Automatic driving prediction decision-making method and device and storage medium

Abstract

The invention relates to an automatic driving prediction decision-making method and device and a storage medium, and the method comprises the steps: obtaining the surrounding environment information of a vehicle, and obtaining the vehicle type and driving behavior characteristics of a target vehicle according to the surrounding environment information of the vehicle; obtaining the driving maturity of the target vehicle according to the vehicle type and the driving behavior characteristics; wherein the driving maturity is an index value used for evaluating the driving level of the driver; when the target vehicle enters the region of interest of the vehicle, if the driving maturity of the target vehicle is greater than or equal to a first preset threshold value and less than a second preset threshold value, obtaining current motion information of the target vehicle, and predicting a driving characteristic track of the target vehicle according to the current motion information and the driving behavior characteristics, and generating a corresponding driving decision according to the driving characteristic track. According to the invention, the maneuvering actions of the surrounding vehicles can be actively pre-judged in advance, and the automatic driving safety is improved.

Description

Technical Field

[0001] The present application relates to the field of autonomous driving technology, and specifically to an autonomous driving prediction decision-making method, a device, and a storage medium thereof. Background Art

[0002] By predicting the trajectories of surrounding vehicles, autonomous vehicles can better predict the driving behaviors of surrounding vehicles, such as lane changes, acceleration, deceleration, and parking, thereby better planning their own driving strategies.

[0003] Current autonomous driving prediction and decision-making methods rely solely on the historical trajectories and driving behavior characteristics of surrounding vehicles to infer the future movements of autonomous driving vehicles on surrounding vehicles. This is a passive maneuver, meaning it responds to the behavior of surrounding vehicles rather than actively predicting it. This often results in late predictions and may cause the planned driving strategy to react slowly, posing certain safety risks. Summary of the Invention

[0004] This application aims to propose an autonomous driving prediction decision-making method, its device, and storage medium, so as to proactively predict the maneuvers of surrounding vehicles in advance and improve the safety of autonomous driving.

[0005] According to the first aspect of the present application, a method for predicting and making decisions on autonomous driving is proposed, the method comprising:

[0006] Acquiring information about the surrounding environment of the own vehicle, and obtaining the vehicle type and driving behavior characteristics of the target vehicle based on the information about the surrounding environment of the own vehicle;

[0007] Obtaining the driving maturity of the target vehicle according to the vehicle type and driving behavior characteristics; wherein the driving maturity is an index value used to evaluate the driving level of the driver;

[0008] When the target vehicle enters the area of interest of the own vehicle, if the driving maturity of the target vehicle is greater than or equal to a first preset threshold and less than a second preset threshold, the current motion information of the target vehicle is obtained, the driving characteristic trajectory of the target vehicle is predicted based on the current motion information and the driving behavior characteristics, and a corresponding driving decision is generated based on the driving characteristic trajectory.

[0009] Compared with traditional passive prediction methods, the above-mentioned autonomous driving prediction decision-making method can actively utilize the driving maturity and driving behavior characteristics of the target vehicle to predict the driving intention of the target vehicle earlier, and obtain a driving characteristic trajectory that conforms to the personalized driving characteristics of the target vehicle, so as to make corresponding driving decisions more timely and accurately. This can improve the planning and response speed of autonomous driving vehicles, thereby actively predicting the maneuvering actions of surrounding vehicles in advance, reducing conflicts and dangers between the vehicle and the target vehicle, and improving the safety of autonomous driving.

[0010] According to the second aspect of the present application, an autonomous driving prediction and decision-making device is proposed, comprising a module for executing the above-mentioned autonomous driving prediction and decision-making method.

[0011] According to the third aspect of the present application, a computer-readable storage medium is proposed, on which a computer program is stored. When the computer program is executed by a processor, the autonomous driving prediction decision method described in the embodiment is implemented.

[0012] Other features and advantages of the present application will be set forth in the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0014] Figure 1 This is a flowchart of an autonomous driving prediction decision method in one embodiment of the present application.

