Vehicle cut-in probability prediction method and system, storage medium and electronic equipment

Through the vehicle entry probability prediction method, the linear regression model is used to predict the entry probability, which solves the safety risks and comfort problems of the adaptive cruise control system when entering adjacent vehicles, achieves higher driving safety and comfort, and reduces the Computing resource requirements.

CN119090084AInactive Publication Date: 2024-12-06SHANGHAI BAOLONG AUTOMOTIVE CORP
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
CN202411249863.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-06
Publication Date
2024-12-06
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing adaptive cruise control system has problems of safety risks and reduced comfort when facing adjacent vehicles, especially the frequent sudden braking phenomenon, and the computing efficiency and real-time performance of deep learning models are poor.

Method used

The vehicle cutting probability prediction method is adopted to obtain the bicycle characteristics and interactive characteristics of the target vehicle, create timing characteristics, and use a linear regression model to predict the cutting probability, and update the model in real time to improve prediction accuracy.

Benefits of technology

Effectively reduce or avoid sudden braking and collisions, improve driving safety and comfort, and the linear regression model has low computing power occupation and strong interpretability, and is suitable for low-cost intelligent driving assistance systems.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119090084A_ABST
    Figure CN119090084A_ABST
Patent Text Reader

Abstract

The invention provides a vehicle cut-in probability prediction method and system, a storage medium and electronic equipment. The vehicle cut-in probability prediction method comprises the steps that single vehicle features and interaction features of a target vehicle are acquired, the single vehicle features comprise motion state features of the target vehicle, and the interaction features comprise scene features around the target vehicle; time sequence features are created according to the bicycle features and the interaction features; based on the time sequence characteristics, predicting the cut-in probability of the target vehicle by adopting a cut-in probability prediction model; wherein the cut-in probability is represented by a label y in the following formula: # imgabs0 # T is a time window, t0 is a cut-in moment, and t1 is a current moment. According to the vehicle cut-in probability prediction method, the probability that the target vehicle cuts in the lane of the vehicle can be predicted in advance, so that self-adaptive cruise control is better carried out.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application belongs to the field of vehicle assisted driving technology and relates to a vehicle cut-in probability prediction method, and in particular to a vehicle cut-in probability prediction method, system, storage medium and electronic device. Background Art

[0002] Adaptive Cruise Control (ACC) is an intelligent driver assistance function designed to improve driving safety and comfort. The ACC system uses advanced sensor technology such as radar, laser or camera to monitor the road conditions ahead in real time, especially the distance and relative speed to the vehicle ahead. The main function of the ACC system is to automatically adjust the speed according to the dynamic changes of the vehicle ahead while maintaining the preset speed to ensure a safe distance from the vehicle ahead. This automatic adjustment can reduce the driver's need to frequently operate the accelerator and brake, reducing driving fatigue.

[0003] When a vehicle is driving, adjacent vehicles often cut into its lane. The existing ACC system has certain safety risks when facing this situation, and is prone to sudden braking, resulting in reduced comfort. Summary of the invention

[0004] The embodiments of the present application provide a vehicle cut-in probability prediction method, system, storage medium and electronic device to solve the above-mentioned problems existing in the prior art.

[0005] In a first aspect, an embodiment of the present application provides a vehicle cut-in probability prediction method, the vehicle cut-in probability prediction method comprising: obtaining a single vehicle feature and an interaction feature of a target vehicle, the single vehicle feature comprising a motion state feature of the target vehicle, and the interaction feature comprising a scene feature around the target vehicle; creating a time series feature according to the single vehicle feature and the interaction feature; based on the time series feature, using a cut-in probability prediction model to predict the cut-in probability of the target vehicle; wherein the cut-in probability is represented by the label y in the following formula: T is the time window, t0 is the cut-in time, and t1 is the current time.

[0006] In an implementation of the first aspect, the single-vehicle characteristics include a lateral distance, a lateral speed, a heading angle and / or a turn signal status of the target vehicle.

[0007] In an implementation of the first aspect, the interaction features include a first interaction feature, a second interaction feature, a third interaction feature, a fourth interaction feature, a fifth interaction feature, a sixth interaction feature and / or a seventh interaction feature; the first interaction feature is a relative vehicle speed. When there is a vehicle in front of the own vehicle, the relative vehicle speed is the speed of the first vehicle relative to the second vehicle. The first vehicle is the vehicle in front of the own vehicle, and the second vehicle is the vehicle in front of the target vehicle. When there is no vehicle in front of the own vehicle, the relative vehicle speed is a preset vehicle speed value; the second interaction feature is a longitudinal distance. When the first vehicle is in front of the own vehicle, the longitudinal distance is the preset vehicle speed value. The longitudinal distance of the first vehicle relative to the second vehicle, when the first vehicle does not exist in front of the own vehicle, the longitudinal distance is a preset longitudinal distance value; the third interaction feature is the type of the second vehicle; the fourth interaction feature is the collision time change rate between the target vehicle and the second vehicle; the fifth interaction feature is the product of the relative vehicle speed and the acceleration of the target vehicle; the sixth interaction feature is the product of the distance of the first vehicle relative to the second vehicle and the acceleration of the target vehicle; the seventh interaction feature is the type of the target lane line, and the target lane line is the lane line between the own vehicle and the target vehicle.

