectory dataset of the target road section. This embodiment corrects the static data in the static data sequence corresponding to each tracking ID to eliminate perspective errors, so that the corrected target static data can be closer to the actual situation, further improving the accuracy of the traffic participant trajectory dataset.

[0071] In some embodiments, the traffic participant detection results of the bird's-eye view image may include the directional detection box, orientation angle, and traffic participant category of each traffic participant in the bird's-eye view image. Traffic participant categories may include pedestrians, bicycles, electric motorcycles, electric tricycles, cars, trucks, and buses. It should be noted that traffic participant categories may also include cars, trucks, buses, tricycles (including electric tricycles or gasoline tricycles), motorcycles (including electric motorcycles or gasoline motorcycles), bicycles, or pedestrians.

[0072] In some embodiments, the preset traffic participant detection model is obtained by iteratively training the target detection model in advance based on a training sample data set, where the sample data in the training sample set includes a sample image and an annotated directional detection box, an annotated orientation angle, and an annotated traffic participant category. The target detection model may include a YOLO model, an SSD (Single Shot MultiBox Detector) model, or a Faster R-CNN model, and the YOLO model may include a YOLOv11 model. For example, the YOLOv11 model may include five different versions: YOLOv11n-obb, YOLOv11m-obb, YOLOv11l-obb, YOLOv11s-obb, and YOLOv11x-obb.

[0073] For example, the preset traffic participant detection model is obtained by iteratively training the YOLOv11m-obb model based on the training sample data set in advance to ensure the detection accuracy of the traffic participant detection model. For example, the process of iteratively training the target detection model based on the training sample data set may include: obtaining a training sample from the training sample data set as a target training sample; inputting the sample image in the target training sample into the YOLOv11m-obb model to obtain the predicted directional detection frame, the predicted orientation angle, and the predicted probability that the traffic participant in the sample image belongs to each traffic participant category; determining the intersection-over-union ratio between the predicted directional detection frame and the labeled directional detection frame in the target training sample, calculating the cross entropy loss value based on the labeled traffic participant category in the target training sample and the predicted probability that the traffic participant in the sample image belongs to each traffic participant category, and calculating the distribution focus loss value based on the predicted orientation angle and the labeled orientation angle in the target training sample; when the intersection-over-union ratio is less than a preset intersection-over-union ratio threshold , update the parameters of the YOLOv11m-obb model that affect the intersection-union ratio according to the intersection-union ratio, when the cross-entropy loss value is less than the first loss value threshold, update the parameters of the YOLOv11m-obb model that affect the cross-entropy loss value according to the cross-entropy loss value, when the distribution focus loss value is less than the second loss value threshold, update the parameters of the YOLOv11m-obb model that affect the distribution focus loss value according to the distribution focus loss value, and then return to execute the step of obtaining a training sample from the training sample dataset as the target training sample; when the intersection-union ratio is greater than or equal to the preset intersection-union ratio threshold, the cross-entropy loss value is greater than or equal to the first loss value threshold, and the distribution focus loss value is greater than or equal to the second loss value threshold, stop iterative training of the YOLOv11m-obb model to obtain a traffic participant detection model.

[0074] In some embodiments, the traffic participant detection results of a bird's-eye view image may include the position coordinates, orientation angle and traffic participant category of the four corner points of the directional detection box of each traffic participant in the bird's-eye view image. Based on the traffic participant detection results of each bird's-eye view image, determining the static data of each traffic participant in each bird's-eye view image may include: for each traffic participant, based on the position coordinates of the four corner points of the directional detection box of the traffic participant, calculating the position coordinates of the center point of the traffic participant and the length and width of the traffic participant, and using the position coordinates of the center point of the traffic participant, the position coordinates of the four corner points of the directional detection box of the traffic participant, the orientation angle of the traffic participant and the length and width of the traffic participant as the static data of the traffic participant.

[0075] In some embodiments, performing multi-target tracking on the traffic participants in each bird's-eye view image based on the static data of each traffic participant in each bird's-eye view image to assign a tracking ID to each successfully tracked traffic participant may include: utilizing a multi-target tracking algorithm to perform multi-target tracking on the traffic participants in each bird's-eye view image based on the static data of each traffic participant in each bird's-eye view image to assign a tracking ID to each successfully tracked traffic participant in each bird's-eye view image. The multi-target tracking algorithm may include a Kalman filter-based multi-target tracking algorithm, a Hungarian algorithm, or a deep learning-based multi-target tracking algorithm, and the deep learning-based multi-target tracking algorithm may include a ByteTrack algorithm or a TransTrack algorithm.

[0076] In some embodiments, the static data of each traffic participant in each bird's-eye view image can be saved according to the format of Table 2. The meanings of the parameters in Table 2 are: the tracking id in the i-th frame bird's-eye view image is id i The position coordinates of the center point of the traffic participant are (x i ,y i ), the length and width of the oriented detection box are (w i , h i ) and the rotation angle of the directional detection frame around the positive direction of the x-axis (the direction of the traffic participant) is r i .

[0077] Table 2

[0078] frame id x y width height rotation i <![CDATA[id i ]]> <![CDATA[x i ]]> <![CDATA[y i ]]> <![CDATA[w i ]]> <![CDATA[h i ]]> <![CDATA[r i ]]>

[0079] In some embodiments, an improved L-shape algorithm can be used to correct the static data in the static data sequence corresponding to each tracking ID. Correcting the static data in the static data sequence corresponding to the tracking ID to obtain a target static data sequence corresponding to the tracking ID includes: obtaining a directional detection frame closest to the minimum perspective distortion area from the directional detection frames contained in each static data in the static data sequence as a reference detection frame, and determining the size of the reference detection frame as a correction reference size; determining the directional detection frames other than the reference detection frame in the directional detection frames contained in each static data in the static data sequence as detection frames to be corrected; for each detection frame to be corrected, determining the corner point closest to the reference detection frame among the four corner points of the detection frame to be corrected as a correction reference corner point; and correcting the position and shape of each detection frame to be corrected in the static data sequence according to the correction reference corner point and correction reference size of each detection frame to be corrected to obtain a target static data sequence. This embodiment corrects the position and shape of each detection frame to be corrected in the static data sequence, thereby eliminating perspective errors so that the corrected target static data can be closer to the actual situation, further improving the accuracy of the traffic participant trajectory dataset.

[0080] For example, see Figure 6 , Figure 6 FIG. 1 is a schematic diagram of a process for correcting the position and shape of a detection frame to be corrected according to an embodiment of the present invention. Figure 6 The bird's-eye view image shown in a includes a detection frame 11 of a traffic participant to be corrected. Figure 6 The bird's-eye view image shown in b includes an L-shape correction reference 12 based on the correction reference size and the correction reference corner points of the detection frame to be corrected 11. The position and shape of the detection frame to be corrected 11 are corrected according to the L-shape correction reference 12, and the following can be obtained: Figure 6 c shows the corrected directional detection frame 13 , which is shorter than the detection frame 11 to be corrected.

[0081] In some embodiments, when there are multiple target bird's-eye view image sequences, for each target bird's-eye view image sequence, a preset traffic participant detection model is called to perform traffic participant detection on each bird's-eye view image in the target bird's-eye view image sequence to obtain traffic participant detection results for each bird's-eye view image in each target bird's-eye view image sequence; based on the traffic participant detection results for each bird's-eye view image in each target bird's-eye view image sequence, static data of each traffic participant in each bird's-eye view image in each target bird's-eye view image sequence is determined; based on the static data of each traffic participant in each bird's-eye view image in each target bird's-eye view image sequence, multiple traffic participant detection results are performed on each traffic participant in each bird's-eye view image in each target bird's-eye view image sequence. Target tracking is to assign a tracking ID to each successfully tracked traffic participant; based on the tracking ID of each successfully tracked traffic participant, the static data of each traffic participant in each bird's-eye view image in each target bird's-eye view image sequence are associated to obtain a static data sequence corresponding to each tracking ID; the static data in the static data sequence corresponding to each tracking ID is corrected to obtain a target static data sequence corresponding to each tracking ID; based on the target static data sequence corresponding to each tracking ID, the speed data sequence of the traffic participant corresponding to each tracking ID is determined; the target static data sequence and speed data sequence corresponding to the same tracking ID are used as the trajectory data of the corresponding traffic participant.

[0082] In some embodiments, determining a speed data sequence of a traffic participant based on a target static data sequence corresponding to a tracking ID may include: for any target static data collection moment in the target static data sequence, obtaining a position information sequence at that collection moment from the target static data sequence based on a preset window length, wherein the length of the position information sequence is the same as the preset window length; calculating the instantaneous speed of the traffic participant at that collection moment based on the position information sequence to obtain the instantaneous speeds of the traffic participant at multiple collection moments; smoothing the instantaneous speeds of the traffic participant at the multiple collection moments to obtain the target instantaneous speeds of the traffic participant at the multiple collection moments; and sorting the target instantaneous speeds of the traffic participant at the multiple collection moments in chronological order to obtain the speed data sequence of the traffic participant. The instantaneous speed of the traffic participant may include the instantaneous lateral speed and the instantaneous longitudinal speed of the traffic participant. The preset window length may be set by the user or a default value, for example, 5. The instantaneous speeds of the traffic participant at the multiple collection moments may be smoothed based on a Rauch-Tung-Striebel (RTS) smoothing algorithm. This embodiment calculates the instantaneous speed of traffic participants in a sliding window manner, effectively suppressing a large amount of instantaneous noise. In addition, by smoothing the instantaneous speed, the speed data obtained after smoothing is closer to the actual speed data, thereby improving the accuracy of the speed data of traffic participants.

