CIPV vehicle identification method and device, vehicle and storage medium

Abstract

The invention relates to the technical field of vehicles, in particular to a CIPV recognition method and device, a vehicle and a storage medium, and the method comprises the steps: obtaining the surrounding environment information and navigation information of a current vehicle, recognizing the driving position of the current vehicle, and if the current vehicle is located at a main road section and is close to a ramp road section, collecting the current scene information of the ramp road section, and preprocessing the current scene information, inputting the preprocessed current scene information into a preset vehicle classification model to obtain a vehicle in the main road section, and determining a CIPV vehicle according to the vehicle in the main road section. Therefore, the problems of wrong braking of the vehicle, active lane changing driving and the like caused by the fact that the vehicle with a long transverse distance is easily identified as the CIPV by mistake are solved, and non-CIPV vehicles on the ramp are identified by training the image features of the ramp scene through the navigation information, the limitation of the transverse distance threshold value and the longitudinal distance threshold value of the lane line, so that the error identification rate of the CIPV vehicles is reduced.

Description

Technical Field

[0001] The present application relates to the field of vehicle technology, and in particular to a method, device, vehicle and storage medium for identifying a CIPV (Closest In-Path Moveable Target) vehicle. Background Art

[0002] The entrance ramps of highways and urban elevated roads are bottleneck areas that affect traffic flow and capacity. Usually, the speed is adjusted to adapt to traffic conditions based on the adaptive cruise control system. During vehicle driving, this function will detect whether there is a target vehicle on the road ahead, and adjust its own speed based on the detection results and the status of the target vehicle.

[0003] In related technologies, adaptive cruise control systems rely on forward-looking cameras to identify objects on the road.

[0004] However, when the vehicle approaches the ramp, other vehicles may enter the field of view of the forward-looking camera. After the adaptive cruise control function is turned on, vehicles that are laterally far away may be easily misidentified as CIPVs, which may easily lead to incorrect braking of the vehicle or even active lane changes, thus affecting driving comfort and safety, which urgently needs to be solved. Summary of the invention

[0005] The present application provides a CIPV vehicle identification method, device, vehicle and storage medium to solve the problem that when other vehicles approach a ramp, a vehicle that is laterally far away is easily misidentified as a CIPV, thereby causing the vehicle to brake incorrectly or even actively change lanes.

[0006] The first embodiment of the present application provides a method for identifying a CIPV vehicle, comprising the following steps:

[0007] Obtain the surrounding environment information and navigation information of the current vehicle;

[0008] Identifying the current driving position of the vehicle based on the surrounding environment information and the navigation information, and determining whether the current driving position of the vehicle is in a main road section and whether the current vehicle is close to a ramp section;

[0009] If the current vehicle is on the main road section and the current vehicle is close to the ramp section, the current scene information of the ramp section is collected, the current scene information is preprocessed, and the preprocessed current scene information is input into a preset vehicle classification model to obtain the vehicle on the main road section, and the CIPV vehicle is determined based on the vehicle on the main road section.

[0010] According to one embodiment of the present application, the acquiring of the surrounding environment information and navigation information of the current vehicle includes:

[0011] Determining whether new surrounding environment information and navigation information of the current vehicle are received within a preset time;

[0012] If the new navigation information of the current vehicle is not received within the preset time, the surrounding environment information and navigation information of the current vehicle within the last preset time are sent; otherwise, the new surrounding environment information and navigation information of the current vehicle are sent.

[0013] According to one embodiment of the present application, before inputting the preprocessed current scene information into the preset vehicle classification model, the method further includes:

[0014] Acquire a public dataset of actual scene information, wherein the public dataset of actual scene information includes multiple image information of ramp road sections;

[0015] Preprocess each image information, annotate each preprocessed image information, and classify the annotated image information according to a preset ratio to obtain a training data set and a test data set of the image information;

[0016] Based on a preset deep learning algorithm, the training data set is trained to obtain an initial vehicle classification model, and the initial vehicle classification model is tested using the test data set. If the test result meets the preset test conditions, the initial vehicle classification model is used as the preset vehicle classification model.