[0008] In an implementation of the first aspect, creating a time series feature based on the bicycle feature and the interaction feature includes: determining at least two time periods; and creating the time series feature based on an average value of the bicycle feature and the interaction feature in each of the time periods.

[0009] In an implementation of the first aspect, the vehicle cut-in probability prediction method further includes: preprocessing the single vehicle feature and / or the interaction feature.

[0010] In an implementation manner of the first aspect, the cut-in probability prediction model is a linear regression model.

[0011] In an implementation of the first aspect, the vehicle cut-in probability prediction method further includes: iteratively updating the cut-in probability prediction model according to data collected by the vehicle.

[0012] In a second aspect, an embodiment of the present application provides a vehicle cut-in probability prediction system, the vehicle cut-in probability prediction system comprising: a feature acquisition module, used to acquire a single vehicle feature and an interaction feature of a target vehicle, the single vehicle feature comprising a motion state feature of the target vehicle, and the interaction feature comprising a scene feature around the target vehicle; a time series feature acquisition module, used to create a time series feature based on the single vehicle feature and the interaction feature; a cut-in probability prediction module, used to predict the cut-in probability of the target vehicle based on the time series feature using a cut-in probability prediction model; wherein the cut-in probability is represented by the label y in the following formula: T is the time window, t0 is the cut-in time, and t1 is the current time.

[0013] In a third aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the vehicle cut-in probability prediction method described in any one of the first aspects of the embodiments of the present application is implemented.

[0014] In a fourth aspect, an embodiment of the present application provides an electronic device, comprising: a memory storing a computer program; and a processor communicatively connected to the memory, for executing the vehicle cut-in probability prediction method described in any one of the first aspects of the embodiments of the present application when the computer program is called.

[0015] As described above, the vehicle cut-in probability prediction method, system, storage medium and electronic device provided by the embodiments of the present application have the following beneficial effects:

[0016] The embodiment of the present application can predict the vehicle cut-in probability, which is a continuously changing value between 0 and 1. Based on the vehicle cut-in probability, the motion state of the vehicle can be smoothly controlled, thereby reducing or even avoiding the possibility of sudden braking and / or collision, which is conducive to improving driving safety and comfort.

[0017] In the embodiment of the present application, a linear regression model can be used as a probability prediction model to perform probability prediction. This model has the characteristics of low computing power usage, strong interpretability, and high reliability, and can be applied to a lower-cost intelligent driving assistance system. In addition, the above characteristics also enable this model to be deployed in an embedded controller in the car, and can be continuously updated in real time based on the data collected by the car in the real environment, thereby improving the prediction accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1A and Figure 1B Shown is a schematic diagram of an application scenario of an embodiment of the present application.

[0019] Figure 1C Shown is a schematic diagram of the structure of the ACC system in an embodiment of the present application.

[0020] Figure 2 Shown is a flow chart of a vehicle cut-in probability prediction method provided in an embodiment of the present application.

[0021] Figure 3 Shown is a flowchart of creating timing features in an embodiment of the present application.

[0022] Figure 4 Shown is a structural schematic diagram of a vehicle cut-in probability prediction system provided in an embodiment of the present application.

[0023] Figure 5 Shown is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0024] Component number description

[0025] 400 Vehicle Cut-in Probability Prediction System

[0026] 410 Feature Acquisition Module

[0027] 420 Time Series Feature Acquisition Module

[0028] 430 Cut into the probability prediction module

[0029] 500 Electronic equipment

[0030] 501 Processor

[0031] 502 Memory

[0032] 5021 Operating System

[0033] 5022 Applications

[0034] 503 Network Interface

[0035] 504 System Bus

[0036] 505 User Interface

[0037] Steps S21 to S23

[0038] Steps S31 to S32 DETAILED DESCRIPTION

[0039] The following describes the embodiments of the present application through specific examples, and those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. The present application can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0040] It should be noted that the illustrations provided in the following embodiments are only used to illustrate the basic concept of the present application in a schematic manner, and therefore the illustrations only show components related to the present application rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0041] In the embodiments of the present application, the words "exemplary" or "for example" represent examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0042] In the embodiments of the present application, "at least one" refers to one or more, and "plurality" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone, where A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can represent: a, b, c, ab, ac, bc or abc, where a, b, c can be single or multiple.