[0015] Figure 2 A schematic diagram of the driving characteristic trajectory of a target vehicle in one embodiment of the present application.

[0016] Figure 3 This is a schematic diagram of a scenario in which a target vehicle in an adjacent lane is traveling close to the lane in an embodiment of the present application.

[0017] Figure 4 This is a schematic diagram of a scenario in which a target vehicle in an adjacent lane crosses the lane in accordance with an embodiment of the present application.

[0018] Figure 5 This is a schematic diagram of a prediction model in one embodiment of the present application.

[0019] Figure 6 This is a flowchart of an autonomous driving prediction decision method in another embodiment of the present application.

[0020] Figure 7 This is a structural diagram of an autonomous driving prediction and decision-making device in one embodiment of the present application. DETAILED DESCRIPTION

[0021] Various exemplary embodiments, features, and aspects of the present disclosure will be described in detail below with reference to the accompanying drawings. The same reference numerals in the accompanying drawings represent elements with the same or similar functions. Although various aspects of the embodiments are shown in the accompanying drawings, the drawings are not necessarily drawn to scale unless otherwise indicated.

[0022] In addition, in order to better illustrate the present application, numerous specific details are provided in the specific examples below. Those skilled in the art will understand that the present application can be practiced without certain specific details. In some examples, means well known to those skilled in the art are not described in detail in order to highlight the main purpose of the present application.

[0023] like Figure 1 As shown, the embodiment of the present application proposes an autonomous driving prediction decision method, which includes the following steps:

[0024] Step S10, obtaining the vehicle's surrounding environment information, and obtaining the vehicle type and driving behavior characteristics of the target vehicle based on the vehicle's surrounding environment information;

[0025] Specifically, the vehicle's surrounding environment information includes information on obstacles such as pedestrians, vehicles, animals, roadblocks, and lane lines, traffic lights, traffic signs, etc., which can be perceived by sensing equipment (such as cameras and radars) installed on the autonomous vehicle, or obtained through V2X technology through V2X equipment installed on the autonomous vehicle. At the same time, high-precision map data can be combined to assist in obtaining information such as lane lines, traffic lights, traffic signs, etc. The obstacle information is, for example, the longitudinal position, lateral position, lateral speed, longitudinal speed, lateral acceleration, longitudinal acceleration, heading angle, yaw angular velocity, etc. of surrounding vehicles; the vehicle types include but are not limited to large trucks, tankers, taxis, buses, private cars, special vehicles, trailers, etc., and the driving behavior characteristics include but are not limited to driving on the line, driving in a dragon pattern, following a vehicle too closely, driving close to the left lane, driving close to the right lane, sudden acceleration, sudden deceleration, frequent lane changes, changing lanes without turning on the light, etc.

[0026] Step S20, obtaining the driving maturity of the target vehicle according to the vehicle type and driving behavior characteristics; wherein the driving maturity is an indicator value used to evaluate the driving level of the driver;

[0027] Specifically, the driving behavior characteristics specifically refer to swerving, frequent lane changes, sudden acceleration, sudden deceleration, and driving close to the line. The driving behavior characteristics of the target vehicle can be determined by analyzing the operating parameters of the target vehicle. For example, the acceleration of the target vehicle satisfies T≤3s and X≥3m / s 2 In the case of the target vehicle, the driving behavior characteristic is determined to be rapid acceleration. For example, the acceleration of the target vehicle satisfies T≤3s, X≤-3m / s 2 For example, if the target vehicle satisfies T≤6s and changes lanes twice, the target vehicle's driving behavior characteristic is determined to be frequent lane changing; where T is time, s is seconds, m is meters, and m / s 2 Indicates the unit of acceleration.

