Lane change prediction method and system, and vehicle

Abstract

A lane change prediction method and system, and a vehicle. The method comprises: when a vehicle is in a travelling state, periodically determining a leftward lane change probability, a rightward lane change probability and a straight-ahead movement probability of the vehicle on the basis of one or more feature signals, the one or more feature signals being feature signal(s) used for representing the travelling state of the vehicle; and when the leftward lane change probability or the rightward lane change probability is less than a first threshold and greater than a second threshold, and the turn signal of the vehicle is in an inactive state, sending a first reminder to a user, the first reminder being used for reminding the user to activate the turn signal. According to the method and apparatus, and the vehicle, lane change prediction is performed on the basis of various effective features, so that lane change prediction accuracy can be improved, a lane change preparation stage is involved in the lane change prediction, and a user can be reminded to turn on a turn signal in the lane change preparation stage, so that the forward driving behavior of the user can be guided, and the safety accident occurrence rate can also be reduced.

Description

Lane change prediction method, system and vehicle

[0001] This application claims priority to the Chinese patent application filed with the China Patent Office on September 28, 2023, with application number 202311282230.7 and application name “Method, system and vehicle for lane change prediction”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of vehicles, and more specifically, to a method, system and vehicle for lane change prediction. Background Art

[0003] For vehicles traveling on the road, if the vehicle does not turn on the turn signal when changing lanes, it will not be able to signal the lane change intention to other vehicles in the lane and the target lane. The deceleration during lane change may very likely cause the vehicle behind in the lane to rear-end. In addition, for other vehicles in the target lane, if the vehicle in the adjacent lane changes lanes to the target lane without turning on the turn signal, it is easy for other vehicles in the target lane to crash due to the inability to make the expected response. Moreover, a considerable number of users do not turn on the turn signal after completing the lane change, which can easily cause confusion to vehicles in the lane and the adjacent lane, resulting in a decrease in traffic efficiency.

[0004] Summary of the Invention

[0005] The present application provides a method, system, and vehicle. Through this method, system, and vehicle, lane change prediction is performed based on multiple effective features, which can improve the accuracy of lane change prediction. In addition, the lane change prediction can cover the lane change preparation stage and remind the user to turn on the turn signal during the lane change preparation stage. This can not only guide the user's positive driving behavior, but also reduce the occurrence rate of safety accidents.

[0006] In a first aspect, a method for lane change prediction is provided, the method comprising: when a vehicle is in a driving state, periodically determining a left lane change probability, a right lane change probability, and a straight-going probability of the vehicle based on one or more characteristic signals, the one or more characteristic signals being characteristic signals used to characterize the driving state of the vehicle; when the left lane change probability or the right lane change probability is less than a first threshold and greater than a second threshold, and the turn signal of the vehicle is in an inactivated state, issuing a first reminder to the user, the first reminder being used to remind the user to activate the turn signal, wherein the second threshold is less than the first threshold.

[0007] For example, the first threshold and the second threshold may be turning thresholds.

[0008] Lane change prediction based on multiple effective features can improve the accuracy of lane change prediction. Lane change prediction can also cover the lane change preparation stage, and can remind users to turn on the turn signal during the lane change preparation stage. This can not only guide users' positive driving behavior, but also reduce the occurrence rate of safety accidents.

[0009] In combination with the first aspect, in certain implementations of the first aspect, the method further includes: when the left lane change probability or the right lane change probability is greater than a first threshold and the turn signal of the vehicle is in an inactivated state, activating the turn signal of the vehicle.

[0010] In an embodiment of the present application, when the user does not turn on the turn signal during the lane change preparation stage and detects that the current vehicle enters the lane change execution stage, the vehicle system can automatically turn on the corresponding turn signal to provide emergency alerts to surrounding vehicles, thereby reducing the frequency of safety accidents.

[0011] In combination with the first aspect, in certain implementations of the first aspect, the method further includes: when the straight-moving probability of the vehicle is greater than a third threshold and the turn signal of the vehicle is in an activated state, switching the state of the turn signal of the vehicle to an inactivated state.

[0012] For example, the third threshold may be a straight-moving threshold.

[0013] In an embodiment of the present application, when it is detected that the user has not turned on the indicator light after the lane change is completed, the indicator light can be automatically turned on to avoid confusing vehicles in the lane and the adjacent lane, thereby avoiding a decrease in traffic efficiency.