[0043] Figure 1A and Figure 1B The following is a schematic diagram of a scenario of an embodiment of the present application, wherein Figure 1A The display shows a situation where a vehicle on the left cuts into the lane of the vehicle. Figure 1B It shows a situation where the vehicle on the right cuts into the lane of the own vehicle. In the above scenario, the vehicle performing adaptive cruise control is called the own vehicle. The vehicle that cuts into the lane of the own vehicle is called the target vehicle. The vehicle located in front of the own vehicle before the cut-in occurs is called the first vehicle. The first vehicle may exist in some scenarios, and may not exist in other scenarios, and this application does not limit this. The vehicle in front of the target vehicle is called the second vehicle. The lane line between the target vehicle and the own vehicle is called the target lane line, and the moment when the center of the rear of the target vehicle crosses the target lane line is called the cut-in moment. For ease of explanation, in the embodiment of the present application, the direction parallel to the lane line is defined as the longitudinal direction, and the direction perpendicular to the lane line is defined as the lateral direction.

[0044] Figure 1C The following is a schematic diagram of the ACC system of the vehicle. Figure 1C As shown, the ACC system includes an information perception unit, an electronic control unit (ECU), an execution unit and a human-computer interaction interface.

[0045] The information perception unit is used to perceive information such as the state of the vehicle and the driving environment. Exemplarily, the information perception unit may include an environmental sensor, a distance sensor, a speed sensor, a steering angle sensor, a throttle position sensor, a brake pedal sensor, etc. Environmental sensors are, for example, millimeter wave radars, cameras, etc., which are used to detect environmental information around the vehicle. The distance sensor is used to obtain the vehicle-to-vehicle distance signal, which can be achieved through a laser radar or a millimeter wave radar. The speed sensor is used to obtain the real-time speed of the vehicle, which can be achieved through a Hall-type speed sensor. The steering angle sensor is used to obtain the steering signal of the vehicle, so as to determine the direction of travel of the vehicle. The throttle position sensor is used to obtain the throttle opening signal. The brake pedal sensor is used to obtain the brake pedal action signal. The wheel speed sensor is used to sense the driving speed of the vehicle.

[0046] The electronic control unit is communicatively connected with the information perception unit, the execution unit and the human-computer interaction interface, and is used to receive and process the data transmitted by the information perception unit, make speed adjustment and distance control decisions according to the preset control strategy and the received data, and generate the control instructions required by the execution unit to adjust the throttle, brake and other actuators of the vehicle.

[0047] In some implementations, the electronic control unit may include a microprocessor (MCU), a memory, and a communication interface. The microprocessor is used to execute complex control algorithms and real-time data processing. The memory is used to store runtime data and program code. The communication interface is responsible for data communication with other systems (such as power systems, brake systems, etc.) and sensors.

[0048] The execution unit is used to execute the control instructions issued by the electronic control unit to adjust the driving state of the vehicle. Figure 1C In the ACC system shown, the execution unit includes a motor controller, a brake controller, and a steering controller. The motor controller is used to drive the throttle and brake actuators. The brake controller is responsible for managing the vehicle's brake system and controls the brake pressure to slow down or stop the vehicle. The steering controller is used to control the vehicle's steering system to adjust the vehicle's driving direction.

[0049] The Human Machine Interface (HMI) is the communication bridge between the ACC system and the driver, responsible for displaying the system status and receiving driver input.

[0050] When the target vehicle cuts into the lane of the ego vehicle, the information perception unit collects data about the ego vehicle, the target vehicle, and the surrounding environment, and sends this data to the electronic control unit. The electronic control unit replaces the target vehicle with the main target in front and makes decisions to generate adjustment instructions. The execution unit adjusts the ego vehicle's speed and the safe distance from the vehicle in front according to these adjustment instructions.

[0051] When replacing the main target in some technical solutions, the electronic control unit needs to make a judgment based on the lateral speed of the target vehicle and the degree of overlap between the target vehicle and the lane. However, when the target vehicle is close to the vehicle, the switching of the main target is likely to cause emergency braking, thus affecting driving comfort. In addition, these technical solutions only make cut-in judgments based on the current state of the target vehicle, and cannot anticipate the cut-in of the target vehicle and take timely measures, thus posing certain safety hazards.

[0052] Some technical solutions use deep learning-based methods to train artificial intelligence models to predict the target vehicle's driving trajectory through historical traffic data, environmental information data, and road structure data. However, the implementation of these technical solutions relies on a large amount of historical traffic data to train deep learning models. If the data is insufficient or unrepresentative, the performance of the deep learning model may be affected. In addition, the prediction results of the deep learning model have a high degree of uncertainty, and these technical solutions cannot effectively handle this uncertainty. Furthermore, the autonomous driving system needs to predict the intention to change lanes in real time in order to make timely decisions. Under the premise of limited computing power, the computing efficiency and real-time performance of the deep learning model are poor.