[0028] Among them, the driving level specifically refers to the driver's reaction ability in various situations and the degree of attention he pays to road safety. The higher the driving maturity, the better the driver's vehicle control ability, the ability to accelerate, decelerate and turn smoothly, and the high degree of understanding and compliance with traffic rules and road signs, and the ability to accurately judge and enforce traffic rules.

[0029] It is understandable that vehicles of different types have different driving requirements when driving on the road. For example, on a highway, large trucks usually drive in the slow lane, that is, the rightmost lane. If a large truck always occupies the leftmost lane, it means that the driver of the large truck has a low driving maturity. For example, if the lane has a speed limit, the more the vehicle speed exceeds the lane speed limit, the lower the driver's driving maturity. For example, if a vehicle changes lanes without turning on the turn signal or changes lanes continuously, it means that the driver has a low degree of compliance with traffic rules, and accordingly, the driving maturity is also relatively low. Based on the above examples, it can be known that when determining the driving maturity of the target vehicle, it can be determined specifically according to the vehicle type and operating parameters of the target vehicle.

[0030] In some specific embodiments, a driving maturity model can be designed to analyze the vehicle type and driving behavior characteristics of the target vehicle and output the corresponding driving maturity; for example, the driving maturity model may include a type database and a behavior database, the type database contains attribute data of different vehicle types, including but not limited to large trucks, tankers, taxis, buses, private cars, special vehicles, trailers, etc., and the behavior database contains attribute data of different driving behavior characteristics, including but not limited to driving on the line, driving in a dragon pattern, following the vehicle too closely, driving close to the left lane, driving close to the right lane, sudden acceleration, sudden deceleration, and frequent lane changes. , changing lanes without turning on the indicator, etc. Different vehicle types are combined with different driving behavior characteristics to obtain driving maturity. For example, the driving maturity of a bus that changes lanes frequently is 20, and the driving maturity of a taxi that changes lanes frequently is 25. The vehicle volume of the bus is larger, and the vehicle volume of the taxi is smaller. Therefore, when the bus and the taxi have the same frequent lane-changing characteristics, the bus is more dangerous than the taxi, and the driving maturity of the bus will be lower than that of the taxi. The above is only an example. Based on the description of this embodiment, those skilled in the art can easily think of other specific embodiments, which should be understood to be within the scope of protection of this application.

[0031] Step S30: When the target vehicle enters the area of interest of the own vehicle, if the driving maturity of the target vehicle is greater than or equal to a first preset threshold and less than a second preset threshold, the current motion information of the target vehicle is obtained, the driving characteristic trajectory of the target vehicle is predicted based on the current motion information and the driving behavior characteristics, and a corresponding driving decision is generated based on the driving characteristic trajectory.

[0032] Specifically, the area of interest of the ego vehicle can be a preset distance range of 20 meters horizontally and 200 meters vertically centered on the ego vehicle; the driving maturity output by the driving maturity model can be any value in the range of 0 to 1, the first preset threshold can be set to 0.3, for example, and the second preset threshold can be set to 0.7, for example. If the driving maturity of the target vehicle is in the range of 0.3 to 0.7, a pre-trained trajectory prediction model can be used to predict the driving characteristic trajectory of the target vehicle based on the current motion information of the target vehicle and the driving behavior characteristics, and generate a corresponding driving decision based on the driving characteristic trajectory.

[0033] It should be noted that, unlike traditional trajectory prediction technology which predicts driving characteristic trajectories based on the historical trajectories of the vehicle, the driving characteristic trajectories predicted in this embodiment are related to the driving behavior characteristics of the target vehicle, for example Figure 2 As shown, Figure 2 The driving characteristic trajectory of the target vehicle driving along the line and the driving characteristic trajectory of the target vehicle driving across the line are shown. If the traditional trajectory prediction method is used, it is impossible to obtain Figure 2 As shown, the driving characteristic trajectory that conforms to the personalized driving characteristics of the target vehicle is obtained. Therefore, compared with the traditional passive prediction method, the method of this embodiment can actively utilize the driving maturity and driving behavior characteristics of the target vehicle to predict the driving intention of the target vehicle earlier, and obtain the driving characteristic trajectory that conforms to the personalized driving characteristics of the target vehicle, so as to make corresponding driving decisions more timely and accurately. This can improve the planning and response speed of the autonomous driving vehicle, thereby realizing the active advance prediction of the maneuvering actions of surrounding vehicles, reducing conflicts and dangers with the target vehicle, and improving the safety of autonomous driving.