[0014] In combination with the first aspect, in certain implementations of the first aspect, the method also includes: in response to a user clicking or pressing a first control of the vehicle, determining the vehicle's left lane change probability, right lane change probability and straight-going probability based on the one or more characteristic signals; when it is determined that the vehicle's left lane change probability is greater than a second threshold or greater than the first threshold, activating the vehicle's left turn signal; or, when it is determined that the right lane change probability is greater than the second threshold or greater than the first threshold, activating the vehicle's right turn signal; and, when it is determined that the vehicle's straight-going probability is greater than a third threshold, turning off the left turn signal or the right turn signal.

[0015] The first control may be, for example, a control located on a steering wheel, and the function of the first control may be a turn signal triggering function.

[0016] In addition, the vehicle's left lane change probability, right lane change probability, and straight-going probability can be determined based on the one or more characteristic signals in response to voice commands or gesture commands issued by the user. This application does not limit this.

[0017] In an embodiment of the present application, the user's operation can be combined with the lane change prediction method provided by the present application. For example, the vehicle's left lane change probability, right lane change probability and straight-ahead probability can be determined based on one or more characteristic signals based on the user's operation, and subsequent operations can be further performed. Based on this, the vehicle can be controlled to turn on the left turn signal, the right turn signal or the return signal through a control, which can avoid the user pressing the wrong steering control and improve the user's operating convenience.

[0018] In combination with the first aspect, in some implementations of the first aspect, the method further includes: periodically acquiring the one or more characteristic signals.

[0019] In combination with the first aspect, in certain implementations of the first aspect, the left lane change probability, right lane change probability and straight-going probability of the vehicle are periodically determined based on one or more characteristic signals, including: periodically determining one or more lane change results based on one or more characteristic signals, the lane change results including left lane change, right lane change or straight-going; determining the left lane change probability, right lane change probability and straight-going probability of the vehicle based on the one or more lane change results.

[0020] In the embodiment of the present application, since the acquisition of characteristic signals is carried out in real time, within each cycle, for a characteristic signal, the number of characteristic signal values ​​obtained is one or more. Each time a characteristic signal value is obtained, a lane change result can be determined. Therefore, within each cycle, one or more lane change results can be obtained, and the vehicle's left lane change probability, right lane change probability and straight-going probability can be further determined based on the one or more lane change results.

[0021] In combination with the first aspect, in certain implementations of the first aspect, the one or more characteristic signals include a distance characteristic signal of the vehicle relative to the adjacent lane line, and / or, an angle characteristic signal of the vehicle relative to the adjacent lane line.

[0022] In some examples, the vehicle can determine the current lane based on the adjacent lane markings, which may include a left lane marking and a right lane marking. The adjacent lane markings may include the left lane marking or the right lane marking; for example, the right lane marking is obscured by another vehicle.

[0023] In some implementations, a vehicle is traveling in a first lane of two lanes from left to right. The vehicle can detect two lane markings that define the first lane. The two lane markings can serve as adjacent lane markings.

[0024] In some implementations, a vehicle is traveling in a first lane of two lanes from left to right. The vehicle may detect three lane markings that define the first lane and the second lane. The three lane markings may serve as adjacent lane markings.

[0025] In certain implementations, a vehicle is traveling in the first lane of a two-lane road, moving from left to right. Ahead of the vehicle, the two lanes become three lanes, with a third lane to the left of the first lane. As the vehicle moves from the two lanes to the three lanes, it moves from the first lane to the third lane. The vehicle can detect three lane markings in the first and third lanes, which can serve as adjacent lane markings.

[0026] In some implementations, there may be two lane lines between the first lane and the second lane, such as double solid lines or double dashed lines, etc. In this case, there may be four lane lines between the first lane and the second lane.

[0027] In some implementations, the characteristic signal of the distance of the vehicle relative to the adjacent lane line includes any one or more of the following characteristic signals: lane width; the distance between the vehicle's forward-looking camera and the ringing lane line; the distance between the vehicle's wheels and the adjacent lane line; the distance between the vehicle's center of mass and the adjacent lane line; the distance between the vehicle's centroid and the adjacent lane line.

[0028] In some implementations, the angular characteristic signal of the vehicle relative to the adjacent lane line includes any one or more of the following characteristic signals: the angle between the central axis of the vehicle's forward-looking camera and the adjacent lane line; the head turning angle of the driver driving the vehicle; and the wheel speed difference between the wheels of the vehicle.

[0029] In addition, one or more characteristic signals may also include other characteristic signals with properties similar or close to the above-mentioned characteristic signals, which is not limited in this application.