[0053] At least in view of the above problems, the present application provides a method for predicting vehicle cut-in probability, which can be applied to Figure 1C The ACC system shown. Figure 2 The flowchart of the vehicle cut-in probability prediction method provided by the embodiment of the present application is shown. Figure 2 As shown, the vehicle cut-in probability prediction method includes the following steps S21 to S23.

[0054] S21, obtaining the single vehicle features and interactive features of the target vehicle. The single vehicle features include the motion state features of the target vehicle and / or some other features related to lane change, and the interactive features include the scene features around the target vehicle. Among them, the scene features around the target vehicle are, for example, scene features within a certain distance range around the target vehicle as the center, and the distance range can be set according to actual needs, for example, 20 meters, 50 meters, etc.

[0055] S22, creating a time series feature according to the single vehicle feature and the interaction feature. The time series feature is used to describe the dynamic changes of the single vehicle feature and the interaction feature.

[0056] S23, based on the time series features, using the cut-in probability prediction model to predict the cut-in probability of the target vehicle. The cut-in probability prediction model is a trained model. The cut-in probability is represented by the label y in the following formula: T is the time window, t0 is the cut-in time, and t1 is the current time. The larger the value of the time window T, the larger the amount of data that needs to be processed during the prediction of the cut-in probability, and the earlier the prediction can be achieved. The smaller the value of the time window T, the smaller the amount of data that needs to be processed during the prediction of the cut-in probability, and the smaller the prediction lead time that can be achieved. In specific applications, the value of the time window T can be selected according to the actual computing power, for example, it can be 3 seconds or 5 seconds.

[0057] As described above, the label y is used in the embodiment of the present application to represent the cut-in probability. The label y is a value that changes continuously over time, and its value range is [0, 1]. The closer the current moment t1 is to the cut-in moment t0, the larger the value of the label y is, and the greater the cut-in probability of the target vehicle is. For the target vehicle that has not cut-in behavior, the corresponding label y value is 0. Therefore, the cut-in probability of the target vehicle predicted in the embodiment of the present application is a continuous value that changes over time. According to the continuous value, the motion state of the vehicle can be smoothly controlled, thereby reducing or even avoiding the possibility of sudden braking and / or collision, which is conducive to improving driving safety and comfort. In comparison, in some technical solutions, the judgment result of whether the target vehicle cuts in is only yes or no. If the judgment is yes, the braking operation is performed, and if the judgment result is otherwise, no operation is performed. This method is prone to sudden braking or even collision, thereby bringing safety hazards and reducing driving comfort.

[0058] In some implementations, the single-vehicle characteristics include the target vehicle's lateral distance, lateral speed, heading angle, and / or turn signal status.

[0059] For example, the lateral distance of the target vehicle can be represented by the lateral distance between the target vehicle and the target lane line. The closer the lateral distance of the target vehicle is, the higher the probability that the target vehicle cuts into the lane of the vehicle.

[0060] For example, the lateral speed of the target vehicle can be represented by the lateral speed of the target vehicle to the own vehicle lane. The greater the lateral speed of the target vehicle, the higher the probability that the target vehicle cuts into the own vehicle lane.

[0061] For example, the heading angle of the target vehicle can be represented by the heading angle of the target vehicle to the lane of the own vehicle. The larger the heading angle of the target vehicle is, the higher the probability that the target vehicle cuts into the lane of the own vehicle is.

[0062] For example, if the turn signal of the target vehicle on the side of the own vehicle lane flashes, the probability that the target vehicle cuts into the own vehicle lane is high.

[0063] The probability of the target vehicle cutting in can be determined based on the single vehicle characteristics of the target vehicle. However, when the ego vehicle recognizes the single vehicle characteristics of the target vehicle, the cutting in behavior of the target vehicle may have already occurred. To address this issue, in the embodiment of the present application, the interaction characteristics of the target vehicle and the surrounding scene are also obtained, and the possibility of the target vehicle cutting in can be determined as early as possible by comprehensively considering the single vehicle characteristics and interaction characteristics of the target vehicle.

[0064] In some implementations, the interaction features include a first interaction feature, a second interaction feature, a third interaction feature, a fourth interaction feature, a fifth interaction feature, a sixth interaction feature, and / or a seventh interaction feature.

[0065] The first interactive feature is the relative speed. When there is a vehicle in front of the vehicle, the relative speed is the speed of the first vehicle relative to the second vehicle, the first vehicle is the vehicle in front of the vehicle, and the second vehicle is the vehicle in front of the target vehicle. When there is no vehicle in front of the vehicle, the relative speed is a preset speed value, such as 50 kph. The greater the relative speed, the higher the probability that the target vehicle cuts into the lane of the vehicle.

[0066] The second interactive feature is the longitudinal distance. When there is a first vehicle in front of the ego vehicle, the longitudinal distance is the longitudinal distance of the first vehicle relative to the second vehicle. When there is no first vehicle in front of the ego vehicle, the longitudinal distance is a preset longitudinal distance value, and the preset longitudinal distance value is, for example, 300 meters. The larger the longitudinal distance, the higher the probability that the target vehicle cuts into the lane of the ego vehicle.