[0083] For example, based on the position information sequence, the process of calculating the instantaneous speed of the traffic participant at the collection moment can be achieved by the formula Indicates that, v t is the instantaneous speed of the traffic participant at the sampling time t, N is the preset window length, N is an odd number, Δt is the sampling period (can be set by the user or use the default value, for example, Δt is 1 / 30s), i is an integer greater than or equal to 1 and less than or equal to (N-1) / 2, p t is the position information sequence of traffic participants at the collection time t, p t+i The position information sequence of traffic participants at the collection time t+i, p t-i The position information sequence of traffic participants at the collection time ti.

[0084] In some embodiments, obtaining a target trajectory dataset from a traffic participant trajectory dataset may include: filtering out motor vehicle trajectory data having a trajectory duration greater than or equal to a target duration threshold from the traffic participant trajectory dataset to obtain a first motor vehicle trajectory dataset; performing standardization processing on each motor vehicle trajectory data in the first motor vehicle trajectory dataset to obtain a second motor vehicle trajectory dataset; performing clustering processing on the second motor vehicle trajectory dataset to obtain a cluster of target centroids, where the cluster is a subset of the second motor vehicle trajectory dataset; and filtering the cluster of target centroids based on lane information of a target road section to obtain at least one target cluster, where one target cluster corresponds to one target trajectory dataset. This embodiment filters out motor vehicle trajectory data with trajectory duration greater than or equal to a target duration threshold from the traffic participant trajectory dataset, thereby eliminating irregular trajectory data of pedestrians and non-motor vehicles, as well as trajectory data of motor vehicles that have been stationary for a long time and have incomplete trajectories, to improve the accuracy of the motor vehicle trajectory dataset and reduce the amount of data during clustering, thereby improving clustering efficiency and accuracy. Moreover, based on the lane information of the target road section, target cluster clusters corresponding to each route of at least one route of the target road section can be screened from the cluster clusters of the target number of centroids, so that the final target trajectory dataset is more consistent with the corresponding actual route of the target road section.

[0085] In some embodiments, the second vehicle trajectory dataset may be clustered based on a preset clustering algorithm to obtain clusters of target centroids. The preset clustering algorithm may be user-configured and is not specifically limited in this embodiment of the present invention. For example, the preset clustering algorithm includes a traditional K-means algorithm or a K-means++ algorithm.

[0086] Exemplarily, clustering the second motor vehicle trajectory data set to obtain clusters with a target number of centroids may include: randomly selecting a motor vehicle trajectory data from the second motor vehicle trajectory data set as the cluster centroid; determining the minimum distance between each motor vehicle trajectory data other than the cluster centroid and the cluster centroid in the second motor vehicle trajectory data set; determining the probability that each motor vehicle trajectory data other than the cluster centroid in the second motor vehicle trajectory data set is a new cluster centroid based on the minimum distance between each motor vehicle trajectory data other than the cluster centroid in the second motor vehicle trajectory data set and the cluster centroid, and using the motor vehicle trajectory data corresponding to the maximum probability as the new cluster centroid; returning to the step of determining the minimum distance between each motor vehicle trajectory data other than the cluster centroid in the second motor vehicle trajectory data set and the cluster centroid until the number of determined cluster centroids reaches the target number; for each motor vehicle trajectory data other than all cluster centroids in the second motor vehicle trajectory data set, For each motor vehicle trajectory data outside the initial cluster, determine the distance between the motor vehicle trajectory data and each cluster centroid, assign the motor vehicle trajectory data to the cluster centroid corresponding to the minimum distance, and obtain an initial cluster with a target number of centroids; for each initial cluster, determine the sum of the distances between each motor vehicle trajectory data in the initial cluster and the other motor vehicle trajectory data in the initial cluster; use the motor vehicle trajectory data corresponding to the minimum distance sum as the new cluster centroid of the initial cluster; determine the distance between the new cluster centroid of the initial cluster and the previous cluster centroid, and when the distance between the new cluster centroid of the initial cluster and the previous cluster centroid is greater than a preset distance threshold, return to the step of determining the sum of the distances between each motor vehicle trajectory data in the initial cluster and the other motor vehicle trajectory data in the initial cluster, until the distance between the new cluster centroid of the initial cluster and the previous cluster centroid is less than or equal to the preset distance threshold, and obtain a cluster with a target number of centroids.

[0087] In some embodiments, normalizing each vehicle trajectory data in the first vehicle trajectory dataset to obtain the second vehicle trajectory dataset may include: normalizing the trajectory lengths of each vehicle trajectory data in the first vehicle trajectory dataset to obtain a candidate vehicle trajectory dataset; and performing mean-variance normalization on each vehicle trajectory data in the candidate vehicle trajectory dataset to obtain the second vehicle trajectory dataset. Normalizing the trajectory lengths of each vehicle trajectory data in the first vehicle trajectory dataset to obtain the candidate vehicle trajectory dataset may include: determining a standard trajectory length based on the trajectory lengths of each vehicle trajectory data in the first vehicle trajectory dataset; downsampling each vehicle trajectory data in the first vehicle trajectory dataset, in response to the trajectory length of the vehicle trajectory data being greater than the standard trajectory length, shortening the trajectory length of the vehicle trajectory data so that the trajectory length of the vehicle trajectory data reaches the standard trajectory length; and performing linear interpolation on the vehicle trajectory data, in response to the trajectory length of the vehicle trajectory data being less than the standard trajectory length, lengthening the trajectory length of the vehicle trajectory data so that the trajectory length of the vehicle trajectory data reaches the standard trajectory length.

[0088] In some embodiments, the interactive behavior extraction method provided by the present invention further includes: clustering the second motor vehicle trajectory data set in sequence based on each number of centroids within a preset range of centroids, and determining the intra-cluster sum of squared errors (SSE) corresponding to each number of centroids based on the clustering results; generating an SSE change trend graph based on the SSE corresponding to each number of centroids, the SSE change trend graph being used to describe the change trend of SSE as the number of centroids increases; determining the SSE change rate corresponding to each number of centroids based on the SSE change trend graph, and determining the number of centroids with an SSE change rate less than the preset change rate as the candidate number of centroids; and determining the target number of centroids based on multiple candidate numbers of centroids. The preset range of centroids and the preset change rate can be set by the user, and the embodiment of the present invention does not specifically limit this. For example, the preset centroid data set includes [1,30]. This embodiment can adaptively determine the target number of centroids based on the motor vehicle trajectory data set, thereby improving the accuracy of subsequent clustering of the motor vehicle trajectory data set.

[0089] In some embodiments, determining the target number of centroids based on multiple candidate number of centroids may include: sorting the multiple candidate number of centroids in descending order to obtain a centroid number sequence, determining the middle candidate number of centroids in the centroid number sequence as the target number of centroids, or determining the first candidate number of centroids in the centroid number sequence as the target number of centroids, or determining the second candidate number of centroids in the centroid number sequence as the target number of centroids.

[0090] In some embodiments, based on the lane information of the target road section, the target number of centroid clusters are screened to obtain at least one target cluster, which may include: determining multiple standard driving routes of motor vehicles on the target road section based on the lane information of the target road section; determining the motor vehicle driving routes corresponding to each cluster based on the motor vehicle trajectory data in each cluster; screening a unique cluster of motor vehicle driving routes from the target number of centroid clusters as a candidate cluster; screening at least one target cluster from the multiple candidate clusters to obtain at least one target trajectory data set, wherein the motor vehicle driving route corresponding to the target cluster is the same as any standard driving route. This embodiment can accurately screen the target number of centroid clusters to obtain a matching cluster corresponding to each route in at least one route of the target road section.

[0091] For example, from Figure 3 The vehicle trajectory data with a trajectory duration greater than or equal to the target duration threshold is filtered out from the traffic participant trajectory data set of the target road section shown in the figure. Then, the filtered vehicle trajectory data set is clustered to obtain 20 clusters. Figure 3 The lane information of the target road section shown can be filtered from these 20 clusters as follows Figure 7 The routes corresponding to the six target clusters shown include a right-hand straight route along the main road, a left-hand straight route along the main road, a route turning right from the branch road to the main road, a route turning left from the main road to the branch road, a route turning left from the branch road to the main road, and a route turning right from the main road to the branch road. Therefore, a target trajectory dataset corresponding to each of the six routes is finally obtained.

[0092] For example, from Figure 4 The vehicle trajectory data sets of the traffic participants on the target road section shown in the figure are screened out, and the vehicle trajectory data sets with a trajectory duration greater than or equal to the target duration threshold are then clustered to obtain 8 clusters. Figure 4 The lane information of the target road section shown can be obtained from these 8 clusters as follows Figure 8 The routes corresponding to the two target clusters shown include a left-going straight route along the main road and a right-going straight route along the main road. Therefore, a target trajectory dataset corresponding to each of the two routes is finally obtained.