[0030] In the embodiment of the present application, lane change prediction is performed based on multiple effective features, which can improve the accuracy of lane change prediction and enable the lane change prediction to cover the lane change preparation stage, lane change execution stage and lane change completion stage.

[0031] In combination with the first aspect, in certain implementations of the first aspect, the head angle of the driver driving the vehicle includes any one or more of the following angles: the driver's angle on the XOY plane, the driver's angle on the XOZ plane, and the driver's angle on the YOZ plane in a spatial rectangular coordinate system established with the center of the driver's facial projection as the origin O, the driver's forward viewing direction as the Z axis, and the horizontal direction as the X axis.

[0032] In combination with the first aspect, in certain implementations of the first aspect, the wheel speed difference between the wheels of the vehicle includes any one or more of the following wheel speed differences: the wheel speed difference between the left front wheel and the right front wheel of the vehicle; the wheel speed difference between the left rear wheel and the right rear wheel of the vehicle; the wheel speed difference between the left front wheel and the left rear wheel of the vehicle; the wheel speed difference between the right front wheel and the right rear wheel of the vehicle; the wheel speed difference between the left front wheel and the right rear wheel of the vehicle; and the wheel speed difference between the left rear wheel and the right front wheel of the vehicle.

[0033] Among them, this implementation method can be applicable to four-wheel vehicles, for example. When the number of wheels of the vehicle changes, the above-mentioned corresponding characteristic signals may also change. For example, characteristic signals such as the wheel speed difference between the wheels of the vehicle, the distance between the wheels of the vehicle and the lane line, etc., when the number of wheels is different, the number of corresponding characteristic signals will also be different.

[0034] In combination with the first aspect, in certain implementations of the first aspect, determining the vehicle's left lane change probability, right lane change probability and straight-going probability based on one or more characteristic signals includes: determining one or more time domain feature sets based on the one or more characteristic signals, the one or more time domain feature sets corresponding one-to-one to the one or more characteristic signals, the time domain features including one or more of the mean, standard deviation, median, maximum value, root mean square value, crest factor, shape factor, impact factor and margin factor; determining the vehicle's left lane change probability, right lane change probability and straight-going probability based on the one or more time domain feature sets.

[0035] In one example, taking the first characteristic signal among one or more characteristic signals as an example, multiple first characteristic signals within a characteristic acquisition period are first obtained, and then the time domain characteristics corresponding to the multiple first characteristic signals (one or more of the mean, standard deviation, median, maximum value, root mean square value, crest factor, shape factor, impact factor, and margin factor) are calculated, and then the vehicle's left lane change probability, right lane change probability, and straight-going probability are determined based on the time domain characteristics corresponding to the multiple first characteristic signals.

[0036] In combination with the first aspect, in certain implementations of the first aspect, the issuing of the first reminder to the user includes: issuing the first reminder to the user by voice; displaying the first reminder on the vehicle's display screen; or issuing the first reminder to the user by vibration.

[0037] In addition, the vehicle can also send a first reminder to the user in other ways, such as: through vibration, emitting a special smell, lighting, etc., which is not limited in this application.

[0038] In a second aspect, a device is provided, comprising: one or more processors; one or more memories; and one or more computer programs, wherein the one or more computer programs are stored in the one or more memories, and the one or more computer programs include instructions that, when executed by the one or more processors, cause the device to perform the method in the above-mentioned first aspect or any possible implementation of the first aspect.

[0039] In a third aspect, a vehicle-mounted electromechanical system is provided, which includes: one or more processors; one or more memories; and one or more computer programs, wherein the one or more computer programs are stored in the one or more memories, and the one or more computer programs include instructions. When the instructions are executed by the one or more processors, the vehicle-mounted electromechanical system executes the method in the above-mentioned first aspect or any possible implementation of the first aspect.

[0040] In a fourth aspect, a vehicle is provided, comprising an on-board electromechanical system and a steering system, wherein the on-board electromechanical system is used to control the steering system to implement the method in the above-mentioned first aspect or any possible implementation of the first aspect.

[0041] In a fifth aspect, a computer-readable storage medium is provided, in which a computer program or instruction is stored. When the computer program or instruction is executed, the method in the above-mentioned first aspect or any possible implementation method of the first aspect is implemented.

[0042] In a sixth aspect, a chip is provided, in which instructions are stored. When the instructions are executed on a device, the chip executes the method in the above-mentioned first aspect or any possible implementation of the first aspect.