[0067] The third interactive feature is the type of the second vehicle. For example, when the second vehicle is a large truck, the probability of the target vehicle cutting into the lane of the own vehicle is high.

[0068] The fourth interactive feature is a time-to-collision (TTC) change rate between the target vehicle and the second vehicle.

[0069] The fifth interactive feature is the product of the relative vehicle speed and the acceleration of the target vehicle. For example, when the first vehicle is slower and the second vehicle is faster, if the target vehicle decelerates, it means that the probability of the target vehicle cutting into the lane of the own vehicle is higher.

[0070] The sixth interactive feature is the product of the distance between the first vehicle and the second vehicle and the acceleration of the target vehicle. For example, when the first vehicle is closer to the target vehicle and the second vehicle is farther from the target vehicle, if the target vehicle decelerates, it means that the probability of the target vehicle cutting into the lane of the self-vehicle is high.

[0071] The seventh interactive feature is the type of the target lane line. For example, if the target lane line is a solid line, the probability of the target vehicle cutting into the own vehicle lane is low. If the target lane line is a dotted line, the probability of the target vehicle cutting into the own vehicle lane is relatively high.

[0072] The cut-in behavior of the target vehicle is highly correlated with the temporal process of the scene. Based on this, in some implementations, a sliding window method can be used to create temporal features.

[0073] For example, see Figure 3 ,Creating time series features according to the bicycle features and the interaction features includes the following steps S31 and S32.

[0074] S31, determining at least two time periods. The length of each time period can be set according to actual needs.

[0075] S32, creating time series features according to the average values ​​of the single-vehicle features and the interaction features in each time period.

[0076] Exemplarily, for any bicycle feature xi, the average value xi_0.5 of the bicycle feature xi within 0.5 seconds before the current moment, the average value xi_1 within 1 second before the current moment, and the average value xi_3 within 3 seconds before the current moment can be obtained as the bicycle time series feature corresponding to xi. For any interaction feature xj, the average value xj_0.5 of the interaction feature xj within 0.5 seconds before the current moment, the average value xj_1 within 1 second before the current moment, and the average value xj_3 within 3 seconds before the current moment can be obtained as the interaction time series feature corresponding to xj. In step S32, all bicycle time series features and interaction time series features can be obtained as the time series features.

[0077] In some implementations, the vehicle cut-in probability prediction method may further include: preprocessing the single vehicle features and / or interaction features. Specifically, data preprocessing is used to detect, correct or delete damaged, inaccurate or inappropriate data from the original data, so that the data is more suitable for the cut-in probability prediction model and meets the requirements of the model.

[0078] Exemplarily, data preprocessing may include missing value processing. For example, statistical methods, interpolation methods, high-dimensional mapping methods, etc. may be selected to fill missing values ​​according to data distribution characteristics and importance.

[0079] Exemplarily, data preprocessing may include outlier processing. For example, outliers may be identified and processed using methods such as drawing, standard deviation, box plot, clustering, etc.

[0080] Exemplarily, data preprocessing may include noise processing. For example, data may be smoothed by binning, regression, etc. to reduce noise.

[0081] Exemplarily, data preprocessing may include feature scaling. For example, standardization or normalization may be used to map the numerical ranges of different features to the same scale.

[0082] Exemplarily, data preprocessing may include label encoding. For example, a categorical variable may be converted into a numerical value so that the cut-in probability prediction model can process it. In one example, for the type of target lane line, a solid line may be encoded as 1, a dashed line may be encoded as 2, a double solid line may be encoded as 3, and a deceleration line may be encoded as 4.

[0083] In some implementations, the cut-in probability prediction model may be a linear regression model, but the present application is not limited thereto. In some other embodiments, the cut-in probability prediction model may also be a ridge regression, a support vector machine, a neural network, etc., or may be a fusion model of different models.

[0084] The linear regression model can be expressed as follows:

[0085] y=w1x1+w2x2+…+w n x n +b

[0086] Where y represents the data label, w i is the slope, indicating the characteristic x i The weight of the linear regression model, b is the intercept of the linear regression model, its value is a constant, n is the number of time series features, x i is the i-th time series feature.

[0087] Exemplarily, the training process of the linear regression model includes data collection, data set division and model training.