[0017] According to one embodiment of the present application, after testing the initial vehicle classification model using the test data set, the method further includes:

[0018] If the test result does not meet the preset test condition, the preset ratio is updated, and a new training data set and a new test data set are divided according to the updated ratio. Based on the preset deep learning algorithm, the new training data set is retrained until the new test result meets the preset test condition, thereby obtaining the preset vehicle classification model.

[0019] According to one embodiment of the present application, determining the CIPV vehicle based on the vehicles on the main road section includes:

[0020] Obtaining non-CIPV vehicles on the ramp section according to an output result of the preset vehicle classification model;

[0021] Based on the vehicles on the main road section, the non-CIPV vehicles, and the lane line information on the main road section of the current vehicle, output the driving speed, yaw rate, and lane line information of the current vehicle, and perform risk judgment on the driving speed, yaw rate, and lane line information of the current vehicle to obtain a predicted driving trajectory of the current vehicle according to the risk judgment result;

[0022] Based on a preset test cycle, the lateral distance and longitudinal distance between the current vehicle and the target vehicle are calculated, and the CIPV vehicle is obtained based on a preset lateral distance threshold and a preset longitudinal distance threshold.

[0023] According to the identification method of the CIPV vehicle of the embodiment of the present application, the surrounding environment information and navigation information of the current vehicle are obtained and the driving position of the current vehicle is identified. If the current vehicle is in the main road section and close to the ramp section, the current scene information of the ramp section is collected, the current scene information is preprocessed, and the preprocessed current scene information is input into the preset vehicle classification model to obtain the vehicle in the main road section, and the CIPV vehicle is determined based on the vehicle in the main road section. In this way, the problem of misidentifying a vehicle with a long lateral distance as a CIPV, thereby causing the vehicle to brake incorrectly and change lanes actively, is solved. By combining navigation information, lane line lateral distance threshold and longitudinal distance threshold limits, the ramp scene image features are trained to identify non-CIPV vehicles in the ramp, so as to reduce the misidentification rate of CIPV vehicles and improve the safety and comfort of users.

[0024] A second aspect of the present application provides a CIPV vehicle identification device, including:

[0025] An acquisition module is used to obtain the surrounding environment information and navigation information of the current vehicle;

[0026] a judgment module, configured to identify the driving position of the current vehicle based on the surrounding environment information and the navigation information, and to judge whether the driving position of the current vehicle is in a main road section and whether the current vehicle is close to a ramp road section;

[0027] An identification module is used to collect current scene information of the ramp section if the current vehicle is in the main road section and the current vehicle is close to the ramp section, preprocess the current scene information, and input the preprocessed current scene information into a preset vehicle classification model to obtain the vehicle in the main road section, and determine the CIPV vehicle based on the vehicle in the main road section.

[0028] According to one embodiment of the present application, the acquisition module includes:

[0029] A determination unit, used to determine whether new surrounding environment information and navigation information of the current vehicle are received within a preset time;

[0030] The transmission unit is used to send the surrounding environment information and navigation information of the current vehicle within the previous preset time if the new navigation information of the current vehicle is not received within the preset time, otherwise, send the new surrounding environment information and navigation information of the current vehicle.

[0031] According to one embodiment of the present application, before inputting the preprocessed current scene information into the preset vehicle classification model, the recognition module further includes:

[0032] A first acquisition unit is used to acquire a public dataset of actual scene information, wherein the public dataset of actual scene information includes multiple image information of ramp road sections;

[0033] A division unit is used to preprocess each image information, annotate each preprocessed image information, and classify the annotated image information according to a preset ratio to obtain a training data set and a test data set of the image information;

[0034] A training unit is used to train the training data set based on a preset deep learning algorithm to obtain an initial vehicle classification model, and use the test data set to test the initial vehicle classification model. If the test result meets the preset test condition, the initial vehicle classification model is used as the preset vehicle classification model.

[0035] According to one embodiment of the present application, after testing the initial vehicle classification model using the test data set, the training unit further includes:

[0036] The judgment subunit is used to update the preset ratio if the test result does not meet the preset test condition, and divide the new training data set and the new test data set according to the updated ratio, and retrain the new training data set based on the preset deep learning algorithm until the new test result meets the preset test condition, thereby obtaining the preset vehicle classification model.