[0043] In a seventh aspect, a computer program product comprising instructions is provided, which, when run on a computer, enables the computer to execute the method in the above-mentioned first aspect or any possible implementation of the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] FIG1 is a schematic functional block diagram of a vehicle provided in an embodiment of the present application;

[0045] FIG2 is a schematic diagram of a vehicle lane changing process provided by an embodiment of the present application;

[0046] FIG3 is a schematic diagram of a lane change prediction method provided by an embodiment of the present application;

[0047] FIG4 is a schematic flow chart of a lane change prediction method provided by an embodiment of the present application;

[0048] FIG5 is a method for obtaining a vehicle's straight-moving probability, left lane-changing probability, and right lane-changing probability, provided by an embodiment of the present application;

[0049] FIG6 is a schematic diagram showing one or more characteristic signals in a coordinate system established based on the vehicle's posture while taking into account the vehicle's yaw angle, provided by an embodiment of the present application;

[0050] FIG7 is a schematic diagram showing one or more characteristic signals in a coordinate system based on a lane centerline without considering the yaw angle of the vehicle, provided by an embodiment of the present application;

[0051] FIG8 is a schematic diagram of a head rotation angle provided by an embodiment of the present application;

[0052] FIG9 is a schematic diagram of the layout of two steering wheels provided in an embodiment of the present application;

[0053] FIG10 is a schematic flowchart of a lane change prediction device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0054] The technical solution in this application will be described below with reference to the accompanying drawings.

[0055] The terms used in the following embodiments are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the specification of this application and the appended claims, the singular expressions "a", "an", "the", "above", "the" and "this" are intended to also include expressions such as "one or more", unless there is a clear contrary indication in the context. It should also be understood that in the following embodiments of the present application, "at least one", "one or more" refer to one, two or more. The term "and / or" is used to describe the association relationship of associated objects, indicating that three relationships can exist; for example, A and / or B can mean: 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 subsequent associated objects are in an "or" relationship.

[0056] References to "one embodiment" or "some embodiments" in this specification mean that a particular feature, structure, or characteristic described in conjunction with that embodiment is included in one or more embodiments of the present application. Thus, phrases such as "in one embodiment," "in some embodiments," "in other embodiments," and "in yet other embodiments" appearing in various places in this specification do not necessarily refer to the same embodiment, but rather mean "one or more but not all embodiments," unless otherwise specifically emphasized. The terms "including," "comprising," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0057] FIG1 is a schematic functional block diagram of a vehicle 100 provided in an embodiment of the present application. As shown in FIG1 , the vehicle 100 may include a perception system 120, a mapping module 130, a peripheral device 140, and a computing platform 150.

[0058] The perception system 120 may include several sensors for sensing information about the environment surrounding the vehicle 100. For example, the perception system 120 may include a positioning system, which may be one or more of a global positioning system (GPS) 121, a BeiDou system, or other positioning systems, an inertial measurement unit (IMU) 122, a laser radar 123, a millimeter-wave radar 124, an ultrasonic radar 125, and a camera 126.

[0059] The vehicle 100 may include a mapping module 130. The mapping module 130 may use techniques such as object recognition algorithms, structure from motion (SFM) algorithms, video tracking, and simultaneous localization and mapping (SLAM) to map the environment.

[0060] The vehicle 100 interacts with a mobile terminal, external sensors, other vehicles, other computer systems, or users through the peripheral devices 140. The peripheral devices 140 may include a wireless communication system 141, an onboard screen 142, a microphone 143, and / or a speaker 144.

[0061] In some possible implementations, the peripheral devices 140 provide a means for the user of the vehicle 100 to interact with the user interface 160. For example, the in-vehicle screen 142 can provide information to the user of the vehicle 100. The user interface 160 can also operate the in-vehicle screen 142 to receive user input. The in-vehicle screen 142 can be operated via a touch screen. In other cases, the peripheral devices 140 can provide a means for the vehicle 100 to communicate with other devices located within the vehicle. For example, the microphone 143 can receive audio (e.g., voice commands or other audio input) from the user of the vehicle 100. Similarly, the speaker 144 can output audio to the user of the vehicle 100.