[0093] In some embodiments, after step S101, the method further includes: generating a vehicle trajectory image for a route corresponding to the target road segment based on the target trajectory dataset, extracting a driving area outline for the route from the vehicle trajectory image; smoothing the driving area outline to obtain a target driving area outline, and filling the driving area contained in the target driving area outline to obtain a driving area map corresponding to the route. If there are multiple target trajectory datasets, driving area maps for multiple routes of the target road segment can be constructed based on the multiple target trajectory datasets, with each target trajectory dataset corresponding to a driving area map. The driving area outline can be smoothed using a Gaussian filter. By converting the target trajectory dataset into a vehicle trajectory image before constructing the driving area map, this embodiment does not rely on lane markings on the target road segment, nor does it require vehicles to strictly follow lane markings on the target road segment. This allows accurate construction of a driving area map for the corresponding route, regardless of whether the target road segment is a structured section with clear lane markings and good traffic order, an unstructured section with blurred or even no lane markings, or an unstructured section where vehicles do not strictly follow lane markings, effectively improving the accuracy and universality of driving area map construction.

[0094] In some embodiments, generating a motor vehicle trajectory image for a corresponding route of a target road segment based on a target trajectory dataset may include: plotting each target motor vehicle trajectory data in the target trajectory dataset into a blank image to obtain a motor vehicle trajectory image for the corresponding route of the target road segment. In some embodiments, extracting a driving area contour of the route from the motor vehicle trajectory image may include: binarizing the motor vehicle trajectory image to obtain a binarized image, performing image erosion on the binarized image to obtain a candidate binarized image; detecting outliers in the candidate binarized image using an outlier detection algorithm, and deleting outliers from the candidate binarized image to obtain a target binarized image; and performing image dilation on the target binarized image to obtain a driving area contour of the route. This embodiment, through image erosion and outlier processing, can effectively remove noise from the trajectory and improve the accuracy of the extracted driving area contour.

[0095] In some embodiments, the outlier detection algorithm may include a DBSCAN algorithm. Using the outlier detection algorithm, detecting outliers in a candidate binary image may include obtaining a neighborhood radius and a minimum number of points, and executing the DBSCAN algorithm on the candidate binary image based on the neighborhood radius and the minimum number of points to obtain outliers in the candidate binary image. The neighborhood radius and the minimum number of points may be user-configurable and are not specifically limited in the present embodiment. For example, the neighborhood radius may be 10 pixels and the minimum number of points may be 6.

[0096] In some embodiments, the size of the structuring element used in the image dilation process for the target binary image is determined based on the average width of the vehicles in the target trajectory dataset and the size of the pixels contracted due to image erosion. For example, the size of the structuring element used in the image dilation process for the target binary image is determined to be half the average width of the vehicles in the target trajectory dataset and twice the sum of the size of the pixels contracted due to image erosion.

[0097] In some embodiments, filling the driving area included in the target driving area outline may include: taking each white pixel point on the target driving area outline as a starting point, and for each starting point, taking the up, down, left, and right sides of the starting point as expansion directions; for each expansion direction, performing a traversal search between the starting point and the boundary of the target driving area outline along the expansion direction, and if another white pixel point different from the starting point is found, marking the expansion direction, recording the starting point, and recording the other white pixel point different from the starting point as the end point; if another white pixel point different from the starting point is not found, determining whether the boundary of the target driving area outline has been searched; if the boundary of the target driving area outline has been searched, determining whether the traversal of the four expansion directions has been completed; if the traversal of the four expansion directions has been completed, determining whether the number of marked expansion directions is greater than 1, and if the number of marked expansion directions is greater than 1, filling all pixel points from each starting point along the marked expansion direction to the corresponding end point to obtain a target driving area map.

[0098] The following generates Figure 7 The process of constructing the driving area map is described by taking the driving area map corresponding to the route from the branch road to the main road as an example. Figure 7 Each target vehicle trajectory data in the target trajectory dataset corresponding to the route of turning left from the branch road to the main road is drawn in the blank image, and the Figure 9 The vehicle trajectory image corresponding to the route from the branch road to the main road is then Figure 7 The vehicle trajectory image corresponding to the route from the branch road to the main road is binarized and the Figure 9 a shows the binary image; Figure 9 The binary image shown in a is subjected to image corrosion processing to obtain Figure 9 b shows the candidate binary image; using the outlier detection algorithm, detect Figure 9 The outliers in the candidate binary image shown in b are Figure 9 Remove the outliers from the candidate binary image shown in b and get Figure 9 c shows the target binary image; Figure 9 The target binary image shown in c is subjected to image dilation processing to obtain Figure 9d shows the driving area outline (the driving area outline of the route from the branch road to the main road); Figure 9 The contour of the driving area shown in d is smoothed to obtain Figure 9 e shows the target driving area outline; Figure 9 Fill the driving area contained in the target driving area outline shown in e, and get Figure 9 The driving area map shown in f (driving area map for turning left from a branch road to a main road).

[0099] In the same way, we can construct Figure 7 The remaining routes in the driving area map, such as Figure 10 As shown, Figure 10 The driving area map a in Figure 7 The right-hand straight route along the main road corresponds to: Figure 10 The driving area map b in Figure 7 The left straight route along the main road corresponds to: Figure 10 The driving area map c in Figure 7 The route of turning right from the branch road to the main road corresponds to: Figure 10 The driving area map d in Figure 7 The route of turning left from the main road to the branch road corresponds to: Figure 10 Driving area map in e and Figure 7 The route of turning left from the branch road to the main road corresponds to: Figure 10 The driving area map f in Figure 7 In the same way, we can construct Figure 8 The driving area map of the route in Figure 11 As shown, Figure 11 The driving area map a in Figure 8 The left straight route along the main road corresponds to: Figure 11 The driving area map b in Figure 8 The route corresponds to the straight right route along the main road.

[0100] The driving area map constructed in the embodiment of the present invention is different from the standard lane area map. The driving area map is larger than the lane area map, which indirectly reflects the irregular driving behavior in the mixed traffic scene in the urban area. For specific differences, please refer to Figure 7 The area difference between the driving area map and the lane area map along the left straight route of the main road, such as Figure 12 As shown, Figure 7The boundary of driving area map 21 for the left-bound straight route along the main road is represented by a solid line, while the boundary of standard lane area map 22 is represented by a dashed line. Driving area map 21 essentially covers the area of ​​standard lane area map 22, and also partially covers the oncoming lane near the intersection. Preliminary analysis suggests that this phenomenon is partly due to the left-bound lane being close to the turning area of ​​the intersection. Most vehicles choose to use the oncoming lane to pass through the intersection to maintain a safe distance from vehicles turning from the branch road. Furthermore, pedestrians frequently cross the street near the intersection, so vehicles also choose to pass through the relatively unobstructed oncoming lane. Compared to standard lane area map 22, driving area map 21 accurately reflects the driving area for the corresponding route in mixed pedestrian and pedestrian traffic scenarios in urban areas. When searching for traffic participants who influence the decision-making of motor vehicles with a clear driving route, i.e., potential interaction objects, searching within the driving area can significantly reduce search time and computational cost.

[0101] Step S102 : searching for multiple potential interaction objects of the target motor vehicle in a driving area map corresponding to the driving route of the target motor vehicle based on the trajectory data of the target motor vehicle and the traffic participant trajectory dataset.

[0102] In this embodiment, potential interaction objects include traffic participants who are simultaneously located with the target vehicle in the driving area map corresponding to the target vehicle's route at the same time as the target vehicle at the same collection moment. The target vehicle's multiple potential interaction objects may include the target vehicle's potential interaction objects at each collection moment. It is understood that at one or some collection moments, the target vehicle's potential interaction objects may not exist in the driving area map corresponding to the target vehicle's route.

[0103] In some embodiments, the target motor vehicle's trajectory data includes a target static data sequence and a speed data sequence for the target motor vehicle. The target static data sequence includes the target static data of the target motor vehicle at each collection moment, and the speed data sequence includes the speed data of the target motor vehicle at each collection moment. A traffic participant trajectory dataset includes trajectory data of multiple traffic participants. The traffic participant's trajectory data includes a target static data sequence and a speed data sequence for the traffic participants. The target static data sequence includes the target static data of the traffic participants at each collection moment, and the speed data sequence includes the speed data of the traffic participants at each collection moment.

[0104] For example, Figure 13As shown, the target vehicle 30's driving route is to turn left from the main road into the branch road 31. The branch road 31 corresponds to the driving area map 40. At the acquisition time t, the target vehicle 30, the first traffic participant 301, and the second traffic participant 302 are all in the driving area map 40. Therefore, the first traffic participant 301 and the second traffic participant 302 can be determined as potential interaction objects of the target vehicle 30. When there is no known driving route, the search effect of the interaction object search based on the standard lane area 41 is as follows. Figure 14 As shown, only the first traffic participant 301 can be searched as a potential interaction object of the target motor vehicle 30, and the second traffic participant 302 cannot be searched as a potential interaction object of the target motor vehicle 30. In comparison, especially when the target motor vehicle 30 is turning, the interaction object search is performed based on the determined driving route and the known driving area map 40. The search range of the interaction object is closer to the driving route, and the accuracy of the interaction object search is higher.

[0105] In some embodiments, based on the target motor vehicle's trajectory data and the traffic participant trajectory data set, searching for multiple potential interaction objects of the target motor vehicle in a driving area map corresponding to the target motor vehicle's driving route may include: obtaining trajectory data of multiple candidate traffic participants from the traffic participant trajectory data set, where the multiple candidate traffic participants are located in the driving area map corresponding to the target motor vehicle's driving route; determining the position information and motion information of the target motor vehicle and each candidate traffic participant at each acquisition moment based on the target motor vehicle's trajectory data and the trajectory data of each candidate traffic participant; and determining multiple potential interaction objects of the target motor vehicle from the multiple candidate traffic participants based on the position information and motion information of the target motor vehicle and each candidate traffic participant at each acquisition moment. The motion information includes motion direction and motion speed. This embodiment can quickly and accurately determine multiple potential interaction objects of the target motor vehicle.