[0037] According to one embodiment of the present application, the identification module includes:

[0038] A second acquisition unit, configured to obtain the non-CIPV vehicles on the ramp section according to an output result of the preset vehicle classification model;

[0039] a third acquisition unit, configured to output the driving speed, yaw rate and lane line information of the current vehicle based on the vehicles on the main road section, the non-CIPV vehicles and the lane line information in the main road section of the current vehicle, and to perform risk judgment on the driving speed, yaw rate and lane line information of the current vehicle, so as to obtain a predicted driving trajectory of the current vehicle according to the risk judgment result;

[0040] The fourth acquisition unit is used to calculate the lateral distance and longitudinal distance between the current vehicle and the target vehicle based on a preset test cycle, and obtain the CIPV vehicle based on a preset lateral distance threshold and longitudinal distance threshold.

[0041] According to the CIPV vehicle identification device of the embodiment of the present application, the surrounding environment information and navigation information of the current vehicle are obtained and the driving position of the current vehicle is identified. If the current vehicle is in the main road section and close to the ramp section, the current scene information of the ramp section is collected, the current scene information is preprocessed, and the preprocessed current scene information is input into the preset vehicle classification model to obtain the vehicle in the main road section, and the CIPV vehicle is determined based on the vehicle in the main road section. In this way, the problem of misidentifying a vehicle with a long lateral distance as a CIPV, thereby causing the vehicle to brake incorrectly and change lanes actively, is solved. By combining navigation information, lane line lateral distance threshold and longitudinal distance threshold limits, the ramp scene image features are trained to identify non-CIPV vehicles in the ramp, so as to reduce the misidentification rate of CIPV vehicles and improve the safety and comfort of users.

[0042] A third aspect of the present application provides a vehicle, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the CIPV vehicle identification method as described in the above embodiment.

[0043] A fourth aspect of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute the CIPV vehicle identification method as described in the above embodiment.

[0044] The fifth aspect of the present application provides a computer program product, including a computer program, which is executed to implement the CIPV vehicle identification method described in the above embodiment.

[0045] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0047] Figure 1 A flowchart of a CIPV vehicle identification method provided according to an embodiment of the present application;

[0048] Figure 2 A schematic diagram of a main road section-ramp according to an embodiment of the present application;

[0049] Figure 3 A schematic diagram of an algorithm flow according to an embodiment of the present application;

[0050] Figure 4 A schematic diagram of a deep learning data set according to an embodiment of the present application;

[0051] Figure 5 A schematic diagram of a risk determination algorithm according to an embodiment of the present application;

[0052] Figure 6 is a block diagram of an example of a CIPV vehicle identification device according to an embodiment of the present application;

[0053] Figure 7 It is a schematic diagram of the structure of a vehicle according to an embodiment of the present application. DETAILED DESCRIPTION

[0054] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0055] The following describes the identification method, device, vehicle and storage medium of the CIPV vehicle of the embodiment of the present application with reference to the accompanying drawings. In view of the problem mentioned in the above background technology that when other vehicles approach the ramp, it is easy to misidentify the vehicle with a long lateral distance as a CIPV, thereby causing the vehicle to brake incorrectly or even change lanes actively, the present application provides a CIPV vehicle identification method, in which the surrounding environment information and navigation information of the current vehicle are obtained and the driving position of the current vehicle is identified. If the current vehicle is in the main section and close to the ramp section, the current scene information of the ramp section is collected, the current scene information is preprocessed, and the preprocessed current scene information is input into the preset vehicle classification model to obtain the vehicle in the main section, and the CIPV vehicle is determined according to the vehicle in the main section. In this way, the problem that it is easy to misidentify the vehicle with a long lateral distance as a CIPV, thereby causing the vehicle to brake incorrectly, change lanes actively, etc., is solved. By combining the navigation information, the lane line lateral distance threshold and the longitudinal distance threshold limit, the ramp scene image features are trained to identify non-CIPV vehicles in the ramp, so as to reduce the misidentification rate of CIPV vehicles and improve the safety and comfort of users.

[0056] Specifically, Figure 1 A flowchart of a method for identifying a CIPV vehicle provided in an embodiment of the present application.

[0057] like Figure 1 As shown, the CIPV vehicle identification method includes the following steps:

[0058] In step S101, the surrounding environment information and navigation information of the current vehicle are obtained.

