Vehicle driving scene identification method and device, equipment and storage medium

Abstract

The invention relates to a vehicle driving scene identification method and device, equipment and a storage medium, and relates to the technical field of vehicle control. The method is applied to a vehicle and comprises the steps that driving information of a target vehicle is acquired, and the driving information comprises target position information of the target vehicle; secondly, according to the target position information, the target lane line position where the target vehicle is located is determined from the multiple preset lane line positions, and the target lane line position is the preset lane line position, closest to the target vehicle, in the multiple preset lane line positions; then, according to front and rear lane information of the target lane line position, the driving scene of the target vehicle is determined, the front and rear lane information comprises the number of lanes of a front variable lane of the lane line position and the number of lanes of a rear variable lane of the lane line position, and the driving scene is a ramp shunting scene or a ramp converging scene.

Description

Technical Field

[0001] The present application relates to the field of vehicle control technology, and in particular to a method, device, equipment and storage medium for identifying a vehicle driving scene. Background Art

[0002] In recent years, with the rapid development of intelligent transportation and autonomous driving technologies, people have higher and higher requirements for the safety of autonomous driving vehicles in daily life. In order to ensure the safety of autonomous driving vehicles during driving, the driving scenes of other vehicles can be identified to change the driving operation of the vehicle.

[0003] At present, the recognition of a vehicle's driving scene is usually based on matching the full scene information with the vehicle's driving information to determine the vehicle's driving scene.

[0004] However, in the above technical solution, since the amount of information in the full scene information is large and some scene information may exist in the full scene information that has no impact on driving safety, the recognition speed of driving scenes that may easily affect driving safety (such as ramp scenes) is reduced, thereby reducing the recognition efficiency of driving scenes. Summary of the invention

[0005] The present application provides a method, device, equipment and storage medium for identifying vehicle driving scenes, so as to at least solve the technical problem of how to improve the recognition efficiency of driving scenes in the related art. The technical solution of the present application is as follows:

[0006] According to the first aspect involved in the present application, a method for identifying a vehicle driving scene is provided, which is applied to a vehicle, and includes: obtaining driving information of a target vehicle, the driving information including: target position information of the target vehicle. According to the target position information, a target lane line position where the target vehicle is located is determined from a plurality of preset lane line positions, and the target lane line position is a preset lane line position closest to the target vehicle among the plurality of preset lane line positions. According to the front and rear lane information of the target lane line position, the driving scene of the target vehicle is determined, and the front and rear lane information includes: the number of lanes that can be changed in front of the lane line position, and the number of lanes that can be changed in the rear of the lane line position, and the driving scene is a ramp diversion scene or a ramp merging scene.

[0007] According to the above technical means, by obtaining the vehicle's position information, the lane line position of the vehicle is determined, and based on the number of variable lanes in the front and rear directions of the lane line position, the ramp divergence / merging scene of the vehicle is determined to achieve recognition of the driving scene under the vehicle. In this way, by specifying and identifying driving scenes that are likely to affect driving safety, the recognition efficiency of driving scenes can be improved.

[0008] In a possible implementation, the front and rear lane information includes: the identification of all first lanes that are variable ahead, the identification of all second lanes that are variable behind, and the identification of adjacent lanes. The above-mentioned method of "determining the driving scene of the target vehicle based on the front and rear lane information of the target lane line position" includes: determining the first number of lanes other than adjacent lanes in all first lanes based on the identification of all first lanes and the identification of adjacent lanes. If the first number is greater than the first preset number threshold, the driving scene is determined to be a ramp diversion scene.

[0009] According to the above technical means, when the ramp diversion scenario is not identified at the lane line position of the target vehicle, the target vehicle is predicted in the direction of the subsequent lane to predict the ramp diversion scenario of the target vehicle at the subsequent lane line position.

[0010] In a possible implementation, the method further includes: if the first number is less than or equal to a first preset number threshold, determining a first lane line position in front of the target lane line position among a plurality of preset lane line positions, and determining a driving scenario based on front and rear lane information of the first lane line position.

[0011] According to the above technical means, by determining that the first number is less than or equal to the first preset number threshold, the recognition device can determine the first lane line position located in front of the target lane line position among multiple preset lane line positions, and then redetermine the driving scene based on the front and rear lane information of the first lane line position.

[0012] In a possible implementation, the method further includes: the first lane line position is adjacent to the target lane line position, or the number of lane lines between the first lane line position and the target lane line position is less than a second preset number threshold.

[0013] According to the above technical means, by predicting the driving scenario of the target vehicle at the subsequent lane line position, the first lane line position is adjacent to the target lane line position, or the relationship between the number of lane lines between the first lane line position and the target lane line position and a second preset number threshold, the number of predicted iterations can be determined.

[0014] In a possible implementation, the method of "determining the driving scene of the target vehicle based on the front and rear lane information of the target lane line position" further includes: determining a second number of lanes other than the adjacent lanes in all the second lanes based on the identifiers of all the second lanes and the identifiers of the adjacent lanes. If the second number is greater than the first preset number threshold, the driving scene is determined to be a ramp merging scene.

[0015] According to the above technical means, by determining the relationship between the second number and the first preset number threshold, when the second number is greater than the first preset number threshold, it is determined that the driving scenario of the target vehicle is a ramp merging scenario.

[0016] In a possible implementation, the method further includes: if the second number is less than or equal to a first preset number threshold, determining a first lane line position in front of the target lane line position among a plurality of preset lane line positions, and determining a driving scenario based on front and rear lane information of the first lane line position.

[0017] According to the above technical means, when the ramp merging scene is not identified at the lane line position where the target vehicle is located, the ramp merging scene of the target vehicle at the subsequent lane line position can be predicted based on the lane direction.

[0018] In a possible implementation, after the above-mentioned method of "determining the driving scene of the target vehicle based on the front and rear lane information of the target lane line position", the method also includes: determining the distance value between the target vehicle and the lane line position corresponding to the driving scene based on the target position information.