[0034] In some embodiments, generating a corresponding driving decision based on the driving characteristic trajectory in step S30 specifically includes:

[0035] Step S31: When the target vehicle is in its own lane, if the driving behavior characteristic is a habit of rapid acceleration or rapid deceleration, the target vehicle's driving characteristic trajectory is a rapid acceleration trajectory or a rapid deceleration trajectory, and the generated driving decision is to add a buffer time interval;

[0036] Specifically, when the target vehicle is in the lane where the autonomous driving vehicle is located, if the driving behavior characteristics of the target vehicle show a frequent driving habit of sudden acceleration or sudden deceleration, then it can be considered that the target vehicle may perform sudden acceleration or sudden deceleration operations in future driving; in this case, the generated driving decision can be to add a buffer time interval to cope with the possible sudden acceleration or sudden deceleration behavior of the target vehicle, which means that the autonomous driving vehicle will increase the additional safety interval while maintaining a certain distance from the target vehicle, so as to respond more promptly and avoid conflict with the target vehicle, and prevent vehicle shaking; by adding a driving decision with a buffer time interval, the autonomous driving vehicle can better cope with the sudden acceleration or sudden deceleration behavior of the target vehicle, reduce dangerous situations with the target vehicle, and improve driving safety. This is also a specific driving decision strategy generated when predicting the driving characteristic trajectory of the target vehicle based on its driving behavior characteristics.

[0037] Step S32: When the target vehicle is not in the own lane, if the driving behavior characteristic is a driving habit of following the lane, the target vehicle's driving characteristic trajectory is a following the lane, and it is determined whether the target vehicle is in an adjacent lane. If so, the generated driving decision is smart evasion.

[0038] Specifically, if Figure 3 As shown in the figure, when the target vehicle is not in the lane where the autonomous vehicle is located, if the driving behavior characteristics of the target vehicle show a driving habit of frequently driving close to the line, then it can be considered that the target vehicle may continue to drive close to the line in the adjacent lane in the future. If it is determined that the target vehicle is driving in the adjacent lane, smart dodging is used as the driving decision strategy. Smart dodging means that the autonomous vehicle will choose the appropriate time and method to dodge based on the prediction of the target vehicle driving close to the line to avoid potential collision or dangerous situations with the target vehicle. The purpose of this driving decision is to maintain a safe distance from the target vehicle and adjust its own driving strategy according to the behavior of the target vehicle. Through the driving decision of smart dodging, the autonomous vehicle can better adapt to the target vehicle's driving close to the line, reduce potential conflicts with the target vehicle, and improve driving safety.

[0039] Step S33: When the target vehicle is not in the own lane, if the driving behavior characteristic is a weaving driving habit, the target vehicle's driving characteristic trajectory is a weaving driving trajectory, and it is determined whether the target vehicle is in an adjacent lane. If so, the generated driving decision is to maintain the current speed or decelerate.

[0040] Specifically, if Figure 4As shown, when the target vehicle is not in the lane where the autonomous driving vehicle is located, if the driving behavior characteristics of the target vehicle show a driving habit of frequent weaving, then it can be considered that the target vehicle may continue to weave in the future; if it is determined that the target vehicle is driving in an adjacent lane, then the driving decision can be to maintain the current speed or slow down, which means that the autonomous driving vehicle will choose to maintain the current speed or slow down according to the prediction of the interweaving driving of the target vehicle to ensure a safe interval and driving stability with the target vehicle; by making the driving decision to maintain the current speed or slow down, the autonomous driving vehicle can better cope with the weaving behavior of the target vehicle, reduce potential conflicts with the target vehicle, and improve driving safety.