[0059] According to one embodiment of the present application, obtaining the surrounding environment information and navigation information of the current vehicle includes: determining whether new surrounding environment information and navigation information of the current vehicle are received within a preset time; if new navigation information of the current vehicle is not received within the preset time, sending the surrounding environment information and navigation information of the current vehicle within the previous preset time, otherwise, sending new surrounding environment information and navigation information of the current vehicle.

[0060] The preset time may be set by those skilled in the art according to actual test requirements, or may be obtained through a limited number of computer simulations, and is not specifically limited here.

[0061] Specifically, Figure 2As shown in Figure 1, the adaptive cruise control function is a common ADAS (Advanced Driver Assistance Systems, advanced driver assistance systems) function, which can realize the current vehicle, that is, the current vehicle adaptively following the vehicle in front or driving at a set speed. When the user drives the vehicle on a highway or an urban elevated road and activates the adaptive cruise control function, the current vehicle traveling on the main road section will mistakenly identify the vehicle on the ramp merging into the road as a CIPV target when passing through the merging ramp or intersection, which will cause the current vehicle to brake incorrectly or even actively change lanes, thereby increasing driving risks. According to typical cases in such road scenarios, the cause of such problems is that the boundary selected when other vehicles cut into the front camera is unreasonable, and the vehicle outside the lane is mistakenly identified as the nearest path vehicle. Therefore, in order to avoid the current vehicle traveling on the main road section after the adaptive cruise control function is turned on, the current vehicle may regard the target that is too far away from the ramp merging into the main road section as a CIPV target, which may easily cause the current vehicle to leave the lane, thereby increasing the misrecognition rate of the adaptive cruise function and reducing the comfort and safety of the user's driving, the embodiment of the present application needs to accurately identify the CIPV target on the main road section and the non-CIPV target on the ramp section.

[0062] Specifically, if Figure 3 As shown, the embodiment of the present application uses a visual sensor, such as a front-view camera, to detect the surrounding environment information of the current vehicle, and introduces navigation information to comprehensively judge the surrounding environment information of the current vehicle. Since the navigation information is more accurate, the embodiment of the present application can send new surrounding environment information and navigation information of the current vehicle within a preset time (for example, 30s) if the dashboard of the current vehicle receives new surrounding environment information and navigation information of the current vehicle, that is, as long as any information of either side changes, the new surrounding environment information and navigation information of the current vehicle will be sent. If the new navigation information of the current vehicle is not received within the preset time, the surrounding environment information and navigation information of the current vehicle within the previous preset time will be sent, wherein the moment when the dashboard receives the ramp information is taken as time T0, and 100ms is taken as the minimum counting period, and the state of the surrounding vehicles of the current vehicle within 30 periods is continuously observed.

[0063] In step S102, the current driving position of the vehicle is identified based on the surrounding environment information and the navigation information, and it is determined whether the current driving position of the vehicle is in the main road section and whether the current vehicle is close to the ramp section.

[0064] Specifically, the driving position of the current vehicle is identified based on the surrounding environment information and navigation information obtained above, and it is determined whether the driving position of the current vehicle is in the main road section and close to the ramp section. If the current vehicle is in the main road section and not close to the ramp section, the current vehicle is controlled to continue to follow the preceding vehicle; if the current vehicle is in the main road section and the current vehicle is close to the ramp section, that is, the current vehicle is in the ramp entrance section or the ramp exit section, a prompt is given through the instrument panel.

[0065] In step S103, if the current vehicle is on the main road section and the current vehicle is close to the ramp section, the current scene information of the ramp section is collected, the current scene information is preprocessed, and the preprocessed current scene information is input into the preset vehicle classification model to obtain the vehicle on the main road section, and determine the CIPV vehicle based on the vehicle on the main road section.

[0066] According to one embodiment of the present application, before the preprocessed current scene information is input into a preset vehicle classification model, it also includes: obtaining a public data set of actual scene information, wherein the public data set of actual scene information includes a variety of image information of ramp road sections; preprocessing each image information, and annotating each preprocessed image information, classifying the annotated image information according to a preset ratio to obtain a training data set and a test data set of the image information; based on a preset deep learning algorithm, training the training data set to obtain an initial vehicle classification model, and using the test data set to test the initial vehicle classification model, if the test result meets the preset test conditions, the initial vehicle classification model is used as the preset vehicle classification model.