[0106] In some embodiments, for each collection moment, the relative distance, relative movement direction, and / or relative speed between the target motor vehicle and the candidate traffic participant at the collection moment are determined based on the position information and movement information of the target motor vehicle and the candidate traffic participant at the collection moment. In response to the relative distance, relative movement direction, and / or relative speed meeting the preset interaction conditions, the candidate traffic participant is determined as a potential interaction partner of the target motor vehicle. The preset interaction conditions can be set by the user or adopted as default conditions, which are not specifically limited in the embodiments of the present invention. For example, Figure 13As shown, the relative distance between the target motor vehicle 30 and the first traffic participant 301 and the second traffic participant 302 at the collection time t is less than the preset distance threshold, and the relative distance between the target motor vehicle 30 and the third traffic participant 32 at the collection time t is greater than the preset distance threshold. Therefore, the first traffic participant 301 and the second traffic participant 302 are determined as potential interaction objects of the target motor vehicle 30, while the third traffic participant 32 is not a potential interaction object of the target motor vehicle 30.

[0107] Step S103 : Determine the value of the interaction risk measurement index between the target motor vehicle and each potential interaction object based on the trajectory data of the target motor vehicle and the trajectory data of each potential interaction object.

[0108] In this embodiment, interaction risk metrics may include surrogate safety management (SSM), among others. These surrogate safety metrics may include post-encroachment time (PET), time to collision (TTC), and vector-based time to collision (VTTC). It is understood that different surrogate safety metrics may be selected as interaction risk metrics between the target vehicle and potential interaction objects in different scenarios. For example, when extracting interaction behaviors between traffic participants on structured roads with clear lane markings and good traffic order, PET, TTC, or VTTC may be selected as interaction risk metrics between the target vehicle and potential interaction objects. For another example, when extracting interaction behaviors between traffic participants on unstructured roads where vehicles do not strictly follow lane markings (e.g., mixed traffic scenarios), VTTC may be selected as interaction risk metrics between the target vehicle and potential interaction objects. This addresses the limitations of traditional metrics (PET and TTC) in complex traffic environments and enhances the accuracy and reliability of interaction behavior extraction.

[0109] In some embodiments, a numerical value of an interaction risk measurement indicator between the target motor vehicle and each potential interaction object is determined based on the trajectory data of the target motor vehicle and the trajectory data of each potential interaction object, including: determining the position vector and velocity vector of the target motor vehicle at each collection moment based on the trajectory information at each collection moment in the trajectory data of the target motor vehicle; determining the position vector and velocity vector of each potential interaction object at each collection moment based on the trajectory information at each collection moment in the trajectory data of each potential interaction object; and determining the numerical value of the vector collision time between the target motor vehicle and each potential interaction object at each collection moment based on the position vector and velocity vector of the target motor vehicle at each collection moment and the position vector and velocity vector of each potential interaction object at each collection moment. This embodiment determines the numerical value of the vector collision time between the target motor vehicle and the potential interaction object by comprehensively considering the velocity vector and position vector of the target motor vehicle and the potential interaction object, thereby more accurately representing the collision risk between the target motor vehicle and the potential interaction object, thereby facilitating more effective subsequent identification of potential interaction objects with interaction influence.

[0110] In some embodiments, the numerical value of the vector collision time between the target motor vehicle and each potential interactive object at each collection moment is determined based on the position vector and velocity vector of the target motor vehicle at each collection moment and the position vector and velocity vector of each potential interactive object at each collection moment, including: determining the relative position vector of the target motor vehicle and the potential interactive object at the collection moment based on their respective position vectors at the same collection moment; determining the relative velocity vector of the target motor vehicle and the potential interactive object at the collection moment based on their respective velocity vectors at the same collection moment; and obtaining the numerical value of the vector collision time between the target motor vehicle and the potential interactive object at the collection moment by dividing the square of the modulus of the relative position vector by the modulus of the target vector obtained by vector multiplication of the relative position vector and the relative velocity vector.

[0111] The value of the vector collision time represents the time it would take for the target vehicle and the potential interactive object to collide at the same acquisition moment. Specifically, a negative vector collision time value indicates that the target vehicle and the potential interactive object are moving away from each other, and the collision risk is low. A vector collision time value of 0 indicates that the relative movement direction and relative position direction of the target vehicle and the potential interactive object are perpendicular, and the collision risk is low. A positive vector collision time value indicates that the target vehicle and the potential interactive object are approaching. The smaller the vector collision time value, the higher the collision risk between the target vehicle and the potential interactive object.

[0112] In some embodiments, the setting of the target numerical range of the interaction risk measurement indicator is crucial to ensure the accuracy of the interaction behavior extraction. The target numerical range can be determined by analyzing the box plot of the numerical distribution of the interaction risk measurement indicator. Exemplarily, based on the numerical values ​​of the interaction risk measurement indicator of the target motor vehicle and each potential interaction object at each acquisition moment, a box plot of the numerical values ​​of the interaction risk measurement indicator is generated; based on the box plot, the minimum value, the first quartile value, the third quartile value and the maximum value of the interaction risk measurement indicator are determined; based on the minimum value, the first quartile value, the third quartile value and the maximum value of the interaction risk measurement indicator, the target numerical range of the interaction risk measurement indicator is determined. This embodiment can accurately determine the target numerical range of the interaction risk measurement indicator.

[0113] In some embodiments, the target numerical range of the interaction risk measurement indicator is determined based on the minimum value, first quartile value, third quartile value and maximum value of the interaction risk measurement indicator, including: subtracting the first quartile value from the third quartile value to obtain the interquartile range; determining the maximum of the minimum value and the difference between the first quartile value and 1.5 times the interquartile range as the lower numerical threshold of the interaction risk measurement indicator; determining the minimum of the maximum value and the sum of the third quartile value and 1.5 times the interquartile range as the upper numerical threshold of the interaction risk measurement indicator; and determining the lower numerical threshold, the upper numerical threshold and the values ​​between the lower numerical threshold and the upper numerical threshold as the target numerical range.

[0114] Step S104: Determine the potential interaction objects whose interaction risk measurement index values ​​are within the target value range as the target interaction objects of the target motor vehicle, and use the trajectory data of the target motor vehicle and the trajectory data of the corresponding target interaction objects as interaction behavior scenarios, and add them to the interaction behavior scenario set of the target motor vehicle.

[0115] For example, at collection time t, the target vehicle has a first potential interaction partner and a second potential interaction partner, and the value of the interaction risk measurement indicator between the target vehicle and the first potential interaction partner is within the target value range, while the value of the interaction risk measurement indicator between the target vehicle and the second potential interaction partner is outside the target value range. Therefore, the trajectory data of the target vehicle between collection time tn and collection time t+n, and the trajectory data of the second potential interaction partner between collection time tn and collection time t+n, can be used as interaction behavior scenarios and added to the target vehicle's interaction behavior scenario set. Here, n is an integer greater than or equal to 0.

[0116] In some embodiments, after step S104, the method further includes: determining the target vehicle category, the type of the target vehicle's route, the direction of interaction between the target vehicle and the target interaction object, and the category of the target interaction object based on the interaction behavior scenario; and classifying the interaction behavior scenario based on the target vehicle category, the type of the target vehicle's route, the direction of interaction between the target vehicle and the target interaction object, and the category of the target interaction object. By classifying the interaction behavior scenarios, this embodiment can obtain a library of interaction behavior scenarios of different categories, which can provide a reliable scenario data foundation for driver behavior research and scenario-based autonomous driving testing research.

[0117] In some embodiments, the target motor vehicle category may include cars, trucks, and buses, the target motor vehicle's route type may include straight or turning, and the interaction direction may include vertical interaction, same-direction interaction, or reverse interaction. The target interaction object category may include cars, trucks, buses, cyclists, electric motorcycles, tricycles, and pedestrians. For example, the interaction behavior scenario may be as follows: Figure 15 For example, the interactive scene containing a car and a cyclist is classified according to the following classification diagram. Figure 15 The classification diagram shown in the figure can be classified as follows Figure 16 The six classification results shown include the vertical interaction between the car and the cyclist when turning, the opposite interaction (reverse interaction) between the car and the cyclist when turning, the same-direction interaction between the car and the cyclist when turning, the vertical interaction between the car and the cyclist when driving straight, the reverse interaction between the car and the cyclist when driving straight, and the same-direction interaction between the car and the cyclist when driving straight.

[0118] Because the interactive behavior scenario classification incorporates the type of driving route, this classification can clearly distinguish the target vehicle's driving purpose at the intersection and better reflect the driving behavior differences between a target vehicle traveling straight and a target vehicle turning at the same intersection location. A target vehicle traveling straight typically focuses on the safe distance from the vehicle ahead, visibility at the intersection ahead, and the presence of obstacles or other traffic interference. During straight driving, the target vehicle's decision-making primarily depends on its current speed, road conditions ahead, and its relative position to the vehicle ahead. The system must adjust based on real-time traffic flow information and intersection conditions to ensure smooth intersection passage while maintaining a safe distance. Compared to a target vehicle traveling straight, a target vehicle turning faces more dynamic factors. These factors include not only the distance to the vehicle ahead, speed control, and visibility, but also the constant consideration of the motion interference of traffic participants within the intersection. A target vehicle turning often faces more complex traffic environments. For example, when changing lanes or turning left or right, it must assess traffic flow in the oncoming lane, space availability during the turn, and potential interference from pedestrians or non-motorized vehicles.