[0019] According to the above technical means, after identifying the ramp divergence / merging scenario, the distance between the vehicle's location and the divergence point / merging point is determined so that a response can be made to the ramp divergence / merging scenario within the driving time of the distance (such as behavioral decision-making for autonomous driving / assisted driving).

[0020] In a possible implementation manner, the driving information is the driving information of the target vehicle at a current moment, or the driving information is the driving information of the target vehicle at a future moment.

[0021] According to the above technical means, the driving information of the target vehicle can be the driving information of the target vehicle at the current moment, or the driving information of the target vehicle can be the driving information of the target vehicle at a future moment, which can improve the accuracy of parsing the driving information of the target vehicle.

[0022] In a possible implementation manner, the driving information further includes: a heading angle of the target vehicle.

[0023] According to the above technical means, by referring to the heading angle of the target vehicle in the process of responding to the identified ramp divergence / merging scenario, the accuracy of the response to the ramp divergence / merging scenario can be improved.

[0024] According to a second aspect provided by the present application, a vehicle driving scene recognition device is provided, which is applied to a vehicle, and the device includes: an acquisition module and a processing module.

[0025] The acquisition module is used to acquire the driving information of the target vehicle, and the driving information includes: the target position information of the target vehicle. The processing module is used to determine the target lane line position of the target vehicle from multiple preset lane line positions according to the target position information, and the target lane line position is the preset lane line position closest to the target vehicle among the multiple preset lane line positions. The processing module is also used to determine the driving scene of the target vehicle according to the front and rear lane information of the target lane line position, and the front and rear lane information includes: the number of lanes that can be changed in front of the lane line position, and the number of lanes that can be changed in the rear of the lane line position, and the driving scene is a ramp diversion scene or a ramp merging scene.

[0026] In a possible implementation, the processing module is specifically configured to determine a first number of lanes other than adjacent lanes in all first lanes according to the identifiers of all first lanes and the identifiers of adjacent lanes. The processing module is further configured to determine that the driving scenario is a ramp diversion scenario if the first number is greater than a first preset number threshold.

[0027] In a possible implementation, the processing module is further configured to determine a first lane line position located in front of the target lane line position among a plurality of preset lane line positions if the first number is less than or equal to a first preset number threshold. The processing module is further configured to determine a driving scenario based on front and rear lane information of the first lane line position.

[0028] In a possible implementation, the processing module is specifically configured to determine a second number of lanes other than adjacent lanes in all second lanes according to the identifiers of all second lanes and the identifiers of adjacent lanes. The processing module is further configured to determine that the driving scenario is a ramp merging scenario if the second number is greater than a first preset number threshold.

[0029] In a possible implementation, the processing module is specifically configured to determine a first lane line position located in front of the target lane line position among a plurality of preset lane line positions if the second number is less than or equal to a first preset number threshold. The processing module is further configured to determine a driving scenario based on front and rear lane information of the first lane line position.

[0030] In a possible implementation, the processing module is further configured to determine, based on the target position information, a distance value between the target vehicle and the lane line position corresponding to the driving scene.

[0031] According to a third aspect provided by the present application, an electronic device is provided, comprising: a processor and a memory for storing instructions executable by the processor, wherein the processor is configured to execute the instructions to implement a method of any possible implementation of the first aspect above.

[0032] According to the fourth aspect provided by the present application, a computer-readable storage medium is provided. When the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute a method of any possible implementation of the above-mentioned first aspect.

[0033] According to a fifth aspect provided by the present application, a computer program product is provided, the computer program product comprising computer instructions, and when the computer instructions are executed on an electronic device, the electronic device executes a method of any possible implementation of the first aspect above.

[0034] Therefore, the above technical features of the present application have the following beneficial effects:

[0035] (1) By acquiring the vehicle's position information, the lane line position of the vehicle is determined, and based on the number of variable lanes in the front and rear directions of the lane line position, the ramp divergence / merging scene of the vehicle is determined to achieve recognition of the driving scene under the vehicle. In this way, by specifying and identifying driving scenes that are likely to affect driving safety, the recognition efficiency of driving scenes can be improved.

[0036] (2) When the ramp diversion scenario is not identified at the lane line position of the target vehicle, the target vehicle is predicted in the direction of the subsequent lane to predict the ramp diversion scenario of the target vehicle at the subsequent lane line position.

[0037] (3) By determining that the first number is less than or equal to a first preset number threshold, the recognition device can determine a first lane line position in front of the target lane line position among multiple preset lane line positions, and then redetermine the driving scene based on the front and rear lane information of the first lane line position.

[0038] (4) The number of prediction iterations can be determined by, in the process of predicting the target vehicle's driving scenario at a subsequent lane line position, the first lane line position is adjacent to the target lane line position, or the number of lane lines spaced between the first lane line position and the target lane line position is related to a second preset number threshold.

[0039] (5) By determining the relationship between the second number and the first preset number threshold, when the second number is greater than the first preset number threshold, determining that the driving scenario of the target vehicle is a ramp merging scenario.

[0040] (6) When the ramp merging scene is not identified at the lane line position where the target vehicle is located, the ramp merging scene of the target vehicle at the subsequent lane line position can be predicted based on the lane direction.

[0041] (7) After identifying the ramp divergence / merging scenario, the distance between the vehicle location and the divergence point / merging point is determined so as to respond to the ramp divergence / merging scenario (such as the behavioral decision of autonomous driving / assisted driving) within the driving time of the distance.

[0042] (8) The driving information of the target vehicle may be the driving information of the target vehicle at the current moment, or the driving information of the target vehicle may be the driving information of the target vehicle at a future moment. This can improve the accuracy of parsing the driving information of the target vehicle.

[0043] (9) By referring to the heading angle of the target vehicle when responding to the identified ramp divergence / merging scenario, the accuracy of the response to the ramp divergence / merging scenario can be improved.