[0088] In the data collection stage, the information perception unit of the vehicle can be used to collect the required data, which include, for example: the vehicle's speed, acceleration, turn signal state, and lateral distance from the target lane line; the first vehicle's longitudinal distance, longitudinal absolute speed, longitudinal absolute acceleration, lateral distance, lateral absolute speed, lateral absolute acceleration, turn signal state, vehicle type, lateral distance from the target lane line, and heading angle; the target vehicle's longitudinal distance, longitudinal absolute speed, longitudinal absolute acceleration, lateral distance, lateral absolute speed, lateral absolute acceleration, turn signal state, vehicle type, lateral distance from the target lane line, and heading angle; the second vehicle's longitudinal distance, longitudinal absolute speed, longitudinal absolute acceleration, lateral distance, lateral absolute speed, lateral absolute acceleration, turn signal state, vehicle type, lateral distance from the target lane line, and heading angle; the type of the target lane line (dashed line, solid line, double solid line, deceleration line, etc.). The data closely related to the cut-in probability prediction is selected from the collected data to obtain the target data set.

[0089] In the data set division stage, the target data set is divided into training set and test set according to a certain ratio.

[0090] In the model training stage, the training set is used to train the cut-in probability prediction model, and the test set is used to test the performance of the cut-in probability prediction model.

[0091] Exemplarily, a loss function may be used to measure the difference between the predicted value and the true value of the cut-in probability. The loss function is, for example: Where m is the number of samples, y i The true value of the label, Represents the predicted value of the label.

[0092] For example, the parameter value may be determined using a gradient descent method, which optimizes the objective function in an iterative manner to find the minimum or maximum value of the function and gradually approaches the optimal solution by continuously adjusting the parameters.

[0093] Specifically, the gradient descent method repeatedly performs the following steps until convergence: initializing parameters, which include the slope and intercept in the linear regression model; calculating the gradient of the objective function (loss function) with respect to the parameters; and updating the parameters.

[0094] Exemplarily, the parameter update is as follows:

[0095]

[0096] Where α is the learning rate, which indicates the step size of each update in each iteration. J(θ0,θ1) is the objective function (i.e., loss function). j The parameters to be updated.

[0097] In some implementations, the vehicle cut-in probability prediction method may further include: iteratively updating the cut-in probability prediction model according to data collected by the vehicle.

[0098] For example, the linear regression model has the characteristics of strong interpretability and low computing power usage. When using the linear regression model as the cut-in probability prediction model, the model can be deployed in the embedded controller in the vehicle and continuously updated iteratively. Specifically, during the driving of the vehicle, the following process is repeated: the cut-in probability prediction model is updated in real time using the data collected by the vehicle, and the next prediction is made using the updated cut-in probability prediction model. In this way, the prediction accuracy can be further improved.

[0099] The protection scope of the vehicle cut-in probability prediction method provided in the embodiment of the present application is not limited to the execution order of the steps listed in this embodiment. All solutions implemented by adding, reducing or replacing steps in the prior art based on the principles of the present application are included in the protection scope of the present application.

[0100] An embodiment of the present application also provides a vehicle cut-in probability prediction system, which can implement the vehicle cut-in probability prediction method described in the present application, but the implementation device of the vehicle cut-in probability prediction method described in the present application includes but is not limited to the structure of the vehicle cut-in probability prediction system listed in the present embodiment. All structural deformations and replacements of the prior art made according to the principles of the present application are included in the protection scope of the present application.

[0101] Figure 4 is a schematic block diagram of a vehicle cut-in probability prediction system provided by an embodiment of the present application. Figure 4 As shown, the vehicle cut-in probability prediction system 400 includes a feature acquisition module 410, a time series feature acquisition module 420 and a cut-in probability prediction module 430. Among them, the feature acquisition module 410 is used to obtain the single vehicle feature and interaction feature of the target vehicle, the single vehicle feature includes the motion state feature of the target vehicle, and the interaction feature includes the scene feature around the target vehicle. The time series feature acquisition module 420 is used to create time series features based on the single vehicle feature and the interaction feature. The cut-in probability prediction module 430 is used to predict the cut-in probability of the target vehicle based on the time series feature using the cut-in probability prediction model. The cut-in probability is represented by the label y in the following formula: T is the time window, t0 is the cut-in time, and t1 is the current time.

[0102] It should be understood that the specific process of each module executing the above corresponding steps has been described in detail in the above method embodiment, and for the sake of brevity, it will not be repeated here.

[0103] The embodiment of the present application also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by the processor, the vehicle cut-in probability prediction method provided by the embodiment of the present application is implemented. A person of ordinary skill in the art can understand that all or part of the steps in the method for implementing the above embodiment can be completed by instructing the processor through a program, and the program can be stored in a computer-readable storage medium, and the storage medium is a non-transitory medium, such as a random access memory, a read-only memory, a flash memory, a hard disk, a solid-state hard disk, a magnetic tape, a floppy disk, an optical disc, and any combination thereof. The above storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that includes one or more available media integrated. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a digital video disc (DVD)), or a semiconductor medium (e.g., a solid-state disk (SSD)), etc.