[0119] In some embodiments, vertical interaction refers to the interaction between traffic participants in the longitudinal direction of the road (such as sideways crossing, cutting in, etc.). For example, pedestrians suddenly crossing the road, non-motor vehicles cutting into motor vehicle lanes, non-motor vehicles cutting in, etc., these situations usually lead to sudden interruptions in traffic flow or emergency avoidance. Vertical interaction scenarios pose challenges to the prediction and avoidance strategies of autonomous driving systems, especially in scenarios involving vulnerable traffic participants (such as pedestrians and cyclists). Same-direction interaction refers to the interaction between traffic participants traveling in the same direction, usually involving speed differences, following behaviors, lane changes, etc. For example, emergency braking of the vehicle in front, vehicle merging, following, overtaking, etc., this type of interaction places high demands on the real-time decision-making and distance control of the autonomous driving system. Especially on highways or congested urban roads, small decision-making differences in same-direction interactions may lead to safety risks. Reverse interaction refers to the interaction between traffic participants in the opposite direction of the road, usually involving lane misalignment or traffic violations. For example, avoiding oncoming vehicles, mistakenly entering the opposite lane, meeting vehicles in the wrong lane, etc. These behaviors require the autonomous driving system to have stronger path planning and risk prediction capabilities to ensure that it can take evasive measures in time when facing oncoming vehicles.

[0120] In some embodiments, after step S104, the following further includes: adding the target motor vehicle's interactive behavior scene set to the interactive behavior scene library. When constructing the interactive behavior scene library, the target motor vehicle (primary participant) is first traversed, and the driving area map corresponding to the driving route is matched. Potential interactive objects (secondary participants) are searched in the driving area map, and the existence of interactive behavior is determined based on the VTTC value between the target motor vehicle and the potential interactive object. This process is repeated until the target motor vehicle is traversed and the interactive behavior scene library is obtained. The specific construction process can be as follows: Figure 17 shown.

[0121] like Figure 17As shown, the process of constructing an interactive behavior scenario library may include: traversing a target motor vehicle Veh with a clear driving route; matching a driving area map corresponding to the driving route; traversing the trajectory data of the target motor vehicle Veh at each collection moment; searching for potential interactive objects Obj in the driving area map of the target motor vehicle Veh at the current collection moment; calculating the value of VTTC between the target motor vehicle Veh and the potential interactive object Obj based on the trajectory data of the target motor vehicle Veh and the potential interactive object Obj at the current collection moment; judging whether the value of VTTC is within the target value range; if the value of VTTC is within the target value range, judging whether the target motor vehicle Veh and the potential interactive object Obj are interacting for the first time, and if the target motor vehicle Veh and the potential interactive object Obj are interacting for the first time, creating an interactive behavior scenario set between the target motor vehicle Veh and the potential interactive object Obj; if the target motor vehicle Veh and the potential interactive object Obj are not interacting for the first time, adding the trajectory data of the target motor vehicle Veh and the potential interactive object Obj to the corresponding interactive behavior scenario set; if VTTC If the value of is not within the target value range, it is determined whether there is an interaction behavior scene set between the target motor vehicle Veh and the potential interaction object Obj. If there is an interaction behavior scene set between the target motor vehicle Veh and the potential interaction object Obj, the interaction behavior scene set between the target motor vehicle Veh and the potential interaction object Obj is added to the interaction behavior scene library; if there is no interaction behavior scene set between the target motor vehicle Veh and the potential interaction object Obj, it is determined whether the trajectory data of each acquisition moment of the currently traversed target motor vehicle Veh is traversed; if the trajectory data of each acquisition moment of the currently traversed target motor vehicle Veh is not traversed, the step of traversing the trajectory data of the target motor vehicle Veh at each acquisition moment is returned to; if the trajectory data of each acquisition moment of the currently traversed target motor vehicle Veh is traversed, it is determined whether all target motor vehicles are traversed; if not all target motor vehicles are traversed, the step of traversing the target motor vehicle Veh with a clear driving route is returned to; if all target motor vehicles are traversed, the construction of the interaction behavior scene library is completed, and the final interaction behavior scene library is obtained.

[0122] See also Figure 18 , Figure 18 This is a schematic block diagram of the structure of a computer device provided by an embodiment of the present invention.

[0123] like Figure 18 As shown, the computer device 100 includes a processor 101 and a memory 102 , and the processor 101 and the memory 102 are connected via a bus 103 , such as an I 2 C (Inter-Integrated Circuit) bus.

[0124] Specifically, the processor 101 is used to provide computing and control capabilities to support the operation of the entire computer device. The processor 101 can be a central processing unit (CPU), and the processor 101 can also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc.

[0125] Specifically, the memory 102 may be a Flash chip, a read-only memory (ROM) disk, an optical disk, a USB flash drive, or a mobile hard disk.

[0126] Those skilled in the art will understand that Figure 18 The structure shown in the figure is merely a block diagram of a portion of the structure related to the embodiment of the present invention, and does not constitute a limitation on the computer device to which the embodiment of the present invention is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0127] The processor 101 is configured to run a computer program stored in the memory 102 and implement any one of the methods for extracting interactive behaviors based on a driving area map provided by the embodiments of the present invention when executing the computer program.

[0128] In some embodiments, the processor 101 is configured to run a computer program stored in a memory, and implement the following steps when executing the computer program:

[0129] Acquire a traffic participant trajectory dataset of a target road section, and acquire a target trajectory dataset from the traffic participant trajectory dataset, wherein the target trajectory dataset includes trajectory data of a plurality of target motor vehicles with clear travel routes;

[0130] searching, based on the target motor vehicle's trajectory data and the traffic participant trajectory dataset, for a plurality of potential interaction objects of the target motor vehicle in a driving area map corresponding to the target motor vehicle's driving route, wherein the driving area map is constructed based on the target trajectory dataset and is used to represent drivable areas of the plurality of target motor vehicles on the target road segment;

[0131] determining a value of an interaction risk measurement index between the target motor vehicle and each of the potential interaction objects based on the trajectory data of the target motor vehicle and the trajectory data of each of the potential interaction objects;

[0132] The potential interaction object whose value of the interaction risk measurement index is within the target value range is determined as the target interaction object of the target motor vehicle, and the trajectory data of the target motor vehicle and the corresponding trajectory data of the target interaction object are used as interaction behavior scenes and added to the interaction behavior scene set of the target motor vehicle.

[0133] In some embodiments, when searching for multiple potential interaction objects of the target motor vehicle in a driving area map corresponding to the driving route of the target motor vehicle based on the trajectory data of the target motor vehicle and the traffic participant trajectory dataset, the processor 101 is configured to implement:

[0134] Acquiring trajectory data of a plurality of candidate traffic participants from the traffic participant trajectory dataset, wherein the plurality of candidate traffic participants are located in a driving area map corresponding to the driving route of the target motor vehicle;

[0135] determining, based on the trajectory data of the target motor vehicle and the trajectory data of each candidate traffic participant, position information and motion information of the target motor vehicle and each candidate traffic participant at each collection moment;

[0136] Based on the position information and movement information of the target motor vehicle and each of the candidate traffic participants at each collection moment, a plurality of potential interaction objects of the target motor vehicle are determined from the plurality of candidate traffic participants.

[0137] In some embodiments, when determining the value of the interaction risk measurement indicator between the target motor vehicle and each of the potential interaction objects based on the trajectory data of the target motor vehicle and the trajectory data of each of the potential interaction objects, the processor 101 is configured to implement:

[0138] Determining a position vector and a velocity vector of the target motor vehicle at each collection moment based on the trajectory information of the target motor vehicle at each collection moment;

[0139] determining, according to the trajectory information at each collection moment in the trajectory data of each potential interactive object, a position vector and a velocity vector of each potential interactive object at each collection moment;

[0140] The value of the vector collision time between the target motor vehicle and each potential interactive object at each collection moment is determined based on the position vector and velocity vector of the target motor vehicle at each collection moment and the position vector and velocity vector of each potential interactive object at each collection moment.

[0141] In some embodiments, when determining the value of the vector collision time between the target motor vehicle and each potential interactive object at each collection moment based on the position vector and velocity vector of the target motor vehicle at each collection moment and the position vector and velocity vector of each potential interactive object at each collection moment, the processor 101 is configured to implement:

[0142] determining a relative position vector of the target motor vehicle and the potential interactive object at the acquisition time based on the position vectors of the target motor vehicle and the potential interactive object at the same acquisition time;

[0143] determining a relative velocity vector between the target motor vehicle and the potential interactive object at the acquisition moment based on the velocity vectors of the target motor vehicle and the potential interactive object at the same acquisition moment;

[0144] The square value of the modulus of the relative position vector is divided by the modulus of the target vector obtained by vector multiplication of the relative position vector and the relative velocity vector to obtain the value of the vector collision time between the target motor vehicle and the potential interactive object at the collection moment.

[0145] In some embodiments, the processor 101 is further configured to implement the following steps:

[0146] generating a box plot of the values ​​of the interaction risk measurement indexes according to the values ​​of the interaction risk measurement indexes between the target motor vehicle and each of the potential interaction objects at each collection moment;

[0147] Determining, based on the box plot, a minimum value, a first quartile value, a third quartile value, and a maximum value of the interaction risk measure;

[0148] The target numerical range of the interaction risk measurement indicator is determined according to the minimum value, the first quartile value, the third quartile value and the maximum value of the interaction risk measurement indicator.