[0067] Among them, the preset test conditions can be set by technical personnel in this field according to actual test requirements and are not specifically limited here.

[0068] Specifically, Figure 4 As shown, if the current vehicle is in the main road section and the current vehicle is close to the ramp road section, the current scene information of the ramp road section is first collected, that is, the actual scene information public data set is obtained, wherein the actual scene information public data set includes a variety of image information of the ramp road section, such as ramp signs, main lane vehicles, ramp vehicles and clear lane lines, etc., and then each distorted image information is preprocessed, for example, the distorted image information is preprocessed by resizing, perspective transformation, etc. to ensure the quality of the image information, and the DarkLabel annotation software is used to annotate the vehicle targets, ramp signs and diversion islands in the preprocessed image information, and the annotated image information is randomly classified according to a preset ratio (for example, 75% and 25%) to obtain the training data set and the test data set of the preprocessed image information.

[0069] Secondly, the embodiment of the present application takes the moment when the instrument receives the current scene information of the ramp section as moment T0, takes 100ms as the minimum counting period, and continuously observes the status of vehicles around the current vehicle within 30 periods. Starting from moment T0, the training data set is trained based on the deep learning algorithm preset in the embodiment of the present application to obtain the initial vehicle classification model, and the test data set is used to test the initial vehicle classification model. If the test result meets the preset test conditions, that is, the vehicles on the main road section and the ramp section are correctly identified respectively, the initial vehicle classification model is used as the preset vehicle classification model, such as a CIPV vehicle or a non-CIPV vehicle, and the preprocessed current scene information is input into the preset vehicle classification model to obtain the vehicle on the main road section, and the CIPV vehicle is determined based on the vehicle on the main road section, that is, the lane line is identified using the preset deep learning algorithm to obtain the lane line position of the current vehicle, and then the vehicle targets are classified and screened to obtain the vehicle on the main road section, and the CIPV vehicle is determined based on the vehicle on the main road section.

[0070] According to one embodiment of the present application, CIPV vehicles are determined based on vehicles on the main road section, including: obtaining non-CIPV vehicles on the ramp section based on the output results of a preset vehicle classification model; outputting the current vehicle's driving speed, yaw rate and lane line information based on vehicles on the main road section, non-CIPV vehicles and lane line information on the main road section of the current vehicle, and performing risk judgment on the current vehicle's driving speed, yaw rate and lane line information to obtain a predicted driving trajectory of the current vehicle based on the risk judgment results; calculating the lateral distance and longitudinal distance between the current vehicle and the target vehicle based on a preset test cycle, and obtaining the CIPV vehicle based on preset lateral distance thresholds and longitudinal distance thresholds.

[0071] Among them, the preset test cycle, preset lateral distance threshold and longitudinal distance threshold can be set by technical personnel in this field according to actual test requirements, or can be obtained through a limited number of computer simulations, and are not specifically limited here.

[0072] Specifically, Figure 5As shown, in the deep learning process of the embodiment of the present application, the selected ramp vehicles will be marked as non-CIPV vehicles to suppress them from being selected as CIPV vehicles, and the marking will be cancelled when the ramp vehicles are about to merge into the main road section and after the merging, so that they can be used as candidate CIPV vehicles. At the same time, the embodiment of the present application usually introduces lane line data, which is usually divided into three areas on expressways and highways, namely, overtaking lanes, fast lanes and slow lanes, so that the lane line information of the current vehicle can be obtained based on the lane line data, and the future trajectory of the current vehicle can be predicted in combination with the speed information and yaw angular velocity of the current vehicle, and the preset lateral distance threshold between the current vehicle and the vehicle in front is calculated, wherein the preset lateral distance threshold can be calibrated according to the actual width of the lane, for example, the lateral distance threshold is set to ±1.7m, and the longitudinal distance can be directly obtained by the sensor more accurately, for example, the longitudinal distance threshold is 15m, so that the CIPV vehicles on the main road section are obtained after these conditions are restricted to reduce the misrecognition rate of CIPV vehicles.

[0073] According to one embodiment of the present application, after testing the initial vehicle classification model using the test data set, it also includes: if the test result does not meet the preset test conditions, updating the preset ratio, and dividing the new training data set and the new test data set according to the updated ratio, and retraining the new training data set based on the preset deep learning algorithm until the new test result meets the preset test conditions, thereby obtaining the preset vehicle classification model.