[0062] Some or all functions of vehicle 100 may be controlled by computing platform 150. Computing platform 150 may include processors 151 to 15n (n is a positive integer). A processor is a circuit capable of processing signals. In one implementation, the processor may be a circuit capable of reading and executing instructions, such as a central processing unit (CPU), a microprocessor, a graphics processing unit (GPU) (which can be understood as a microprocessor), or a digital signal processor (DSP). In another implementation, the processor may implement certain functions through the logical relationships of hardware circuits. The logical relationships of the hardware circuits may be fixed or reconfigurable. For example, the processor may be a hardware circuit implemented by an application-specific integrated circuit (ASIC) or a programmable logic device (PLD), such as an FPGA. In a reconfigurable hardware circuit, the process of the processor loading a configuration file to implement the hardware circuit configuration can be understood as the process of the processor loading instructions to implement the functions of some or all of the above units. In addition, it can also be a hardware circuit designed for artificial intelligence, which can be understood as an ASIC, such as a neural network processing unit (NPU), a tensor processing unit (TPU), a deep learning processing unit (DPU), etc. In addition, the computing platform 150 can also include a memory for storing instructions. Some or all of the processors 151 to 15n can call the instructions in the memory and execute them to achieve corresponding functions.

[0063] Alternatively, one or more of these components may be installed or associated separately from the vehicle 100. The components may be communicatively coupled together in a wired and / or wireless manner.

[0064] Optionally, the above components are only an example. In actual applications, the components in the above modules may be added or deleted according to actual needs. Figure 1 should not be understood as a limitation to the embodiments of the present application.

[0065] Optionally, the vehicle 100 may include one or more different types of vehicles, or may include one or more different types of transportation vehicles or movable objects that operate or move on land (e.g., roads, highways, railways, etc.), water (e.g., waterways, rivers, oceans, etc.), or space. For example, the vehicle may include a car, a bicycle, a motorcycle, a train, a subway, an airplane, a ship, an aircraft, a robot, or other types of transportation vehicles or movable objects, etc., and the embodiments of the present application are not limited to this.

[0066] As people's living standards improve, more and more families own cars, which also leads to an increasing number of vehicles on the road. According to statistics, among the vehicles on the road, more than half of them do not use turn signals when changing lanes. The habit of not using turn signals when changing lanes may seem insignificant, but it poses a serious safety hazard. Without using turn signals when changing lanes, it is impossible to send a lane change expectation to other vehicles in the lane and the target lane. The deceleration when changing lanes may very likely cause a rear-end collision with the vehicle behind in the lane. In addition, for other vehicles in the target lane, if the vehicle in the adjacent lane changes lanes to the target lane without using turn signals, it is easy for other vehicles in the target lane to crash due to the inability to make the expected response. Moreover, a considerable number of users do not turn on the turn signal after completing the lane change, which can easily cause confusion to vehicles in the lane and the adjacent lane, resulting in a decrease in traffic efficiency.

[0067] In order to more easily understand the lane-changing assistance method provided in the present application, the lane-changing process of a vehicle while driving is first introduced with reference to FIG. 2 .

[0068] Figure 2 shows a schematic diagram of a vehicle lane changing process provided by an embodiment of the present application. As shown in Figure 2, the vehicle's lane changing process can be divided into a lane changing intention generating stage, a lane changing preparation stage and a lane changing execution stage. Among them, the lane changing intention generating stage and the lane changing preparation stage are completed in the current lane (i.e., lane 1), and the lane changing execution stage is completed in the current lane and the target lane (i.e., lane 2). The lane changing execution stage can be specifically subdivided into a lane changing stage, a target lane reaching stage and a vehicle body adjusting stage.

[0069] Among them, the lane change intention generation stage refers to the stage when the driver of the vehicle generates the lane change intention. This stage is the stage before time T1, and it can also be understood that time T1 is the moment when the driver generates the lane change intention; the lane change preparation stage refers to the stage when the vehicle prepares to change lanes from the current lane to the target lane. For example, the vehicle can turn on the turn signal in advance and approach the target lane. The stage from T1 to T2 in Figure 2 is the lane change preparation stage; the lane change execution stage is the stage when the vehicle executes the operation of changing lanes from the current lane to the target lane. The stage from T2 to T4 in Figure 2 is the lane change execution stage. Specifically, the stage from T2 to T3 is the lane change stage, the stage from T3 to T4 is the stage of reaching the target lane, and the stage after T4 is the stage of adjusting the vehicle body. Time T2 is the moment when the lane change to the target lane begins, time T3 is the moment when the lane change begins to cross the dividing line between the current lane and the target lane to enter the target lane, and time T4 is the moment when the lane change is completed.