[0149] In some embodiments, when determining the target numerical range of the interaction risk measurement indicator based on the minimum value, the first quartile value, the third quartile value, and the maximum value of the interaction risk measurement indicator, the processor 101 is configured to implement:

[0150] Subtracting the first quartile value from the third quartile value to obtain the interquartile range;

[0151] determining the maximum of the minimum value and the difference between the first quartile value and 1.5 times the interquartile range as the lower numerical threshold of the interaction risk measurement indicator;

[0152] Determine the minimum of the maximum value and the sum of the third quartile value and 1.5 times the interquartile range as the upper limit numerical threshold of the interaction risk measurement indicator;

[0153] The lower numerical threshold, the upper numerical threshold, and numerical values ​​between the lower numerical threshold and the upper numerical threshold are determined as the target numerical range.

[0154] In some embodiments, after adding the trajectory data of the target motor vehicle and the corresponding trajectory data of the target interactive object as an interactive behavior scene to the interactive behavior scene set of the target motor vehicle, the processor 101 is further configured to:

[0155] Determining, based on the interaction behavior scenario, the category of the target motor vehicle, the type of the target motor vehicle's travel route, the interaction direction between the target motor vehicle and the target interaction object, and the category of the target interaction object;

[0156] The interactive behavior scenario is classified according to the category of the target motor vehicle, the type of the driving route of the target motor vehicle, the interaction direction between the target motor vehicle and the target interactive object, and the category of the target interactive object.

[0157] In some embodiments, when acquiring the target trajectory dataset from the traffic participant trajectory dataset, the processor 101 is configured to:

[0158] Filtering out motor vehicle trajectory data with a trajectory duration greater than or equal to a target duration threshold from the traffic participant trajectory dataset to obtain a first motor vehicle trajectory dataset;

[0159] performing normalization processing on each motor vehicle trajectory data in the first motor vehicle trajectory dataset to obtain a second motor vehicle trajectory dataset;

[0160] performing clustering processing on the second motor vehicle trajectory dataset to obtain clusters having the same number of target centroids, wherein the clusters are subsets of the second motor vehicle trajectory dataset;

[0161] According to the lane information of the target road section, the clusters of the target centroid number are screened to obtain at least one target cluster, and one target cluster corresponds to one target trajectory data set.

[0162] In some embodiments, after obtaining a traffic participant trajectory dataset of a target road segment and obtaining a target trajectory dataset from the traffic participant trajectory dataset, the processor 101 is further configured to:

[0163] generating a motor vehicle trajectory image of a corresponding route of the target road section according to the target trajectory dataset, and extracting a driving area contour of the route from the motor vehicle trajectory image;

[0164] The driving area outline is smoothed to obtain a target driving area outline, and the driving area included in the target driving area outline is filled to obtain a driving area map corresponding to the route.

[0165] In some embodiments, when extracting the driving area contour of the route from the motor vehicle trajectory image, the processor 101 is configured to implement:

[0166] performing a binarization process on the vehicle trajectory image to obtain a binarized image, and performing an image corrosion process on the binarized image to obtain a candidate binarized image;

[0167] Detecting outliers in the candidate binary image using an outlier detection algorithm, and deleting the outliers in the candidate binary image to obtain a target binary image;

[0168] Perform image expansion processing on the target binary image to obtain the driving area outline of the route.

[0169] In some embodiments, when acquiring a traffic participant trajectory dataset of a target road segment, the processor 101 is configured to:

[0170] Obtain a target bird's-eye view image sequence corresponding to the traffic flow of the target road section;

[0171] A traffic participant trajectory dataset of the target road section is obtained according to the target bird's-eye view image sequence.

[0172] In some embodiments, when acquiring the traffic participant trajectory dataset of the target road section according to the target bird's-eye view image sequence, the processor 101 is configured to implement:

[0173] calling a preset traffic participant detection model to perform traffic participant detection on each bird's-eye view image in the target bird's-eye view image sequence, and obtaining a traffic participant detection result for each bird's-eye view image;

[0174] determining static data of each traffic participant in each of the bird's-eye view images according to the traffic participant detection result of each of the bird's-eye view images;

[0175] performing multi-target tracking on the traffic participants in each of the bird's-eye view images based on static data of each traffic participant in each of the bird's-eye view images, and assigning a tracking ID to each of the traffic participants that is successfully tracked;

[0176] According to the tracking ID of each traffic participant, the static data of each traffic participant in each bird's-eye view image are associated to obtain a static data sequence corresponding to each tracking ID;

[0177] Correcting the static data in the static data sequence corresponding to each tracking ID to obtain a target static data sequence corresponding to each tracking ID;

[0178] Determining a speed data sequence of a traffic participant corresponding to each tracking ID according to a target static data sequence corresponding to each tracking ID;

[0179] The target static data sequence and the speed data sequence corresponding to the same tracking ID are used as trajectory data of the corresponding traffic participant.

[0180] It should be noted that those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the computer device described above can refer to the corresponding process in the aforementioned embodiment of the interactive behavior extraction method based on the driving area map, and will not be repeated here.

[0181] See also Figure 19 , Figure 19 This is a schematic block diagram of the structure of an interactive behavior extraction system based on a driving area map provided by an embodiment of the present invention.

[0182] like Figure 19 As shown, the interactive behavior extraction system 1000 based on the driving area map includes a computer device 100, a height maintenance device 200 and an image acquisition device 300, wherein:

[0183] The height maintaining device 200 is deployed on the target road section and is used to carry the image acquisition device 300, so that the image acquisition device 300 can acquire a plurality of original bird's-eye view image sequences of the traffic flow of the target road section from a bird's-eye view perspective;

[0184] The image acquisition device 300 is mounted on the altitude maintaining device 200 and is used to acquire a plurality of original bird's-eye view image sequences of the traffic flow of the target road section from a bird's-eye view perspective;

[0185] The computer device 100 is configured to obtain the plurality of original bird's-eye view image sequences and obtain a traffic participant trajectory dataset of the target road section based on the plurality of original bird's-eye view image sequences;

[0186] The computer device 100 is further configured to obtain a target trajectory dataset from the traffic participant trajectory dataset, wherein the target trajectory dataset includes trajectory data of a plurality of target motor vehicles with clear driving routes;

[0187] The computer device 100 is further configured to construct a driving area map on the target road segment based on the target trajectory dataset, wherein the driving area map is configured to represent the driving areas of the multiple target motor vehicles on the target road segment;

[0188] The computer device 100 is further configured to search for a plurality of potential interaction objects of the target motor vehicle in the driving area map corresponding to the driving route of the target motor vehicle based on the trajectory data of the target motor vehicle and the traffic participant trajectory dataset;

[0189] The computer device 100 is further configured to determine a value of an interaction risk measurement indicator between the target motor vehicle and each of the potential interaction objects based on the trajectory data of the target motor vehicle and the trajectory data of each of the potential interaction objects;

[0190] The computer device 100 is further used to determine the potential interactive object whose value of the interaction risk measurement index is within the target value range as the target interactive object of the target motor vehicle, and to add the trajectory data of the target motor vehicle and the corresponding trajectory data of the target interactive object as interactive behavior scenarios to the interactive behavior scenario set of the target motor vehicle.

[0191] In some embodiments, obtaining a traffic participant trajectory dataset of the target road section based on the multiple original bird's-eye view image sequences includes:

[0192] Acquire a target bird's-eye view image sequence corresponding to the traffic flow of the target road section according to the multiple original bird's-eye view image sequences;

[0193] A traffic participant trajectory dataset of the target road section is obtained according to the target bird's-eye view image sequence.

[0194] In some embodiments, obtaining a target bird's-eye view image sequence corresponding to the traffic flow of the target road section based on the multiple original bird's-eye view image sequences includes:

[0195] performing image stabilization processing on each of the plurality of original bird's-eye view image sequences;

[0196] The plurality of original bird's-eye view image sequences that have undergone image stabilization processing are aligned to the same image coordinate system to obtain a plurality of target bird's-eye view image sequences.

[0197] In some embodiments, obtaining a traffic participant trajectory dataset of the target road section based on the target bird's-eye view image sequence includes:

[0198] calling a preset traffic participant detection model to perform traffic participant detection on each bird's-eye view image in the target bird's-eye view image sequence, and obtaining a traffic participant detection result for each bird's-eye view image;

[0199] determining static data of each traffic participant in each of the bird's-eye view images according to the traffic participant detection result of each of the bird's-eye view images;

[0200] performing multi-target tracking on the traffic participants in each of the bird's-eye view images based on static data of each traffic participant in each of the bird's-eye view images, and assigning a tracking ID to each of the traffic participants that is successfully tracked;

[0201] According to the tracking ID of each traffic participant, the static data of each traffic participant in each bird's-eye view image are associated to obtain a static data sequence corresponding to each tracking ID;

[0202] Correcting the static data in the static data sequence corresponding to each tracking ID to obtain a target static data sequence corresponding to each tracking ID;

[0203] Determining a speed data sequence of a traffic participant corresponding to each tracking ID according to a target static data sequence corresponding to each tracking ID;

[0204] The target static data sequence and the speed data sequence corresponding to the same tracking ID are used as trajectory data of the corresponding traffic participant.

[0205] In some embodiments, constructing a driving area map on the target road segment based on the target trajectory dataset includes:

[0206] generating a motor vehicle trajectory image of a corresponding route of the target road section according to the target trajectory dataset, and extracting a driving area contour of the route from the motor vehicle trajectory image;

[0207] The driving area outline is smoothed to obtain a target driving area outline, and the driving area included in the target driving area outline is filled to obtain a driving area map corresponding to the route.