[0074] Specifically, in an embodiment of the present application, if the test results do not meet the preset test conditions, that is, the vehicle identification errors on the main road section and the ramp section, the preset ratio can be updated, and the new training data set and the new test data set are divided according to the updated ratio, and the new training data set is retrained based on the preset deep learning algorithm until the new test results meet the preset test conditions, and the preset vehicle classification model is obtained, and the above-mentioned steps of inputting the preprocessed current scene information into the preset vehicle classification model are continued to obtain the vehicles on the main road section, and determining the CIPV vehicles based on the vehicles on the main road section. The specific steps are as described in the above embodiment and will not be discussed in detail here.

[0075] Therefore, through the specific discussion of the above embodiments, the embodiments of the present application can achieve the following beneficial effects:

[0076] (1) The embodiment of the present application optimizes relevant strategies based on typical scenarios prone to misidentification, combines navigation information and a lateral distance threshold method that introduces lane line features, and adds a longitudinal distance threshold limit to reduce the misidentification rate of the adaptive cruise control function and improve driving comfort and safety;

[0077] (2) The embodiment of the present application collects pictures of ramps in actual road scenes to create a data set, and trains the deep learning algorithm in an end-to-end manner to extract non-CIPV target features. Vehicles on the ramp can be marked as non-CIPV, thereby reducing the probability of the vehicle being mistakenly selected as a CIPV and improving the safety and comfort of the current vehicle.

[0078] According to the identification method of the CIPV vehicle of the embodiment of the present application, the surrounding environment information and navigation information of the current vehicle are obtained and the driving position of the current vehicle is identified. If the current vehicle is in the main road section and close to the ramp section, the current scene information of the ramp section is collected, the current scene information is preprocessed, and the preprocessed current scene information is input into the preset vehicle classification model to obtain the vehicle in the main road section, and the CIPV vehicle is determined based on the vehicle in the main road section. In this way, the problem of misidentifying a vehicle with a long lateral distance as a CIPV, thereby causing the vehicle to brake incorrectly and change lanes actively, is solved. By combining navigation information, lane line lateral distance threshold and longitudinal distance threshold limits, the ramp scene image features are trained to identify non-CIPV vehicles in the ramp, so as to reduce the misidentification rate of CIPV vehicles and improve the safety and comfort of users.

[0079] Next, the CIPV vehicle identification device proposed in accordance with the embodiment of the present application is described with reference to the accompanying drawings.

[0080] Figure 6 It is a block diagram of the CIPV vehicle identification device of the embodiment of the present application.

[0081] like Figure 6 As shown, the CIPV vehicle identification device 10 includes: an acquisition module 100, a judgment module 200 and an identification module 300.

[0082] The acquisition module 100 is used to acquire the surrounding environment information and navigation information of the current vehicle;

[0083] A determination module 200, for identifying the current driving position of the vehicle based on the surrounding environment information and the navigation information, and determining whether the current driving position of the vehicle is in the main road section and whether the current vehicle is close to the ramp section;

[0084] The identification module 300 is used to collect current scene information of the ramp section if the current vehicle is in the main road section and the current vehicle is close to the ramp section, pre-process the current scene information, and input the pre-processed current scene information into a preset vehicle classification model to obtain the vehicle in the main road section, and determine the CIPV vehicle based on the vehicle in the main road section.

[0085] According to one embodiment of the present application, the acquisition module 100 includes:

[0086] A determination unit, used to determine whether new surrounding environment information and navigation information of the current vehicle are received within a preset time;

[0087] The transmission unit is used to send the surrounding environment information and navigation information of the current vehicle within the previous preset time if no new navigation information of the current vehicle is received within the preset time, otherwise, send new surrounding environment information and navigation information of the current vehicle.

[0088] According to one embodiment of the present application, before inputting the preprocessed current scene information into the preset vehicle classification model, the recognition module 300 further includes:

[0089] A first acquisition unit is used to acquire a public dataset of actual scene information, wherein the public dataset of actual scene information includes multiple image information of ramp road sections;

[0090] A division unit is used to preprocess each image information, annotate each preprocessed image information, and classify the annotated image information according to a preset ratio to obtain a training data set and a test data set of the image information;

[0091] The training unit is used to train the training data set based on the preset deep learning algorithm to obtain the initial vehicle classification model, and use the test data set to test the initial vehicle classification model. If the test result meets the preset test conditions, the initial vehicle classification model is used as the preset vehicle classification model.