[0044] It should be noted that the technical effects brought about by any implementation method in the second to fifth aspects can refer to the technical effects brought about by the corresponding implementation method in the first aspect, and will not be repeated here.

[0045] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The drawings herein are incorporated into the specification and constitute a part of the specification, illustrate embodiments consistent with the present application, and together with the specification are used to explain the principles of the present application, and do not constitute an improper limitation on the present application.

[0047] Figure 1 is a schematic diagram of a communication system according to an exemplary embodiment;

[0048] Figure 2 is a flow chart of a method for identifying a vehicle driving scene according to an exemplary embodiment;

[0049] Figure 3 is a schematic diagram showing an example of a preset lane line position according to an exemplary embodiment;

[0050] Figure 4 is a schematic diagram showing an example of the distance between a vehicle and a preset lane line according to an exemplary embodiment;

[0051] Figure 5 is a schematic diagram showing an example of the number of variable lanes before and after a target lane line position according to an exemplary embodiment;

[0052] Figure 6 is a schematic diagram showing an example of front and rear lane information of a target lane line position according to an exemplary embodiment;

[0053] Figure 7 is a schematic diagram showing an example of traffic diversion scene recognition according to an exemplary embodiment;

[0054] Figure 8 is a schematic diagram showing an example of confluence scene recognition according to an exemplary embodiment;

[0055] Fig. 9 is a block diagram of a vehicle driving scene recognition device according to an exemplary embodiment;

[0056] Fig.10 It is a block diagram of an electronic device according to an exemplary embodiment. DETAILED DESCRIPTION

[0057] In order to enable ordinary persons in the art to better understand the technical solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0058] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the attached claims.

[0059] Before introducing in detail the vehicle driving scene recognition method of the embodiment of the present application, the implementation environment and application scenario of the embodiment of the present application are first introduced.

[0060] In the autonomous driving algorithm, in order to obtain the future driving trajectory of the target vehicle in advance and make decisions on the behavior of the vehicle earlier and better, a trajectory prediction method based on deep learning is used to reduce problems such as false braking and missed braking. In the process of testing the trajectory prediction model, when there are scenarios involving mixed traffic flow and multi-target interaction, the results obtained are often not very ideal due to the high complexity of the scenario. For example, many problems will be encountered in the diversion and confluence scenarios in the automatic on-ramp and off-ramp function, so it is necessary to analyze these scenarios separately and extract samples of these scenarios for fine-tuning on the trained model. However, the amount of full-scene data used for training is often relatively large, and checking them one by one is undoubtedly like looking for a needle in a haystack. How to efficiently identify the required scenes from hundreds of thousands, millions, or even tens of millions of sets of data and extract effective samples has become an urgent problem to be solved.

[0061] At present, in the existing technical solutions, the lane lines in the converging scene are usually separated and drawn based on the semantic segmentation method to solve the problem of lane line adhesion in the converging area. However, this method requires the training of a semantic segmentation model, which is not efficient for converging scene recognition.

[0062] In addition, an intelligent dispatching strategy based on deep reinforcement learning algorithm can also be adopted. However, this method focuses on dispatching vehicles more safely and efficiently, and does not mention how to identify ramp merging scenarios.

[0063] To sum up, since the amount of information in the full scene information is large and some scene information that has no impact on driving safety may exist in the full scene information, the recognition speed of driving scenes that may easily affect driving safety (such as ramp scenes) is reduced, thereby reducing the recognition efficiency of driving scenes.

[0064] In order to solve the above problems, the embodiment of the present application provides a method for identifying vehicle driving scenes (i.e., a method for identifying convergence and divergence scenes using a deep learning trajectory prediction model), which is applied to the scene recognition of autonomous driving, including: obtaining the vehicle's position information, determining the lane line position of the vehicle, and based on the number of variable lanes in the front and rear directions of the lane line position, further determining the ramp divergence / convergence scene of the vehicle, so as to realize the recognition of the driving scene under the vehicle. In this way, by specifying and identifying driving scenes that are likely to affect driving safety, the recognition efficiency of driving scenes can be improved.

[0065] In other words, the algorithm involved in the embodiment of the present application is simple and has high sample extraction efficiency. It does not require additional example resources. It can identify whether the vehicle is in a converging or diverging scenario by parsing map data and vehicle trajectory information based on existing data. It facilitates targeted training, testing, and analysis of trajectory prediction models, and can be widely used in converging and diverging scenario recognition.

[0066] The implementation environment of the embodiments of the present application is introduced below.

[0067] Figure 1 FIG. 1 is a schematic diagram of a communication system according to an exemplary embodiment. Figure 1 As shown, the communication system includes: a vehicle driving scene recognition device (such as recognition device 101) and a target vehicle 102. The recognition device 101 can perform wired / wireless communication with the target vehicle 102.

[0068] Specifically, the recognition device 101 can obtain the position information of the target vehicle 102, and determine the lane line position of the target vehicle 102 based on the position information. Then, the recognition device 101 can determine the number of variable lanes in the front and rear directions of the target vehicle 102 based on the lane line position of the target vehicle 102, and then determine the ramp divergence / merging scene where the target vehicle 102 is located, so as to realize the recognition of the driving scene of the target vehicle 102.

[0069] In the embodiment of the present application, the recognition device 101 may be a server. Alternatively, the recognition device 101 may be a terminal. Alternatively, the recognition device 101 may be a vehicle (such as a vehicle that performs automatic driving / assisted driving).

[0070] That is to say, the identification device 101 may be a third-party device other than the vehicle (ie, a server or a terminal), or may be the vehicle itself that requires vehicle driving scene identification.

[0071] Among them, when the identification device 101 is a server or a terminal, the identification device 101 can provide the driving scene of the identified target vehicle 102 to other vehicles in need (such as vehicles performing automatic driving / assisted driving), and when the identification device 101 is a vehicle, the identification device 101 can apply the driving scene of the identified target vehicle 102 locally.