[0208] In some embodiments, extracting the driving area contour of the route from the motor vehicle trajectory image includes:

[0209] performing a binarization process on the vehicle trajectory image to obtain a binarized image, and performing an image corrosion process on the binarized image to obtain a candidate binarized image;

[0210] Detecting outliers in the candidate binary image using an outlier detection algorithm, and deleting the outliers in the candidate binary image to obtain a target binary image;

[0211] Perform image expansion processing on the target binary image to obtain the driving area outline of the route.

[0212] In some embodiments, obtaining a target trajectory dataset from the traffic participant trajectory dataset includes:

[0213] Filtering out motor vehicle trajectory data with a trajectory duration greater than or equal to a target duration threshold from the traffic participant trajectory dataset to obtain a first motor vehicle trajectory dataset;

[0214] performing normalization processing on each motor vehicle trajectory data in the first motor vehicle trajectory dataset to obtain a second motor vehicle trajectory dataset;

[0215] performing clustering processing on the second motor vehicle trajectory dataset to obtain clusters having the same number of target centroids, wherein the clusters are subsets of the second motor vehicle trajectory dataset;

[0216] According to the lane information of the target road section, the clusters of the target centroid number are screened to obtain at least one target cluster, and one target cluster corresponds to one target trajectory data set.

[0217] In some embodiments, searching for multiple potential interaction objects of the target motor vehicle in a driving area map corresponding to the driving route of the target motor vehicle based on the trajectory data of the target motor vehicle and the traffic participant trajectory dataset includes:

[0218] Acquiring trajectory data of a plurality of candidate traffic participants from the traffic participant trajectory dataset, wherein the plurality of candidate traffic participants are located in a driving area map corresponding to the driving route of the target motor vehicle;

[0219] determining, based on the trajectory data of the target motor vehicle and the trajectory data of each candidate traffic participant, position information and motion information of the target motor vehicle and each candidate traffic participant at each collection moment;

[0220] Based on the position information and movement information of the target motor vehicle and each of the candidate traffic participants at each collection moment, a plurality of potential interaction objects of the target motor vehicle are determined from the plurality of candidate traffic participants.

[0221] In some embodiments, determining a value of an interaction risk measurement indicator between the target motor vehicle and each of the potential interaction objects based on the trajectory data of the target motor vehicle and the trajectory data of each of the potential interaction objects includes:

[0222] Determining a position vector and a velocity vector of the target motor vehicle at each collection moment based on the trajectory information of the target motor vehicle at each collection moment;

[0223] determining, according to the trajectory information at each collection moment in the trajectory data of each potential interactive object, a position vector and a velocity vector of each potential interactive object at each collection moment;

[0224] The value of the vector collision time between the target motor vehicle and each potential interactive object at each collection moment is determined based on the position vector and velocity vector of the target motor vehicle at each collection moment and the position vector and velocity vector of each potential interactive object at each collection moment.

[0225] In some embodiments, determining the value of the vector collision time between the target motor vehicle and each potential interactive object at each collection moment based on the position vector and velocity vector of the target motor vehicle at each collection moment and the position vector and velocity vector of each potential interactive object at each collection moment includes:

[0226] determining a relative position vector of the target motor vehicle and the potential interactive object at the acquisition time based on the position vectors of the target motor vehicle and the potential interactive object at the same acquisition time;

[0227] determining a relative velocity vector between the target motor vehicle and the potential interactive object at the acquisition moment based on the velocity vectors of the target motor vehicle and the potential interactive object at the same acquisition moment;

[0228] The square value of the modulus of the relative position vector is divided by the modulus of the target vector obtained by vector multiplication of the relative position vector and the relative velocity vector to obtain the value of the vector collision time between the target motor vehicle and the potential interactive object at the collection moment.

[0229] In some embodiments, the computer device is further configured to:

[0230] generating a box plot of the values ​​of the interaction risk measurement indexes according to the values ​​of the interaction risk measurement indexes between the target motor vehicle and each of the potential interaction objects at each collection moment;

[0231] Determining, based on the box plot, a minimum value, a first quartile value, a third quartile value, and a maximum value of the interaction risk measure;

[0232] The target numerical range of the interaction risk measurement indicator is determined according to the minimum value, the first quartile value, the third quartile value and the maximum value of the interaction risk measurement indicator.

[0233] In some embodiments, determining the target numerical range of the interaction risk measurement indicator according to the minimum value, the first quartile value, the third quartile value, and the maximum value of the interaction risk measurement indicator includes:

[0234] Subtracting the first quartile value from the third quartile value to obtain the interquartile range;

[0235] determining the maximum of the minimum value and the difference between the first quartile value and 1.5 times the interquartile range as the lower numerical threshold of the interaction risk measurement indicator;

[0236] Determine the minimum of the maximum value and the sum of the third quartile value and 1.5 times the interquartile range as the upper limit numerical threshold of the interaction risk measurement indicator;

[0237] The lower numerical threshold, the upper numerical threshold, and numerical values ​​between the lower numerical threshold and the upper numerical threshold are determined as the target numerical range.

[0238] In some embodiments, after adding the trajectory data of the target motor vehicle and the corresponding trajectory data of the target interactive object as an interactive behavior scene to the interactive behavior scene set of the target motor vehicle, the computer device 100 is further configured to:

[0239] Determining, based on the interaction behavior scenario, the category of the target motor vehicle, the type of the target motor vehicle's travel route, the interaction direction between the target motor vehicle and the target interaction object, and the category of the target interaction object;

[0240] The interactive behavior scenario is classified according to the category of the target motor vehicle, the type of the driving route of the target motor vehicle, the interaction direction between the target motor vehicle and the target interactive object, and the category of the target interactive object.

[0241] In some embodiments, the altitude maintaining device 200 may include an unmanned aerial vehicle, a stand, or a street lamp, and the image acquisition device 300 may include a digital camera, a single-lens reflex camera, an infrared camera, or a depth camera.

[0242] In some embodiments, the interactive behavior extraction system 1000 based on the driving area map further includes an image storage device, which is used to store multiple original bird's-eye view image sequences obtained by the image acquisition device 300 from capturing the traffic flow of the target road section from a bird's-eye view. The unmanned aerial vehicle is communicatively connected to the image storage device so that the unmanned aerial vehicle transmits the multiple original bird's-eye view image sequences obtained by the image acquisition device 300 from capturing the traffic flow of the target road section from a bird's-eye view to the image storage device for storage. Alternatively, the image acquisition device is communicatively connected to the image storage device so that the image acquisition device transmits the multiple original bird's-eye view image sequences obtained by capturing the traffic flow of the target road section from a bird's-eye view to the image storage device for storage.

[0243] It should be noted that technical personnel in the relevant field can clearly understand that for the convenience and conciseness of description, the specific working process of the interactive behavior extraction system based on the driving area map described above can refer to the corresponding process in the aforementioned embodiment of the interactive behavior extraction method based on the driving area map, and will not be repeated here.

[0244] An embodiment of the present invention also provides a storage medium for computer-readable storage, wherein the storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement any interactive behavior extraction method based on a driving area map as provided in the description of the embodiment of the present invention.

[0245] The storage medium may be an internal storage unit of the computer device described in the aforementioned embodiment, such as a hard disk or memory of the computer device. The storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the computer device.

[0246] It will be appreciated by those skilled in the art that all or some of the steps, systems, and functional modules / units in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In a hardware embodiment, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or temporary medium). As is well known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable, and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is well known to those skilled in the art that communication media generally embodies computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0247] It should be understood that the term "and / or" used in the present specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, including these combinations. It should be noted that, in this article, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system that includes a series of elements includes not only those elements, but also includes other elements that are not explicitly listed, or also includes elements that are inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "including a..." does not exclude the presence of other identical elements in the process, method, article or system that includes the element.

[0248] The serial numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments. The above description is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be based on the scope of protection of the claims.

Claims

1. A method for extracting interactive behaviors based on a driving area map, characterized in that: include: Acquire a traffic participant trajectory dataset of a target road section, and acquire a target trajectory dataset from the traffic participant trajectory dataset, wherein the target trajectory dataset includes trajectory data of a plurality of target motor vehicles with clear travel routes; searching, based on the target motor vehicle's trajectory data and the traffic participant trajectory dataset, for a plurality of potential interaction objects of the target motor vehicle in a driving area map corresponding to the target motor vehicle's driving route, wherein the driving area map is constructed based on the target trajectory dataset and is used to represent drivable areas of the plurality of target motor vehicles on the target road segment; determining a value of an interaction risk measurement index between the target motor vehicle and each of the potential interaction objects based on the trajectory data of the target motor vehicle and the trajectory data of each of the potential interaction objects; The potential interaction object whose value of the interaction risk measurement index is within the target value range is determined as the target interaction object of the target motor vehicle, and the trajectory data of the target motor vehicle and the corresponding trajectory data of the target interaction object are used as interaction behavior scenes and added to the interaction behavior scene set of the target motor vehicle.