[0041] In some embodiments, the step S10 further includes: obtaining a vehicle identification of the target vehicle;

[0042] The step S20 specifically includes: obtaining the driving maturity of the target vehicle according to the vehicle identification, vehicle type and driving behavior characteristics.

[0043] Specifically, the vehicle identification refers to the vehicle's identification mark, which is used to identify specific information of the vehicle, such as the trainee mark, license plate, etc.; for vehicles with trainee marks, their driving maturity is lower than that of vehicles without trainee marks, because novice drivers lack driving experience; for vehicles with out-of-town license plates, their driving maturity is lower than that of local vehicles, because out-of-town vehicles may not be familiar with local roads.

[0044] This embodiment includes trajectory prediction and driving maturity detection. The principle of this embodiment is as follows: Figure 5 As shown, a prediction model is designed. The prediction model may include a trajectory prediction model and a driving maturity model. The input of the prediction model includes the current motion information of the target vehicle (such as longitudinal position, lateral position, lateral velocity, longitudinal velocity, lateral acceleration, longitudinal acceleration, heading angle and yaw angular velocity), identification result (vehicle identification), type identification result (vehicle type) and behavior identification result (driving behavior characteristics).

[0045] In some embodiments, step S30 further includes:

[0046] When the target vehicle enters the area of interest of the vehicle, if the driving maturity of the target vehicle is less than a first preset threshold, the generated driving decision is not to follow the target vehicle and to actively change lanes to overtake.

[0047] Specifically, the autonomous driving vehicle can determine whether to follow the target vehicle based on the driving maturity of the target vehicle. If the driving maturity of the target vehicle is less than a first preset threshold, it means that the driving level of the target vehicle is low. At this time, there is a greater safety risk in following the target vehicle. Therefore, the autonomous driving vehicle can choose to actively change lanes and overtake to ensure the safety and smoothness of driving. This method can actively predict the maneuvering actions of surrounding vehicles in advance, improve the safety of autonomous driving, and better adapt to complex traffic environments.

[0048] In some embodiments, step S30 further includes:

[0049] When the target vehicle enters the area of interest of the vehicle, if the driving maturity of the target vehicle is greater than or equal to a second preset threshold, a driving decision is generated not to drive parallel to the target vehicle.

[0050] Specifically, when the driving maturity of the target vehicle is greater than or equal to the second preset threshold, it indicates that the target vehicle has a higher or more mature driving ability, which means that the driver of the vehicle has higher driving skills, experience and reaction ability, and can better adapt to various driving situations and deal with potential risks. Therefore, it is only necessary to avoid driving parallel to the target vehicle to avoid possible accidents or traffic conflicts.

[0051] In some embodiments, step S10 further includes:

[0052] Obtain obstacle information from the vehicle's surroundings, input the obstacle information into a pre-trained target detection model for target detection, obtain the target vehicle's type, driving behavior characteristics, and vehicle identification, and label the target vehicle with a corresponding label;

[0053] The attributes of the tag are set to the driving maturity and driving behavior characteristics of the target vehicle, and the tag and the attributes of the target vehicle are stored.

[0054] Specifically, obstacle information about the ego vehicle's surroundings refers to information about other vehicles, pedestrians, or other obstacles detected within the region of interest around the ego vehicle, such as operating parameters such as position, size, and speed. The object of interest detection model analyzes and identifies the obstacle information and identifies obstacles that meet preset conditions as objects of interest, such as vehicles within a certain distance range or vehicles moving faster behind the ego vehicle. In specific applications, the preset conditions can be set according to actual technical requirements, and objects of no interest are ignored.