[0072] It should be noted that the server can be a single physical server or a server cluster consisting of multiple servers. Alternatively, the server cluster can also be a distributed cluster. Alternatively, the server can also be a cloud server. The embodiments of the present application do not limit the specific implementation of the server.

[0073] The terminal may be a mobile phone, tablet computer, desktop, laptop, handheld computer, notebook computer, ultra-mobile personal computer (UMPC), netbook, etc. with transceiver functions. The present application does not impose any special restrictions on the specific form of the terminal. It may interact with the user through one or more methods such as keyboard, touch pad, touch screen, remote control, voice interaction or handwriting device.

[0074] For ease of understanding, the vehicle driving scene recognition method provided by the present application is specifically introduced below with reference to the accompanying drawings. Figure 2 is a flow chart of a method for identifying a vehicle driving scene according to an exemplary embodiment. Figure 2 As shown, the method comprises the following steps:

[0075] S201. Obtain driving information of a target vehicle.

[0076] Among them, when the identification device is a vehicle, the target vehicle can be any vehicle on the road except the vehicle itself, or, when the identification device is a server or terminal (i.e., a third-party device other than the vehicle), the target vehicle can be any vehicle on the road.

[0077] It should be noted that the following uses a vehicle as an example to introduce the vehicle driving scene recognition method provided in the embodiment of the present application.

[0078] In an embodiment of the present application, the driving information may include: target position information of the target vehicle.

[0079] Among them, the target position information can be the position information of the target vehicle in the global coordinate system (such as longitude and latitude information), or the target position information can be the position information of the target vehicle compared to the identification device (that is, the position information in the coordinate system with the identification device as the origin).

[0080] It should be noted that the identification device can actively obtain the driving information of the target vehicle, or the identification device can passively receive the driving information of the target vehicle.

[0081] As a possible implementation method, taking the example of the identification device actively acquiring the driving information of the target vehicle, the identification device can acquire the driving information of the target vehicle through multiple sensors (such as sensors on the vehicle or sensors on both sides of the road).

[0082] Optionally, the identification device may send a query message for driving information to the target vehicle, so that the target vehicle can feed back the driving information recorded by itself to the identification device, so that the identification device can obtain the driving information of the target vehicle.

[0083] As another possible implementation, taking the example of the identification device passively receiving the driving information of the target vehicle, the target vehicle may actively report the driving information of the target vehicle to the identification device, so that the identification device can obtain the driving information of the target vehicle.

[0084] In some embodiments, the driving information of the target vehicle may be the driving information of the target vehicle at the current moment, or the driving information of the target vehicle may be the driving information of the target vehicle at a future moment.

[0085] That is to say, for the vehicle's driving scene recognition, not only can the current driving scene be recognized based on the current real-time information, but the future driving scene can also be recognized based on the future predicted information.

[0086] S202: Determine a target lane line position of a target vehicle from a plurality of preset lane line positions according to the target position information.

[0087] Among them, a preset lane line position can indicate a lane line section in a lane.

[0088] For example, Figure 3 As shown, the multiple preset lane line positions may include: position A, position B, position C, position D and position E. Among them, position A indicates lane line A in lane 1, position B indicates lane line B in lane 1, position C indicates lane line C in lane 2, position D indicates lane line D in lane 2, and position E indicates lane line E in lane 3.

[0089] In an embodiment of the present application, the target lane line position is a preset lane line position closest to the target vehicle among multiple preset lane line positions.

[0090] As a possible implementation, preset position information corresponding to each preset lane line position is pre-stored in the recognition device. The recognition device can determine the first distance value between the target vehicle and each preset lane line position according to the target position information and the multiple preset position information, and based on the first distance value corresponding to each preset lane line position, select the preset lane line position corresponding to the minimum first distance value from the multiple preset lane line positions as the lane line position where the target vehicle is located, so as to obtain the target lane line position.

[0091] It should be noted that the preset position information may be the position information of the center point of the corresponding lane line, or the preset position may include the position information of both end points of the corresponding lane line.

[0092] Wherein, when the preset position information is the position information of the center point of the corresponding lane line, the first distance value is the distance value between the target vehicle and the center point position of the preset lane line.

[0093] In the case where the preset position includes the position information of the two end points of the corresponding lane line, the first distance value includes the second distance values ​​between the target vehicle and the two end points of the preset lane line, that is, one preset lane line position corresponds to two second distance values, then the target lane line position is the preset lane line position corresponding to the smallest second distance value among all the second distance values.

[0094] For example, Figure 4As shown, the multiple preset lane line positions include: position A of lane line 1, position B of lane line 2, position C of lane line 3, and position D of lane line 4. Among them, the two second distance values ​​corresponding to position A are respectively the distance value between endpoint a and the vehicle, and the distance value between endpoint b and the vehicle, the two second distance values ​​corresponding to position B are respectively the distance value between endpoint c and the vehicle, and the distance value between endpoint d and the vehicle, the two second distance values ​​corresponding to position C are respectively the distance value between endpoint e and the vehicle, and the distance value between endpoint f and the vehicle, and the two second distance values ​​corresponding to position D are respectively the distance value between endpoint g and the vehicle, and the distance value between endpoint h and the vehicle.

[0095] S203: Determine the driving scene of the target vehicle according to the front and rear lane information of the target lane line position.

[0096] Among them, the front and rear lane information may include: the number of variable lanes in front of the lane line position, and the number of variable lanes behind the lane line position.

[0097] For example, Figure 5 As shown, taking the target lane line position as the position of lane line 1 in lane 2 as an example, the front and rear lane information of the target lane line position may include: the number of lanes 1 of the front variable lane, and the number of lanes 2 of the rear variable lane. Taking the target lane line position as the position of lane line 2 in lane 2 as an example, the front and rear lane information of the target lane line position may include: the number of lanes 2 of the front variable lane, and the number of lanes 1 of the rear variable lane.