2. The interactive behavior extraction method according to claim 1, characterized in that: The step of searching for a plurality of potential interaction objects of the target motor vehicle in a driving area map corresponding to a driving route of the target motor vehicle based on the trajectory data of the target motor vehicle and the traffic participant trajectory dataset includes: Acquiring trajectory data of a plurality of candidate traffic participants from the traffic participant trajectory dataset, wherein the plurality of candidate traffic participants are located in a driving area map corresponding to the driving route of the target motor vehicle; determining, based on the trajectory data of the target motor vehicle and the trajectory data of each candidate traffic participant, position information and motion information of the target motor vehicle and each candidate traffic participant at each collection moment; Based on the position information and movement information of the target motor vehicle and each of the candidate traffic participants at each collection moment, a plurality of potential interaction objects of the target motor vehicle are determined from the plurality of candidate traffic participants.

3. The interactive behavior extraction method according to claim 1, characterized in that: Determining a value of an interaction risk measurement index between the target motor vehicle and each of the potential interaction objects based on the trajectory data of the target motor vehicle and the trajectory data of each of the potential interaction objects includes: Determining a position vector and a velocity vector of the target motor vehicle at each collection moment based on the trajectory information of the target motor vehicle at each collection moment; determining, according to the trajectory information at each collection moment in the trajectory data of each potential interactive object, a position vector and a velocity vector of each potential interactive object at each collection moment; The value of the vector collision time between the target motor vehicle and each potential interactive object at each collection moment is determined based on the position vector and velocity vector of the target motor vehicle at each collection moment and the position vector and velocity vector of each potential interactive object at each collection moment.

4. The interactive behavior extraction method according to claim 3, characterized in that: Determining a value of a vector collision time between the target motor vehicle and each of the potential interactive objects at each collection moment based on the position vector and velocity vector of the target motor vehicle at each collection moment and the position vector and velocity vector of each of the potential interactive objects at each collection moment includes: determining a relative position vector of the target motor vehicle and the potential interactive object at the acquisition time based on the position vectors of the target motor vehicle and the potential interactive object at the same acquisition time; determining a relative velocity vector between the target motor vehicle and the potential interactive object at the acquisition moment based on the velocity vectors of the target motor vehicle and the potential interactive object at the same acquisition moment; The square value of the modulus of the relative position vector is divided by the modulus of the target vector obtained by vector multiplication of the relative position vector and the relative velocity vector to obtain the value of the vector collision time between the target motor vehicle and the potential interactive object at the collection moment.

5. The interactive behavior extraction method according to claim 1, characterized in that: The method further comprises: generating a box plot of the values ​​of the interaction risk measurement indexes according to the values ​​of the interaction risk measurement indexes between the target motor vehicle and each of the potential interaction objects at each collection moment; Determining, based on the box plot, a minimum value, a first quartile value, a third quartile value, and a maximum value of the interaction risk measure; The target numerical range of the interaction risk measurement indicator is determined according to the minimum value, the first quartile value, the third quartile value and the maximum value of the interaction risk measurement indicator.

6. The interactive behavior extraction method according to claim 5, characterized in that: Determining the target numerical range of the interaction risk measurement indicator according to the minimum value, the first quartile value, the third quartile value, and the maximum value of the interaction risk measurement indicator includes: Subtracting the first quartile value from the third quartile value to obtain the interquartile range; determining the maximum of the minimum value and the difference between the first quartile value and 1.5 times the interquartile range as the lower numerical threshold of the interaction risk measurement indicator; Determine the minimum of the maximum value and the sum of the third quartile value and 1.5 times the interquartile range as the upper limit numerical threshold of the interaction risk measurement indicator; The lower numerical threshold, the upper numerical threshold, and numerical values ​​between the lower numerical threshold and the upper numerical threshold are determined as the target numerical range.

7. The interactive behavior extraction method according to claim 1, characterized in that: After adding the trajectory data of the target motor vehicle and the corresponding trajectory data of the target interactive object as an interactive behavior scene to the interactive behavior scene set of the target motor vehicle, the method further includes: Determining, based on the interaction behavior scenario, the category of the target motor vehicle, the type of the target motor vehicle's travel route, the interaction direction between the target motor vehicle and the target interaction object, and the category of the target interaction object; The interactive behavior scenario is classified according to the category of the target motor vehicle, the type of the driving route of the target motor vehicle, the interaction direction between the target motor vehicle and the target interactive object, and the category of the target interactive object.

8. The interactive behavior extraction method according to claim 1, characterized in that: The acquiring of a target trajectory dataset from the traffic participant trajectory dataset includes: Filtering out motor vehicle trajectory data with a trajectory duration greater than or equal to a target duration threshold from the traffic participant trajectory dataset to obtain a first motor vehicle trajectory dataset; performing normalization processing on each motor vehicle trajectory data in the first motor vehicle trajectory dataset to obtain a second motor vehicle trajectory dataset; performing clustering processing on the second motor vehicle trajectory dataset to obtain clusters having the same number of target centroids, wherein the clusters are subsets of the second motor vehicle trajectory dataset; According to the lane information of the target road section, the clusters of the target centroid number are screened to obtain at least one target cluster, and one target cluster corresponds to one target trajectory data set.

9. The interactive behavior extraction method according to any one of claims 1 to 8, characterized in that: The method of obtaining the traffic participant trajectory dataset of the target road section, after obtaining the target trajectory dataset from the traffic participant trajectory dataset, further includes: generating a motor vehicle trajectory image of a corresponding route of the target road section according to the target trajectory dataset, and extracting a driving area contour of the route from the motor vehicle trajectory image; The driving area outline is smoothed to obtain a target driving area outline, and the driving area included in the target driving area outline is filled to obtain a driving area map corresponding to the route.

10. The interactive behavior extraction method according to claim 9, characterized in that: The step of extracting the travel area contour of the route from the motor vehicle trajectory image comprises: performing a binarization process on the vehicle trajectory image to obtain a binarized image, and performing an image corrosion process on the binarized image to obtain a candidate binarized image; Detecting outliers in the candidate binary image using an outlier detection algorithm, and deleting the outliers in the candidate binary image to obtain a target binary image; Perform image expansion processing on the target binary image to obtain the driving area outline of the route.

11. The interactive behavior extraction method according to any one of claims 1 to 8, characterized in that: The step of obtaining a traffic participant trajectory dataset of a target road section includes: Obtain a target bird's-eye view image sequence corresponding to the traffic flow of the target road section; A traffic participant trajectory dataset of the target road section is obtained according to the target bird's-eye view image sequence.

12. The interactive behavior extraction method according to claim 11, characterized in that: The step of obtaining a traffic participant trajectory dataset of the target road section according to the target bird's-eye view image sequence includes: calling a preset traffic participant detection model to perform traffic participant detection on each bird's-eye view image in the target bird's-eye view image sequence, and obtaining a traffic participant detection result for each bird's-eye view image; determining static data of each traffic participant in each of the bird's-eye view images according to the traffic participant detection result of each of the bird's-eye view images; performing multi-target tracking on the traffic participants in each of the bird's-eye view images based on static data of each traffic participant in each of the bird's-eye view images, and assigning a tracking ID to each of the traffic participants that is successfully tracked; According to the tracking ID of each traffic participant, the static data of each traffic participant in each bird's-eye view image are associated to obtain a static data sequence corresponding to each tracking ID; Correcting the static data in the static data sequence corresponding to each tracking ID to obtain a target static data sequence corresponding to each tracking ID; Determining a speed data sequence of a traffic participant corresponding to each tracking ID according to a target static data sequence corresponding to each tracking ID; The target static data sequence and the speed data sequence corresponding to the same tracking ID are used as trajectory data of the corresponding traffic participant.

13. A computer device, characterized in that: The computer device includes a processor, a memory, a computer program stored in the memory and executable by the processor, and a data bus for realizing connection and communication between the processor and the memory, wherein when the computer program is executed by the processor, the interactive behavior extraction method according to any one of claims 1 to 12 is realized.

14. An interactive behavior extraction system based on a driving area map, characterized in that: include: Computer equipment, altitude maintenance equipment and image acquisition equipment, including: The height maintaining device is deployed on the target road section and is used to carry the image acquisition device, so that the image acquisition device can acquire a plurality of original bird's-eye view image sequences of the traffic flow of the target road section from a bird's-eye view perspective; The image acquisition device is mounted on the altitude maintaining device and is used to acquire a plurality of original bird's-eye view image sequences of the traffic flow of the target road section from a bird's-eye view perspective; The computer device is configured to obtain the plurality of original bird's-eye view image sequences and obtain a traffic participant trajectory dataset of the target road section based on the plurality of original bird's-eye view image sequences; The computer device is further configured to obtain a target trajectory dataset from the traffic participant trajectory dataset, wherein the target trajectory dataset includes trajectory data of a plurality of target motor vehicles with clear driving routes; The computer device is further configured to construct a driving area map on the target road segment based on the target trajectory dataset, wherein the driving area map is configured to represent the driving areas of the multiple target motor vehicles on the target road segment; The computer device is further configured to search for a plurality of potential interaction objects of the target motor vehicle in the driving area map corresponding to the driving route of the target motor vehicle based on the trajectory data of the target motor vehicle and the traffic participant trajectory dataset; The computer device is further configured to determine a value of an interaction risk measurement indicator between the target motor vehicle and each of the potential interaction objects based on the trajectory data of the target motor vehicle and the trajectory data of each of the potential interaction objects; The computer device is also used to determine the potential interactive object whose value of the interaction risk measurement index is within the target value range as the target interactive object of the target motor vehicle, and add the trajectory data of the target motor vehicle and the corresponding trajectory data of the target interactive object as interactive behavior scenes to the interactive behavior scene set of the target motor vehicle.

15. A storage medium for computer-readable storage, characterized in that: The storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the interactive behavior extraction method according to any one of claims 1 to 12.

Images