[0092] According to one embodiment of the present application, after testing the initial vehicle classification model using the test data set, the training unit further includes:

[0093] The judgment subunit is used to update the preset ratio if the test result does not meet the preset test condition, and divide the new training data set and the new test data set according to the updated ratio, and retrain the new training data set based on the preset deep learning algorithm until the new test result meets the preset test condition, thereby obtaining the preset vehicle classification model.

[0094] According to one embodiment of the present application, the identification module 300 includes:

[0095] A second acquisition unit, configured to obtain non-CIPV vehicles on the ramp section according to an output result of a preset vehicle classification model;

[0096] A third acquisition unit is used to output the driving speed, yaw rate and lane line information of the current vehicle based on the vehicles on the main road section, the non-CIPV vehicles and the lane line information of the current vehicle, and to perform risk judgment on the driving speed, yaw rate and lane line information of the current vehicle, so as to obtain the predicted driving trajectory of the current vehicle according to the risk judgment result;

[0097] The fourth acquisition unit is used to calculate the lateral distance and longitudinal distance between the current vehicle and the target vehicle based on a preset test cycle, and obtain the CIPV vehicle based on a preset lateral distance threshold and longitudinal distance threshold.

[0098] According to the CIPV vehicle identification device of the embodiment of the present application, the surrounding environment information and navigation information of the current vehicle are obtained and the driving position of the current vehicle is identified. If the current vehicle is in the main road section and close to the ramp section, the current scene information of the ramp section is collected, the current scene information is preprocessed, and the preprocessed current scene information is input into the preset vehicle classification model to obtain the vehicle in the main road section, and the CIPV vehicle is determined based on the vehicle in the main road section. In this way, the problem of misidentifying a vehicle with a long lateral distance as a CIPV, thereby causing the vehicle to brake incorrectly and change lanes actively, is solved. By combining navigation information, lane line lateral distance threshold and longitudinal distance threshold limits, the ramp scene image features are trained to identify non-CIPV vehicles in the ramp, so as to reduce the misidentification rate of CIPV vehicles and improve the safety and comfort of users.

[0099] Figure 7 A schematic diagram of the structure of a vehicle provided in an embodiment of the present application. The vehicle may include:

[0100] A memory 701 , a processor 702 , and a computer program stored in the memory 701 and executable on the processor 702 .

[0101] When the processor 702 executes the program, the CIPV vehicle identification method provided in the above embodiment is implemented.

[0102] Furthermore, the vehicle also includes:

[0103] The communication interface 703 is used for communication between the memory 701 and the processor 702 .

[0104] The memory 701 is used to store computer programs that can be executed on the processor 702 .

[0105] The memory 701 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0106] If the memory 701, the processor 702 and the communication interface 703 are implemented independently, the communication interface 703, the memory 701 and the processor 702 can be connected to each other through a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 7 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0107] Optionally, in a specific implementation, if the memory 701, the processor 702 and the communication interface 703 are integrated on a chip, the memory 701, the processor 702 and the communication interface 703 can communicate with each other through an internal interface.

[0108] The processor 702 may be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0109] This embodiment also provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the above-mentioned CIPV vehicle identification method is implemented.

[0110] This embodiment also provides a computer program product, including a computer program, which is executed to implement the CIPV vehicle identification method of the above embodiment.

[0111] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0112] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0113] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0114] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or N wirings (electronic devices), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or otherwise processing in a suitable manner if necessary, and then stored in a computer memory.

[0115] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiment, the N steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0116] A person skilled in the art may understand that all or part of the steps in the above-mentioned embodiment method may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0117] In addition, each functional unit in each embodiment of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0118] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present application. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A method for identifying a CIPV vehicle, characterized in that: The following steps are involved: Obtain the surrounding environment information and navigation information of the current vehicle; Identifying the current driving position of the vehicle based on the surrounding environment information and the navigation information, and determining whether the current driving position of the vehicle is in a main road section and whether the current vehicle is close to a ramp section; If the current vehicle is on the main road section and the current vehicle is close to the ramp section, the current scene information of the ramp section is collected, the current scene information is preprocessed, and the preprocessed current scene information is input into a preset vehicle classification model to obtain the vehicle on the main road section, and the CIPV vehicle is determined based on the vehicle on the main road section.