[0098] It should be noted that, in the embodiment of the present application, the driving scenario determined based on the front and rear lane information may be a ramp divergence scenario or a ramp merging scenario.

[0099] Among them, when the number of non-adjacent lanes in front that can be changed indicated by the front and rear lane information is greater than or equal to 1, the driving scene is a ramp diversion scene; when the number of non-adjacent lanes in the rear that can be changed indicated by the front and rear lane information is greater than or equal to 1, the driving scene is a ramp merging scene.

[0100] For example, in combination with the above Figure 5 The positional relationship between the multiple lane lines shown is, for example, that the target lane line position is the position of lane line 1 in lane 2, then the driving scenario of the target vehicle is a ramp merging scenario, and that the target lane line position is the position of lane line 2 in lane 2, then the driving scenario of the target vehicle is a ramp diverging scenario.

[0101] It can be understood that by obtaining the vehicle's position information, determining the lane line position of the vehicle, and based on the number of variable lanes in the front and rear directions of the lane line position, the ramp divergence / merging scene of the vehicle is determined to achieve recognition of the driving scene under the vehicle. In this way, by specifying and identifying driving scenes that are likely to affect driving safety, the recognition efficiency of driving scenes can be improved.

[0102] As a possible implementation manner, the front and rear lane information may include: identifications of all first lanes of the front variable lanes, identifications of all second lanes of the rear variable lanes, and identifications of adjacent lanes.

[0103] For example, in combination with the above Figure 5 The positional relationship between the multiple lane lines shown, taking the position of lane line 2 in lane 2 as an example, the identification of all first lanes in the front and rear lane information includes: lane 1 and lane 3, the identification of all second lanes in the front and rear lane information includes: lane 1, and the identification of the adjacent lane in the front and rear lane information is lane 1.

[0104] In the embodiment of the present application, when the recognition device determines the driving scene of the target vehicle according to the front and rear lane information of the target lane line position, the recognition device can determine the other lanes in all the first lanes except the adjacent lanes (such as the above-mentioned Figure 5 A first number of lanes 3) corresponding to the middle lane line 2, and a magnitude relationship between the first number and a first preset number threshold (such as 0) is determined.

[0105] Among them, if the recognition device determines that the first number is greater than the first preset number threshold, the recognition device can determine that the driving scenario of the target vehicle is a ramp diversion scenario.

[0106] Optionally, if the recognition device determines that the first number is less than or equal to a first preset number threshold, the recognition device can determine a first lane line position among multiple preset lane line positions that is located in front of the target lane line position, and then redetermine the driving scene based on the front and rear lane information of the first lane line position.

[0107] It should be noted that, for the process of the recognition device determining the driving scene based on the front and rear lane information of the first lane line position, reference can be made to the above introduction to the recognition device determining the driving scene based on the front and rear lane information of the target lane line position, which will not be repeated here.

[0108] For example, Figure 6As shown, taking the target lane line position as the position of lane line 1 in lane 2 as an example, the front and rear lane information of the target lane line position may include: lane 1 (i.e., the mark of the first lane), lane 1 (i.e., the mark of the adjacent lane). If the first preset number threshold is 0, the position of lane line 2 in lane 2 is determined to be the first lane line position, and the front and rear lane information of the first lane line position may include: lane 3 (i.e., the mark of the first lane), lane 1 (i.e., the mark of the adjacent lane).

[0109] That is to say, when the ramp diversion scenario is not identified at the lane line position where the target vehicle is located, the recognition device can predict the ramp diversion scenario of the target vehicle at the subsequent lane line position in the lane direction.

[0110] In an embodiment of the present application, in the process of the above-mentioned recognition device redetermining the driving scene based on the front and rear lane information of the first lane line position, if the recognition device determines that the first number corresponding to the first lane line position is also less than or equal to the first preset number threshold, then the second lane line position located in front of the first lane line position among multiple preset lane line positions is determined, and then the driving scene is redetermined based on the front and rear lane information of the second lane line position.

[0111] That is to say, the recognition device can predict the ramp diversion scenario in the lane where the target vehicle is located through multiple position iterations until it is predicted that the target vehicle has a ramp diversion scenario at a subsequent lane line position.

[0112] It should be noted that the number of lane lines between the first lane line position and the target lane line position is less than the second preset number threshold, or the first lane line position is adjacent to the target lane line position (ie, the second preset number threshold is 1).

[0113] That is to say, in the process of predicting the driving scenario of the target vehicle at the subsequent lane line position, the recognition device can determine the number of prediction iterations based on the second preset number threshold.

[0114] In summary, the recognition (or prediction) of the ramp divergence scenario can be determined based on the number of variable lanes in front of the lane line position. Similarly, the recognition (or prediction) of the ramp merging scenario can be determined based on the number of variable lanes behind the lane line position. The specific implementation examples are shown below.

[0115] In some embodiments, when the recognition device determines the driving scene of the target vehicle based on the front and rear lane information of the target lane line position, the recognition device can determine the other lanes in all the second lanes except the adjacent lanes (such as the above-mentioned Figure 5A second number of lanes 4) corresponding to the middle lane line 1, and a magnitude relationship between the second number and a first preset number threshold (such as 0) is determined.

[0116] If the recognition device determines that the second number is greater than the first preset number threshold, the recognition device may determine that the driving scenario of the target vehicle is a ramp merging scenario.

[0117] Optionally, if the recognition device determines that the second number is less than or equal to a first preset number threshold, the recognition device can determine a first lane line position among multiple preset lane line positions that is located in front of the target lane line position, and then redetermine the driving scene based on the front and rear lane information of the first lane line position.

[0118] That is to say, when the ramp merging scene is not recognized at the lane line position where the target vehicle is located, the recognition device can predict the ramp merging scene of the target vehicle at the subsequent lane line position in the lane direction.