[0070] As shown in Figure 2, the process of vehicle 200 changing lanes from lane 1 to lane 2 is specifically as follows: after the driver of vehicle 200 has the intention to change lanes to lane 2 at time T1, he enters the pre-lane change preparation stage, that is, he performs a surrounding vehicle environment detection. When the surrounding vehicle environment detection result meets the lane change conditions, he starts to maneuver vehicle 200 to approach lane 2 and turns on the turn signal. When the driver completes the pre-lane change preparation (that is, time T2), he starts to maneuver vehicle 200 to cross the dividing line between lane 1 and lane 2 to change lanes to lane 2. Time T3 is the moment when the driver starts to cross the dividing line between lane 1 and lane 2. After vehicle 200 successfully changes lanes to lane 2, the turn signal is turned back on, that is: the turn signal is turned back on at time T4, and the vehicle body is adjusted so that vehicle 200 can drive normally in lane 2.

[0071] It can be seen that the lane change process of a vehicle includes the following four key moments:

[0072] T1: the moment when the driver intends to change lanes;

[0073] T2: When the driver completes the vehicle environment inspection and prepares to change lanes, the driver should turn on the turn signal before T2;

[0074] T3: The moment when the driver begins to cross the dividing line between the own lane and the target lane;

[0075] T4: The moment when the driver successfully completes the lane change. After T4, the driver should turn on the indicator light.

[0076] Therefore, it is precisely because a considerable number of drivers fail to turn on their turn signals before T2 that the vehicles behind them and in the target lane cannot anticipate the vehicle's lane change in advance, which can easily lead to safety accidents such as rear-end collisions and collisions. In addition, many users fail to turn on their turn signals in time after T4, which can easily cause confusion to vehicles in their lane and the adjacent lane, resulting in a decrease in traffic efficiency.

[0077] For example, FIG3 shows a schematic diagram of a lane change prediction method 300. As shown in FIG3, the method 300 includes:

[0078] S301: When the vehicle is in motion, if it is detected that the vehicle is about to cross a lane line, step S302 is executed.

[0079] In some examples, when a vehicle camera recognizes that the vehicle is about to leave the current lane, it is determined that the vehicle is about to cross the lane line.

[0080] S302: Determine whether the steering wheel angle is greater than a first angle threshold. If so, further execute step S303.

[0081] In some examples, the steering wheel angle refers to the steering wheel rotation angle of the driver, that is, the steering wheel rotation angle relative to the return angle after the steering wheel is turned.

[0082] S303: Determine whether there are other vehicles near the vehicle, and if so, assist in activating the vehicle's turn signal.

[0083] In some examples, assisting in activating the vehicle's turn signal is performed when it is determined that there are other vehicles near the vehicle and the vehicle's turn signal is not activated.

[0084] In this method, lane change prediction only covers the stage when the lane change is almost completed, that is, time T3 (the moment when the driver starts to cross the dividing line between the own lane and the target lane). Therefore, the vehicle can only make an emergency response of turning on the turn signal before crossing the lane. The turn signal is not turned on in advance before crossing the lane to remind other vehicles in the own lane and vehicles in the target lane. Forward driving behavior is not encouraged, and it is easy for other vehicles in the own lane and vehicles in the target lane to cause rear-end collisions or other traffic accidents because they do not have time to react. This method also does not have a mechanism to turn on the light when the lane change is completed. If the light is not turned on in time after the lane change is completed, it is easy to cause confusion to vehicles in the own lane and the adjacent lanes, resulting in a decrease in traffic efficiency.

[0085] Furthermore, this method only predicts lane changes based on data collected by the vehicle's camera and the steering wheel angle. However, lanes are typically only 2.3-3.75 meters wide. Even when driving straight, the vehicle often sways from side to side, appearing to be about to cross the lane line. The steering wheel angle may also change during lane keeping. For example, when driving on a curve, especially at high speeds, the steering wheel angle during lane changes is very small, making it difficult to distinguish whether the steering wheel angle change is for lane change or for driving on a curve. Therefore, it is virtually impossible to correctly predict the vehicle's lane change behavior based on the steering wheel angle. This means that this method has a very low accuracy rate for lane change predictions.

[0086] It can be seen that in the existing lane change prediction methods, lane change prediction only covers the moment when the vehicle crosses the lane line, does not encourage forward driving behavior, and poses a safety hazard; and lane change prediction is only based on camera data and steering wheel angle, and the prediction results are inaccurate.

[0087] In view of this, embodiments of the present application provide a lane change prediction method, device, and vehicle. This method predicts whether a vehicle will change lanes based on multiple valid features, accurately predicts the vehicle's lane change behavior, and can remind the user to turn on the turn signal before executing the lane change. It can also guide the user to execute the lane change after the turn signal has been on for a specified period of time. This not only guides the user's positive driving behavior, but also sends a lane change expectation to vehicles in the lane and adjacent lanes in advance, thereby reducing the frequency of safety accidents. The method can also automatically turn on the turn signal if it detects that the user has not turned on the turn signal after the lane change is completed, avoiding confusion for vehicles in the lane and adjacent lanes, thereby avoiding a decrease in traffic efficiency.