[0055] In this embodiment, a corresponding label is marked on the target vehicle to facilitate distinguishing different vehicles; at the same time, the self-driving vehicle stores the label and attributes of the target vehicle, that is, the self-driving vehicle is given a memory function for the target vehicle. When a vehicle carrying a label enters the area of interest of the self-driving vehicle, the self-driving vehicle can obtain the attributes of the vehicle's label, that is, quickly obtain the driving maturity and driving behavior characteristics of the vehicle, and then execute the above steps S20 and S30 to make timely driving decisions for the target vehicle. This can further improve the planning and response speed of the autonomous driving vehicle, which is conducive to better realizing the proactive and early prediction of the maneuvering actions of surrounding vehicles, further reducing conflicts and dangers between the self-driving vehicle and the target vehicle, and improving the safety of autonomous driving.

[0056] In some embodiments, the method further comprises:

[0057] When the storage time of the target vehicle's label and its attributes reaches a preset time threshold, or the cumulative mileage of the vehicle from the moment the target vehicle's label and its attributes are stored reaches a preset mileage threshold, the target vehicle's label and its attributes will be deleted.

[0058] Specifically, the storage duration refers to the duration for which the target vehicle's label and attributes are stored. If the storage duration reaches a preset duration threshold, the vehicle will delete the label and attribute information of the target vehicle. For example, the duration threshold is 60 minutes. The vehicle will accumulate its mileage from the moment the target vehicle's label and attributes are stored. If the preset mileage threshold is reached, the vehicle will delete the label and attribute information of the target vehicle. For example, the mileage threshold is 100 kilometers. The label and attributes of the target vehicle are deleted to maintain the effective use of the vehicle's storage space or to avoid the impact of outdated information on system decisions.

[0059] In some embodiments, the method further comprises:

[0060] When the target vehicle does not enter the area of interest of the ego vehicle, the historical trajectory of the target vehicle is obtained, the future trajectory of the target vehicle is predicted based on the historical trajectory, and the ego vehicle path is planned based on the future trajectory.

[0061] Specifically, the historical trajectory refers to the driving trajectory of the target vehicle in the past period of time. By collecting the historical position data of the target vehicle, the trajectory of the target vehicle can be constructed. Using the historical trajectory data of the target vehicle and related prediction algorithms, the future trajectory of the target vehicle can be predicted. In this way, the possible driving path and direction of the target vehicle can be inferred. Based on the predicted trajectory of the target vehicle, the self-vehicle can perform path planning to determine the optimal path or avoid collision with the target vehicle.

[0062] Figure 6 Shown is a flowchart of a specific embodiment of the present application. Figure 6 The steps described in the above embodiments are shown. Figure 6 It can help understand the above embodiments.

[0063] Another embodiment of the present application further proposes an autonomous driving prediction and decision-making device, comprising a module for executing the autonomous driving prediction and decision-making method according to the above embodiment, for example Figure 7 Shown, including:

[0064] Information acquisition module 1, used to execute step S1 in the above embodiment method;

[0065] A driving level assessment module 2, configured to execute step S2 in the above-mentioned embodiment method; and

[0066] The driving decision module 3 is used to execute step S3 in the method of the above embodiment.

[0067] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of this embodiment.

[0068] It should be noted that the apparatus of the above embodiment corresponds to the method of the above embodiment. Therefore, the undescribed parts of the apparatus of the above embodiment can be obtained by referring to the contents of the method of the above embodiment, and will not be repeated here.

[0069] Furthermore, if the autonomous driving prediction and decision-making device of the above-mentioned embodiment 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.

[0070] Another embodiment of the present application further proposes a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the autonomous driving prediction decision-making method described in the above embodiment.

[0071] Specifically, the computer-readable storage medium may include: any entity or device that can carry the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal and software distribution medium, etc.

[0072] While various embodiments of the present application have been described above, the above descriptions are illustrative, non-exhaustive, and not intended to be limiting of the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is selected to best explain the principles of the embodiments, their practical applications, or technological improvements in the marketplace, or to enable others skilled in the art to understand the embodiments disclosed herein.