[0119] In some embodiments, after the recognition device determines the driving scene of the target vehicle based on the front and rear lane information of the target lane line position (ie, S203), the recognition device can determine the distance value between the target vehicle and the lane line position corresponding to the driving scene based on the target position information.

[0120] That is to say, after identifying the ramp divergence / merging scenario, the recognition device can determine the distance between the vehicle's location and the divergence point / merging point, so as to respond to the ramp divergence / merging scenario (such as behavioral decisions of automatic driving / assisted driving) within the driving time of the distance.

[0121] Optionally, the driving information of the target vehicle may also include: a heading angle of the target vehicle.

[0122] That is to say, when the recognition device responds to the recognized ramp divergence / merging scenario, it can also refer to the heading angle of the target vehicle to improve the accuracy of the response to the ramp divergence / merging scenario.

[0123] The following is an introduction to the vehicle driving scene recognition method provided by the embodiment of the present application in conjunction with a specific example, including the following steps 1 to 7.

[0124] Step 1. Use Python to parse all lane data (lanes_data) from a real vehicle data. Define the data structure of lanes_data as a dictionary (dict), whose key is the number of each lane (lane_id), and whose value is the data corresponding to a single lane (lane_data). The single lane data contains the following information: the number of the current lane (id), the number of the subsequent lane of the current lane (successor_id, that is, the identification of all the first lanes of the variable lane in front), the number of the lane before the current lane (predecessor_id, that is, the identification of all the second lanes of the variable lane in the back), the number of the sibling lane of the current lane (neighbor_id, that is, the identification of the adjacent lane), and the point set of the global coordinate system of the lane (points_utm), including the ordinate (point_x) and the abscissa (point_y).

[0125] Step 2: Obtain all target trajectory data (objs_hist_traj_data) from sensor fusion. The ego vehicle (such as a vehicle performing autonomous driving / assisted driving) can predict the future trajectory of the target vehicle based on the target trajectory information. The historical trajectory data is 2s, the sampling frequency is 5, and the target trajectory data contains 10 historical frames of data and current frame data, a total of 11 frames, of which the last frame is the current frame. The data structure of the target trajectory data is also defined as a dictionary, which contains the following information: target ID number (track_id), coordinate system (universal transverse mercator, UTM) coordinate trajectory point of the target vehicle in the ego vehicle coordinate system (history_xy), including the longitudinal coordinate (target_x) and the lateral coordinate (target_y) in the UTM coordinate system, and the heading angle (history_heading) of the target vehicle's heading relative to the direction of the ego vehicle in the ego vehicle coordinate system.

[0126] Step 3: Traverse all targets in a time window data, set the target vehicle as i, the current frame data as -1, and calculate the current position (target_current_xy) and heading angle (target_current_heading) of target i. The current position of target i is the coordinate trajectory point of the target vehicle in the ego vehicle coordinate system when the target trajectory data is the current frame (i.e., target_current_xy[i] = objs_hist_traj_data['history_xy'][i][-1]), and the heading angle of target i is the heading angle of the target heading relative to the direction of the ego vehicle in the ego vehicle coordinate system when the target trajectory data is the current frame (i.e., target_current_heading[i] = objs_hist_traj_data['history_heading'][i][-1]).

[0127] Step 4. According to target_current_xy[i] and lanes_data, the lane number (target_current_laneid[i]) of target i and the closest distance (min_dist[i]) of target i to the global coordinate system point set of the jth lane line are calculated to determine the lane line where the target is located. The lane line where target i is located and the closest distance of target i to the global coordinate system point set of the jth lane line are initialized, and target_current_laneid[i] = 'unknown', min_dist[i] = 10000 are initialized. First, determine whether there is a lane line in the current window data. If the map parsing is incorrect and there is no lane line in the current window, keep initializing target_current_laneid[i] = 'unknown', min_dist[i] = 10000. If there is a lane line, traverse all lane lines and determine whether the length of the global coordinate system point set of the jth lane line is equal to 0. If the length of the global coordinate system point set of the jth lane line is equal to 0, skip the lane line. If the length of the global coordinate system point set of the j-th lane line is not equal to 0, then the closest distance min_dist[i] of the target i relative to the global coordinate system point set of the j-th lane line is calculated.

[0128] By traversing all points on a single lane line, the distance from target i to the kth point of the jth lane line is calculated (i.e., dist[j][k] = sqrt((point_x[j][k] - target_x[i])^2 + (point_y[j][k] - target_y[i])^2)).

[0129] Starting from the second point of the global coordinate system point set of the j-th lane line, a line is formed by the two points before and after, and the length of the n-th line (k>1) of the j-th lane line is calculated (i.e., dist2nextpoint[j][n] = sqrt((point_x(k)-point_x(k-1))^2+(point_y(k)-point_y(k-1))^2).

[0130] If the distance between two points on the jth lane is equal to 0, skip this point and continue, so that all dist[j][k] are different. If dist[j][k] is less than min_dist[i], assign dist[j][k] to min_dist[i] to find the distance from target i to the closest point on the jth lane. Find the lane closest to target i, and target_current_laneid[i] is equal to the lane number lane_id of that lane.

[0131] Step 5: Identify the diversion scene, such as Figure 7 As shown. According to the current lane line number target_current_laneid[i] of target i, the successor relationship of the current lane line is determined. First, determine whether all lane line data lanes_data contains target_current_laneid[i]. If lanes_data does not contain target_current_laneid[i], the branch point branch_point is equal to 0. Otherwise, continue to judge. If the number of subsequent lane line numbers successor_id[i] of the current lane line target_current_laneid[i] of target i is greater than 1, it means that there is a diversion, and the global coordinate branch_point of the diversion point is equal to the last point in the global coordinate system point set of the current lane line of target i. If the number of subsequent lane line numbers successor_id[i] of the current lane line target_current_laneid[i] of target i is equal to 1, then continue to search along this subsequent lane line number successor_id[i]. At this time, target_current_laneid[i] is equal to successor_id[i]. If the number of subsequent lane line numbers successor_id[i] of the current lane line target_current_laneid[i] where target i is located is equal to 0, it means that there is no divergence point.