2. The method according to claim 1, characterized in that The obtaining of the surrounding environment information and navigation information of the current vehicle includes: Determining whether new surrounding environment information and navigation information of the current vehicle are received within a preset time; If the new navigation information of the current vehicle is not received within the preset time, the surrounding environment information and navigation information of the current vehicle within the last preset time are sent; otherwise, the new surrounding environment information and navigation information of the current vehicle are sent.

3. The method according to claim 1, characterized in that: Before inputting the preprocessed current scene information into the preset vehicle classification model, the method further includes: Acquire a public dataset of actual scene information, wherein the public dataset of actual scene information includes multiple image information of ramp road sections; Preprocess each image information, annotate each preprocessed image information, and classify the annotated image information according to a preset ratio to obtain a training data set and a test data set of the image information; Based on a preset deep learning algorithm, the training data set is trained to obtain an initial vehicle classification model, and the initial vehicle classification model is tested using the test data set. If the test result meets the preset test conditions, the initial vehicle classification model is used as the preset vehicle classification model.

4. The method according to claim 3, characterized in that After testing the initial vehicle classification model using the test data set, the method further includes: If the test result does not meet the preset test condition, the preset ratio is updated, and a new training data set and a new test data set are divided according to the updated ratio. Based on the preset deep learning algorithm, the new training data set is retrained until the new test result meets the preset test condition, thereby obtaining the preset vehicle classification model.

5. The method according to claim 1, characterized in that The determining of the CIPV vehicle according to the vehicles on the main road section includes: Obtaining non-CIPV vehicles on the ramp section according to an output result of the preset vehicle classification model; Based on the vehicles on the main road section, the non-CIPV vehicles, and the lane line information on the main road section of the current vehicle, output the driving speed, yaw rate, and lane line information of the current vehicle, and perform risk judgment on the driving speed, yaw rate, and lane line information of the current vehicle to obtain a predicted driving trajectory of the current vehicle according to the risk judgment result; Based on a preset test cycle, the lateral distance and longitudinal distance between the current vehicle and the target vehicle are calculated, and the CIPV vehicle is obtained based on a preset lateral distance threshold and a preset longitudinal distance threshold.

6. A CIPV vehicle identification device, characterized in that: include: An acquisition module is used to obtain the surrounding environment information and navigation information of the current vehicle; a judgment module, configured to identify the driving position of the current vehicle based on the surrounding environment information and the navigation information, and to judge whether the driving position of the current vehicle is in a main road section and whether the current vehicle is close to a ramp road section; An identification module is used to collect current scene information of the ramp section if the current vehicle is in the main road section and the current vehicle is close to the ramp section, preprocess the current scene information, and input the preprocessed current scene information into a preset vehicle classification model to obtain the vehicle in the main road section, and determine the CIPV vehicle based on the vehicle in the main road section.

7. The device according to claim 6, characterized in that The acquisition module comprises: A determination unit, used to determine whether new surrounding environment information and navigation information of the current vehicle are received within a preset time; The transmission unit is used to send the surrounding environment information and navigation information of the current vehicle within the previous preset time if the new navigation information of the current vehicle is not received within the preset time, otherwise, send the new surrounding environment information and navigation information of the current vehicle.

8. The device according to claim 6, characterized in that Before inputting the preprocessed current scene information into the preset vehicle classification model, the recognition module further includes: An acquisition unit, configured to acquire a public data set of actual scene information, wherein the public data set of actual scene information includes multiple image information of a ramp road segment; A division unit is used to preprocess each image information, annotate each preprocessed image information, and classify the annotated image information according to a preset ratio to obtain a training data set and a test data set of the image information; A training unit is used to train the training data set based on a preset deep learning algorithm to obtain an initial vehicle classification model, and use the test data set to test the initial vehicle classification model. If the test result meets the preset test condition, the initial vehicle classification model is used as the preset vehicle classification model.

9. A vehicle, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the CIPV vehicle identification method as described in any one of claims 1-5.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement the CIPV vehicle identification method as described in any one of claims 1-5.

Images