[0104] An embodiment of the present application also provides an electronic device. Figure 5 is a schematic block diagram of an electronic device provided in an embodiment of the present application. Figure 5 As shown, the electronic device 500 includes: at least one processor 501, a memory 502, at least one network interface 503 and a user interface 505. The various components in the electronic device 500 are coupled together via a bus system 504. It can be understood that the bus system 504 is used to realize the connection and communication between these components. In addition to the data bus, the bus system 504 also includes a power bus, a control bus and a status signal bus. However, for the sake of clarity, the bus system 504 is not described in detail. Figure 5 In the specification, various buses are labeled as bus systems.

[0105] The user interface 505 may include a display, a keyboard, a mouse, a trackball, a click gun, keys, buttons, a touch pad or a touch screen.

[0106] It is understood that the memory 502 can be a volatile memory or a non-volatile memory, and can also include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), which is used as an external cache. By way of exemplary but not limiting explanation, many forms of RAM are available, such as static random access memory (SRAM), synchronous static random access memory (SSRAM). The memory described in the embodiments of the present application is intended to include but is not limited to these and any other suitable categories of memory.

[0107] The memory 502 in the embodiment of the present application is used to store various categories of data to support the operation of the electronic device 500. Examples of these data include: any executable program for operating on the electronic device 500, such as an operating system 5021 and an application 5022; the operating system 5021 includes various system programs, such as a framework layer, a core library layer, a driver layer, etc., for implementing various basic services and processing hardware-based tasks. The application 5022 may include various applications, such as a media player (Media Player), a browser (Browser), etc., for implementing various application services. The vehicle cut-in probability prediction method provided in the embodiment of the present application may be included in the application 5022.

[0108] The method disclosed in the above embodiment of the present application can be applied to the processor 501, or implemented by the processor 501. The processor 501 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit or software instructions in the processor 501. The above processor 501 may be a general processor, a digital signal processor (DSP, Digital Signal Processor), or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The processor 501 can implement or execute the various methods, steps and logic block diagrams disclosed in the embodiments of the present application. The general processor 501 can be a microprocessor or any conventional processor, etc. In combination with the steps of the accessory optimization method provided in the embodiment of the present application, it can be directly embodied as a hardware decoding processor to execute, or it can be executed by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium, which is located in a memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0109] In an exemplary embodiment, the electronic device 500 may be implemented by one or more application specific integrated circuits (ASIC), DSP, programmable logic device (PLD), complex programmable logic device (CPLD) to execute the aforementioned method.

[0110] The present application embodiment may also provide a computer program product, the computer program product including one or more computer instructions. When the computer instructions are loaded and executed on a computing device, the process or function described in the embodiment of the present application is generated in whole or in part. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from a website site, computer or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer or data center.

[0111] When the computer program product is executed by a computer, the computer executes the method described in the above method embodiment. The computer program product may be a software installation package, and when the above method is required, the computer program product may be downloaded and executed on a computer.

[0112] The terms "component", "module", "system", etc. used in this specification are used to represent computer-related entities, hardware, firmware, a combination of hardware and software, software, or software in execution. For example, a component can be, but is not limited to, a process running on a processor, a processor, an object, an executable file, an execution thread, a program and / or a computer. By way of illustration, both applications running on a computing device and a computing device can be components. One or more components may reside in a process and / or an execution thread, and a component may be located on a computer and / or distributed between two or more computers. In addition, these components may be executed from various computer-readable media having various data structures stored thereon. Components may, for example, communicate through local and / or remote processes according to signals having one or more data packets (e.g., data from two components interacting with another component between a local system, a distributed system and / or a network, such as the Internet interacting with other systems through signals).

[0113] Those of ordinary skill in the art will appreciate that the various illustrative logical blocks and steps described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or in a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0114] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

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

[0116] In addition, each functional unit in 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.

[0117] In the above embodiments, the functions of each functional unit can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions (programs). When loading and executing computer program instructions (programs) on a computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. Computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, computer instructions can be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center.

[0118] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the 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 for a computer device (which can be a personal computer, server, or network device, etc.) 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 a USB flash drive, a mobile hard disk, a read-only memory, a random access memory, a magnetic disk or an optical disk.

[0119] In summary, the embodiments of the present application provide a method, system, storage medium and electronic device for predicting the probability of vehicle cut-in. The embodiments of the present application can predict the probability of vehicle cut-in, which is a continuously changing value between 0 and 1. Based on the vehicle cut-in probability, the motion state of the vehicle can be smoothly controlled, thereby reducing or even avoiding the possibility of sudden braking and / or collision, which is conducive to improving driving safety and comfort. In addition, in the embodiments of the present application, a linear regression model can be used as a probability prediction model to predict the cut-in probability. This model has the characteristics of small computing power occupation, strong interpretability and high reliability, and can be applied to a lower-cost intelligent driving assistance system. In addition, the above-mentioned characteristics also enable this model to be deployed in an embedded controller in the car, and can be continuously updated in real time according to the data collected by the vehicle in the real environment, thereby improving the prediction accuracy. Therefore, the present application effectively overcomes the various shortcomings in the prior art and has a high industrial utilization value.