[0132] Step 6: Identify the confluence scene, such as Figure 8As shown. According to the current lane line number (target_current_laneid[i]) of target i, the successor relationship of the current lane line of target i is determined. Determine whether (target_current_laneid[i]) is included in all lane line data (lanes_data). If lanes_data does not contain target_current_laneid[i], the confluence point merge_point is equal to 0. Otherwise, continue to judge. If the number of subsequent lane line numbers successor_id[i] of the current lane line target_current_laneid[i] of target i is greater than 1 or equal to 0, it means that there is no confluence point merge_point, that is, merge_point is equal to 0. If the number of subsequent lane line numbers successor_id[i] of the current lane line target_current_laneid[i] of target i is equal to 1, then continue to search forward along this subsequent lane line number successor_id[i]. At this time, target_current_laneid[i] is equal to successor_id[i]. If the number of predecessor lane line numbers predecessor_id[i] of the current lane line target_current_laneid[i] where target i is located is greater than 1, it means that the confluence point is found, and the global coordinates merge_point of the confluence point are equal to the first point in the global coordinate system point set of the current lane line.

[0133] Step 7: Calculate the distance between the diverging point and the converging point.

[0134] The global coordinates of the branch point (branch_point) and the merge point (merge_point) are obtained through steps 5 and 6 respectively. According to the current position (target_current_xy) and heading angle (target_current_heading) of target i obtained in step 3, the distance of the branch point or merge point of target i can be obtained by converting the global coordinate system to the vehicle coordinate system.

[0135] The above mainly introduces the solution provided by the embodiment of the present application from the perspective of the method. In order to achieve the above functions, the identification device or equipment of the vehicle driving scene includes a hardware structure and / or software module corresponding to the execution of each function. It should be easily appreciated by those skilled in the art that, in combination with the units and algorithm steps of each example described in the embodiments disclosed herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0136] The embodiment of the present application can divide the functional modules of the recognition device or equipment for vehicle driving scenes according to the above method. For example, the recognition device or equipment for vehicle driving scenes can include various functional modules corresponding to the various functional divisions, or two or more functions can be integrated into one processing module. The above integrated modules can be implemented in the form of hardware or in the form of software functional modules. It should be noted that the division of modules in the embodiment of the present application is schematic and is only a logical functional division. There may be other division methods in actual implementation.

[0137] Fig. 9 A block diagram of a vehicle driving scene recognition device according to an exemplary embodiment is shown, wherein the vehicle driving scene recognition device is applied to a vehicle and is used to perform Figure 2 The vehicle driving scene recognition device 900 includes: an acquisition module 901 and a processing module 902.

[0138] The acquisition module 901 is used to acquire the driving information of the target vehicle, and the driving information includes: the target position information of the target vehicle. The processing module 902 is used to determine the target lane line position of the target vehicle from multiple preset lane line positions according to the target position information, and the target lane line position is the preset lane line position closest to the target vehicle among the multiple preset lane line positions. The processing module 902 is also used to determine the driving scene of the target vehicle according to the front and rear lane information of the target lane line position, and the front and rear lane information includes: the number of lanes that can be changed in front of the lane line position, and the number of lanes that can be changed in the rear of the lane line position. The driving scene is a ramp diversion scene or a ramp merging scene.

[0139] In a possible implementation, the processing module 902 is specifically configured to determine a first number of lanes other than adjacent lanes in all first lanes according to the identifiers of all first lanes and the identifiers of adjacent lanes. The processing module 902 is further configured to determine that the driving scenario is a ramp diversion scenario if the first number is greater than a first preset number threshold.

[0140] In a possible implementation, the processing module 902 is further configured to determine a first lane line position located in front of the target lane line position among the plurality of preset lane line positions if the first number is less than or equal to a first preset number threshold. The processing module 902 is further configured to determine a driving scenario based on the front and rear lane information of the first lane line position.

[0141] In a possible implementation, the processing module 902 is specifically configured to determine a second number of lanes other than adjacent lanes in all second lanes according to the identifiers of all second lanes and the identifiers of adjacent lanes. The processing module 902 is further configured to determine that the driving scenario is a ramp merging scenario if the second number is greater than a first preset number threshold.

[0142] In a possible implementation, the processing module 902 is specifically configured to determine a first lane line position located in front of the target lane line position among the plurality of preset lane line positions if the second number is less than or equal to the first preset number threshold. The processing module 902 is further configured to determine a driving scenario based on the front and rear lane information of the first lane line position.

[0143] In a possible implementation, the processing module 902 is further configured to determine, according to the target position information, a distance value between the target vehicle and the lane line position corresponding to the driving scene.

[0144] Regarding the device in the above embodiment, the specific manner in which each unit performs the operation has been described in detail in the embodiment of the method, and will not be elaborated here.

[0145] Fig.10 A block diagram of an electronic device is shown according to an exemplary embodiment. Fig.10 As shown, the electronic device 1000 includes but is not limited to: a processor 1001 and a memory 1002 .

[0146] The memory 1002 is used to store executable instructions of the processor 1001. It can be understood that the processor 1001 is configured to execute instructions to implement the vehicle driving scene recognition method in the above embodiment.

[0147] It should be noted that those skilled in the art can understand that Fig.10 The electronic device structure shown in the figure does not constitute a limitation on the electronic device, and the electronic device may include Fig.10 More or fewer components may be shown, or certain components may be combined, or the components may be arranged differently.