[0073] While various embodiments of the present application have been described above, the above descriptions are illustrative, non-exhaustive, and not intended to be limiting of the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is selected to best explain the principles of the embodiments, their practical applications, or technological improvements in the marketplace, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. An automatic driving prediction decision method, characterized in that: The method comprises: Acquiring information about the surrounding environment of the own vehicle, and obtaining the vehicle type and driving behavior characteristics of the target vehicle based on the information about the surrounding environment of the own vehicle; Obtaining the driving maturity of the target vehicle according to the vehicle type and driving behavior characteristics; wherein the driving maturity is an index value used to evaluate the driving level of the driver; When the target vehicle enters the area of interest of the own vehicle, if the driving maturity of the target vehicle is greater than or equal to a first preset threshold and less than a second preset threshold, the current motion information of the target vehicle is obtained, the driving characteristic trajectory of the target vehicle is predicted based on the current motion information and the driving behavior characteristics, and a corresponding driving decision is generated based on the driving characteristic trajectory.

2. The method according to claim 1, characterized in that Generating a corresponding driving decision according to the driving characteristic trajectory specifically includes: When the target vehicle is in its own lane, if the driving behavior characteristic is a habit of rapid acceleration or rapid deceleration, the target vehicle's driving characteristic trajectory is a rapid acceleration trajectory or a rapid deceleration trajectory, and the generated driving decision is to add a buffer time interval; When the target vehicle is not in the own lane, if the driving behavior characteristic is a driving habit of following the lane, the target vehicle's driving characteristic trajectory is a following the lane, and it is determined whether the target vehicle is in an adjacent lane. If so, the generated driving decision is smart evasion; When the target vehicle is not in the current lane, if the driving behavior characteristic is a weaving driving habit, the driving characteristic trajectory of the target vehicle is an weaving driving trajectory, and it is determined whether the target vehicle is in an adjacent lane. If so, the driving decision generated is to maintain the current speed or decelerate.

3. The method according to claim 1, characterized in that The method specifically includes: Obtain the vehicle identification of the target vehicle; The driving maturity of the target vehicle is obtained according to the vehicle identification, vehicle type and driving behavior characteristics.

4. The method according to claim 1, wherein The method further comprises: When the target vehicle enters the area of interest of the vehicle, if the driving maturity of the target vehicle is less than a first preset threshold, the generated driving decision is not to follow the target vehicle and to actively change lanes to overtake.

5. The method according to claim 1, wherein The method further comprises: When the target vehicle enters the area of interest of the vehicle, if the driving maturity of the target vehicle is greater than or equal to a second preset threshold, a driving decision is generated not to drive parallel to the target vehicle.

6. The method according to claim 1, characterized in that The method further comprises: Obtain obstacle information from the vehicle's surroundings, input the obstacle information into a pre-trained target detection model for target detection, obtain the target vehicle's type, driving behavior characteristics, and vehicle identification, and label the target vehicle with a corresponding label; The attributes of the tag are set to the driving maturity and driving behavior characteristics of the target vehicle, and the tag and the attributes of the target vehicle are stored.

7. The method according to claim 6, characterized in that The method further comprises: When the storage time of the target vehicle's label and its attributes reaches a preset time threshold, or the cumulative mileage of the vehicle from the moment the target vehicle's label and its attributes are stored reaches a preset mileage threshold, the target vehicle's label and its attributes will be deleted.

8. The method according to any one of claims 1 to 7, characterized in that The method further comprises: When the target vehicle does not enter the area of interest of the ego vehicle, the historical trajectory of the target vehicle is obtained, the future trajectory of the target vehicle is predicted based on the historical trajectory, and the ego vehicle path is planned based on the future trajectory.

9. An automatic driving prediction and decision-making device, characterized in that: The method comprises a module for executing the autonomous driving prediction decision method according to any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the automatic driving prediction and decision-making method according to any one of claims 1 to 8 is implemented.

Images