[0120] The above embodiments are merely illustrative of the principles and effects of the present application and are not intended to limit the present application. Anyone familiar with the technology may modify or change the above embodiments without violating the spirit and scope of the present application. Therefore, all equivalent modifications or changes made by a person of ordinary skill in the art without departing from the spirit and technical ideas disclosed in the present application shall still be covered by the claims of the present application.

Claims

1. A method for predicting vehicle cut-in probability, It is characterized in that The vehicle cut-in probability prediction method comprises: Acquire a single vehicle feature and an interactive feature of a target vehicle, wherein the single vehicle feature includes a motion state feature of the target vehicle, and the interactive feature includes a scene feature around the target vehicle; Creating a time series feature according to the bicycle feature and the interaction feature; Based on the time series characteristics, a cut-in probability prediction model is used to predict the cut-in probability of the target vehicle; The cut-in probability is represented by the label y in the following formula: T is the time window, t 0 is the cut-in time, t 1 For the current moment.

2. The vehicle cut-in probability prediction method according to claim 1, It is characterized in that The single-vehicle characteristics include the lateral distance, lateral speed, heading angle and / or turn signal status of the target vehicle.

3. The vehicle cut-in probability prediction method according to claim 1, It is characterized in that The interaction features include a first interaction feature, a second interaction feature, a third interaction feature, a fourth interaction feature, a fifth interaction feature, a sixth interaction feature and / or a seventh interaction feature; The first interaction feature is a relative vehicle speed. When there is a vehicle in front of the own vehicle, the relative vehicle speed is the speed of the first vehicle relative to the second vehicle. The first vehicle is the vehicle in front of the own vehicle, and the second vehicle is the vehicle in front of the target vehicle. When there is no vehicle in front of the own vehicle, the relative vehicle speed is a preset vehicle speed value. The second interaction feature is a longitudinal distance. When the first vehicle exists in front of the vehicle, the longitudinal distance is a longitudinal distance between the first vehicle and the second vehicle. When the first vehicle does not exist in front of the vehicle, the longitudinal distance is a preset longitudinal distance value. The third interaction feature is the type of the second vehicle; The fourth interaction feature is a collision time change rate between the target vehicle and the second vehicle; The fifth interaction feature is the product of the relative vehicle speed and the acceleration of the target vehicle; The sixth interaction feature is the product of the distance between the first vehicle and the second vehicle and the acceleration of the target vehicle; The seventh interactive feature is the type of the target lane line, and the target lane line is the lane line between the vehicle and the target vehicle.

4. The vehicle cut-in probability prediction method according to claim 1, It is characterized in that Creating a time series feature according to the bicycle feature and the interaction feature includes: Identify at least two time periods; The time series feature is created according to the average values ​​of the single-vehicle feature and the interaction feature in each of the time periods.

5. The vehicle cut-in probability prediction method according to claim 1, It is characterized in that The vehicle cut-in probability prediction method further includes: preprocessing the single vehicle feature and / or the interaction feature.

6. The vehicle cut-in probability prediction method according to claim 1, It is characterized in that The cut-in probability prediction model is a linear regression model.

7. The vehicle cut-in probability prediction method according to claim 1, It is characterized in that The vehicle cut-in probability prediction method further includes: iteratively updating the cut-in probability prediction model according to data collected by the vehicle.

8. A vehicle cut-in probability prediction system, It is characterized in that The vehicle cut-in probability prediction system comprises: A feature acquisition module, used to acquire single-vehicle features and interactive features of a target vehicle, wherein the single-vehicle features include motion state features of the target vehicle, and the interactive features include scene features around the target vehicle; A time series feature acquisition module, used to create a time series feature according to the bicycle feature and the interaction feature; A cut-in probability prediction module, used to predict the cut-in probability of the target vehicle using a cut-in probability prediction model based on the time series characteristics; The cut-in probability is represented by the label y in the following formula: T is the time window, t 0 is the cut-in time, t 1 For the current moment.

9. A computer-readable storage medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, the vehicle cut-in probability prediction method according to any one of claims 1 to 7 is implemented.

10. An electronic device, It is characterized in that The electronic device comprises: A memory storing a computer program; A processor is communicatively connected to the memory, and executes the vehicle cut-in probability prediction method according to any one of claims 1 to 7 when calling the computer program.

Citation Information

Patent Citations

  • Vehicle lane change prediction

    CN111527013A

  • Vehicle behavior prediction method and device

    CN112562328A

  • Vehicle lane changing behavior prediction model construction, prediction and early warning method and system

    CN113516846A

  • Vehicle control method and device and vehicle

    CN114435389A

  • Lane change prediction regression model training method, lane change prediction method and device

    CN114945961A