[0148] The processor 1001 is the control center of the electronic device. It uses various interfaces and lines to connect various parts of the entire electronic device. By running or executing software programs and / or modules stored in the memory 1002, and calling data stored in the memory 1002, it performs various functions of the electronic device and processes data, thereby monitoring the electronic device as a whole. The processor 1001 may include one or more processing units. Optionally, the processor 1001 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface, and application programs, and the modem processor mainly processes wireless communications. It is understandable that the above-mentioned modem processor may not be integrated into the processor 1001.

[0149] The memory 1002 may be used to store software programs and various data. The memory 1002 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required by at least one functional module (such as a processing unit), etc. In addition, the memory 1002 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0150] In an exemplary embodiment, a computer-readable storage medium including instructions is also provided, such as a memory 1002 including instructions. The above instructions can be executed by the processor 1001 of the electronic device 1000 to implement the vehicle driving scene recognition method in the above embodiment.

[0151] In actual implementation, Fig. 9 The functions of the acquisition module 901 and the processing module 902 can be obtained by Fig.10 The processor 1001 in the memory 1002 calls the computer program stored in the memory 1002. The specific execution process can refer to the description of the vehicle driving scene recognition method in the above embodiment, which will not be repeated here.

[0152] Optionally, the computer-readable storage medium may be a non-temporary computer-readable storage medium, for example, the non-temporary computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0153] In an exemplary embodiment, the embodiment of the present application also provides a computer program product including one or more instructions, and the one or more instructions can be executed by a processor of an electronic device to complete the vehicle driving scene recognition method in the above embodiment.

[0154] It should be noted that when the instructions in the above-mentioned computer-readable storage medium or one or more instructions in the computer program product are executed by the processor of the electronic device, the various processes of the embodiment of the above-mentioned vehicle driving scene recognition method are implemented, and the same technical effect as the above-mentioned vehicle driving scene recognition method can be achieved. To avoid repetition, they will not be repeated here.

[0155] Through the description of the above implementation methods, technical personnel in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0156] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of modules or units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0157] The units described as separate components may or may not be physically separated, and the components shown as units may be one physical unit or multiple physical units, that is, they may be located in one place or distributed in multiple different places. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0158] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0159] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions to enable a device (which can be a single-chip microcomputer, chip, etc.) or a processor (processor) to perform all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard drives, ROM, RAM, magnetic disks or optical disks.

[0160] The above are only specific implementations of the present application, but the protection scope of the present application is not limited thereto, and any changes or substitutions within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for identifying a vehicle driving scene, characterized in that: The method comprises: Acquire driving information of a target vehicle, wherein the driving information includes: target position information of the target vehicle; According to the target position information, determine the target lane line position of the target vehicle from a plurality of preset lane line positions, wherein the target lane line position is the preset lane line position closest to the target vehicle among the plurality of preset lane line positions; The driving scenario of the target vehicle is determined based on the front and rear lane information of the target lane line position, the front and rear lane information includes: the number of lanes that can be changed into lanes in front of the lane line position, the number of lanes that can be changed into lanes behind the lane line position, and the driving scenario is a ramp divergence scenario or a ramp merging scenario.

2. The method according to claim 1, characterized in that The front and rear lane information includes: the marks of all first lanes that can be changed in front, the marks of all second lanes that can be changed in the rear, and the marks of adjacent lanes; The determining the driving scene of the target vehicle according to the front and rear lane information of the target lane line position includes: Determine, according to the identifiers of all the first lanes and the identifiers of the adjacent lanes, a first number of lanes other than the adjacent lanes in the all the first lanes; If the first number is greater than a first preset number threshold, the driving scenario is determined to be the ramp diversion scenario.

3. The method according to claim 2, characterized in that The method further comprises: If the first number is less than or equal to the first preset number threshold, determining a first lane line position located in front of the target lane line position among the plurality of preset lane line positions; The driving scene is determined according to the front and rear lane information of the first lane line position.

4. The method according to claim 3, characterized in that The first lane line position is adjacent to the target lane line position, or the number of lane lines between the first lane line position and the target lane line position is less than a second preset number threshold.

5. The method according to claim 2, characterized in that: The determining the driving scene of the target vehicle according to the front and rear lane information of the target lane line position also includes: Determining a second number of lanes other than the adjacent lanes among all the second lanes according to the identifiers of all the second lanes and the identifiers of the adjacent lanes; If the second number is greater than the first preset number threshold, the driving scenario is determined to be the ramp merging scenario.

6. The method according to claim 5, characterized in that The method further comprises: If the second number is less than or equal to the first preset number threshold, determining a first lane line position in the plurality of preset lane line positions that is located ahead of the target lane line position; The driving scene is determined according to the front and rear lane information of the first lane line position.

7. The method according to any one of claims 1 to 6, characterized in that After determining the driving scene of the target vehicle according to the front and rear lane information of the target lane line position, the method further includes: According to the target position information, a distance value between the target vehicle and the lane line position corresponding to the driving scene is determined.

8. The method according to any one of claims 1 to 6, characterized in that The driving information is the driving information of the target vehicle at a current moment, or the driving information is the driving information of the target vehicle at a future moment.

9. The method according to any one of claims 1 to 6, characterized in that: The driving information also includes: a heading angle of the target vehicle.

10. A vehicle driving scene recognition device, characterized in that: Applied to a vehicle, the device comprises: An acquisition module, used to acquire the driving information of the target vehicle, the driving information including: target position information of the target vehicle; A processing module, used to determine the target lane line position of the target vehicle from a plurality of preset lane line positions according to the target position information; The processing module is further used to determine the driving scene of the target vehicle according to the front and rear lane information of the target lane line position.

11. An electronic device, characterized in that: include: processor; A memory for storing the processor executable instructions; wherein the processor is configured to execute the instructions to implement the method according to any one of claims 1 to 6.

12. A computer-readable storage medium, characterized in that: When the computer-executable instructions stored in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device can perform the method as claimed in any one of claims 1 to 6.

13. A computer program product, characterized in that The computer program product comprises computer program instructions which, when executed, implement the method according to any one of claims 1 to 6.

Images