[0088] For example, FIG4 shows a schematic flow chart of a lane change prediction method 400 provided in an embodiment of the present application. As shown in FIG4 , the method 400 includes:

[0089] S401: When the vehicle is in a driving state, the straight-moving probability, left lane-changing probability, and right lane-changing probability of the vehicle are obtained.

[0090] In some embodiments, the vehicle's straight-moving probability is represented by the vehicle's straight-moving confidence, the vehicle's left lane-changing probability is represented by the vehicle's left lane-changing confidence, and the vehicle's right lane-changing probability is represented by the vehicle's right lane-changing confidence.

[0091] In some examples, the vehicle's straight-going probability, left lane-changing probability, and right lane-changing probability may be expressed as a percentage, a fraction, or a decimal.

[0092] It should be understood that there is a correlation between the vehicle's straight-going probability, left lane-changing probability and right lane-changing probability. When the vehicle's straight-going probability is high, the vehicle's left lane-changing probability and right lane-changing probability are relatively low; when the vehicle's left lane-changing probability is high, the vehicle's straight-going probability and right lane-changing probability are relatively low. For example, when the vehicle changes lanes left, the left lane-changing probability > straight-going probability > right lane-changing probability; when the vehicle's right lane-changing probability is high, the vehicle's straight-going probability and left lane-changing probability are relatively low. For example, when the vehicle changes lanes right, the right lane-changing probability > straight-going probability > left lane-changing probability.

[0093] In some examples, the vehicle's straight-going probability, left lane change probability, and right lane change probability are determined based on one or more signal features acquired during vehicle driving for characterizing the vehicle's driving state. Optionally, the vehicle includes a prediction model whose input is one or more signal features, and whose output is the vehicle's straight-going probability, left lane change probability, and right lane change probability. Alternatively, the output of the prediction model is a decision made based on the vehicle's straight-going probability, left lane change probability, and right lane change probability, such as a decision to remind the user to turn on the turn signal, a decision to control automatic activation or return of the turn signal, etc.

[0094] It should be understood that step S401 is not executed only once, but is executed in real time or periodically while the vehicle is in motion. For example, while the vehicle is driving, the vehicle's straight-moving probability, left lane-changing probability, and right lane-changing probability are determined based on one or more signal features used to characterize the vehicle's driving state acquired in a first acquisition cycle. After acquiring one or more signal features used to characterize the vehicle's driving state in a second acquisition cycle, the vehicle's straight-moving probability, left lane-changing probability, and right lane-changing probability are re-determined based on one or more signal features used to characterize the vehicle's driving state acquired in the second acquisition cycle.

[0095] The explanation of step S401 will be described in detail in subsequent embodiments.

[0096] In some embodiments, when the vehicle is in a driving state, in response to a first operation of the user, a straight-moving probability, a left lane-changing probability, and a right lane-changing probability of the vehicle are obtained.

[0097] In some examples, the user's first operation may be the user pressing a first control, which is used to trigger the acquisition of the vehicle's straight-going probability, left lane change probability, and right lane change probability; in addition, the first operation may also be the user issuing a voice command or gesture command, etc., which is not limited in this application.

[0098] S402: Determine whether the probability of changing lanes to the left or to the right is less than a first threshold and greater than a second threshold, and the vehicle's turn signal is in an inactive state (i.e., the user has not turned on the turn signal). If so, execute step S403; if not, return to step S401.

[0099] In some examples, when the probability of changing lanes left and right is not greater than the second threshold, it means that the vehicle currently has no intention to change lanes, and there is no need to remind the user to activate the turn signal. At this time, it is only necessary to obtain the vehicle's straight-moving probability, left lane change probability, and right lane change probability in real time or periodically, and make continuous judgments.

[0100] In some examples, the first threshold may be any value between 80% and 100%, such as 85% or 90%.

[0101] In some examples, the second threshold may be any value that is less than the first threshold and greater than or equal to 50%, for example, any value between 50% and 80%, such as 60% or 70%.

[0102] In one implementation, when it is detected that the probability of changing lanes left or right is less than a first threshold and greater than a second threshold within a first time period, and the vehicle's turn signal is in an inactive state, step S403 is further executed. The first time period can, for example, be any time period from 1 to 5 seconds starting from the moment when it is detected that the probability of changing lanes left or right is less than the first threshold and greater than the second threshold, and the vehicle's turn signal is in an inactive state.

[0103] S403: Prompt the user to activate the turn signal, and further execute step S404.

[0104] In some examples, the user can be reminded to activate the turn signal by means of a voice reminder. The voice reminder may be, for example: It is detected that you intend to change lanes, please turn on the turn signal after 3 seconds to execute the lane change; for example, it may also be: It is detected that you are about to change lanes, please turn on the turn signal in advance; for example, it may also be other voice prompts, such as music reminders; for example, it may be a user-defined voice, which is not limited in this application.

[0105] In some other examples, the user can be reminded to turn on the turn signal by means of a reminder on the vehicle display screen. For example, a pop-up box may be displayed on the vehicle display screen: It is detected that you are about to change lanes, please turn on the turn signal in advance. The content displayed in the pop-up box can be in any form, and this application does not limit this.

[0106] In some other examples, the user can be reminded to turn on the turn signal through voice reminders and vehicle display screen reminders. For example, a voice prompt can be issued to the user: It is detected that you are about to change lanes, please turn on the turn signal in advance, and a pop-up box can be displayed on the vehicle display screen: It is detected that you are about to change lanes, please turn on the turn signal in advance. In addition, the user can be reminded to activate the turn signal by any other possible means (for example: by vibration, emitting special smells, lighting, etc.), which is not limited in this application.

[0107] S404: Determine whether the left lane change probability or the right lane change probability is greater than a first threshold, and the vehicle's turn signal is inactive (i.e., the user has not activated the turn signal after being reminded). If so, proceed to step S405; if not, continue to step S404.

[0108] It should be understood that: since the probability of changing lanes to the left or the probability of changing lanes to the right is determined based on one or more characteristic signals obtained to characterize the driving state of the vehicle, the degree of change of the one or more characteristic signals gradually increases during the lane changing process. Therefore, during the vehicle's right (left) lane changing process, the probability of changing lanes to the right (left) gradually increases, the probability of changing lanes to the left (right) gradually decreases, and the probability of going straight also gradually decreases. Therefore, the first threshold is greater than the second threshold.

[0109] In one implementation, when it is detected that the probability of changing lanes left or right is greater than the first threshold value within a second time period and the turn signal of the vehicle is in an inactive state, step S405 is further executed. The second time period can be, for example, any time period from 1 to 5 seconds starting from the moment when it is detected that the probability of changing lanes left or right is greater than the first threshold value and the turn signal of the vehicle is in an inactive state.

[0110] S405: Automatically activate the turn signal, and further execute step S406.

[0111] It can be understood that this step is intended to: when the vehicle changes lanes and is about to cross the line, if the user has not turned on the turn signal, in order to ensure driving safety, the vehicle automatically activates the turn signal to serve as an emergency turn prompt, wherein the automatic activation of the turn signal can, for example, only activate the vehicle's turn signal, or only activate the vehicle's one-touch three (five) flash function, or activate the vehicle's one-touch three (five) flash function while activating the vehicle's turn signal.

[0112] S406: Determine whether the straight-driving probability is greater than a third threshold and the turn signal is activated (ie, the user has not turned on the turn signal). If so, execute step S407; if not, continue to execute step S406.

[0113] In some examples, the third threshold may be any value between 50% and 100%.

[0114] In one implementation, when it is detected that the vehicle's straight-moving probability is always greater than a third threshold value during a third time period and the turn signal is in an activated state, step S407 is further executed. The third time period can be, for example, any time period from 1 to 5 seconds starting from the moment when the straight-moving probability is detected to be greater than the third threshold value.

[0115] S407: Automatically turn the turn signal.

[0116] That is to say, after determining that the probability of changing lanes to the left or the probability of changing lanes to the right is less than the first threshold and greater than the second threshold, it indicates that the user has the intention to change lanes and is ready to change lanes (corresponding to time T2: the time of preparing to change lanes), and the user is reminded to activate the turn signal at this time; after the user is reminded to activate the turn signal, the probability of changing lanes to the left or the probability of changing lanes to the right is continued to be detected in real time or periodically to see whether it is greater than the first threshold. When it is detected that the probability of changing lanes to the left or the probability of changing lanes to the right is greater than the first threshold, it indicates that the user is about to cross the line when changing lanes (corresponding to time T3: the time when the dividing line between the current lane and the target lane begins to cross). If the user still does not turn on the turn signal at this time, the vehicle automatically activates the turn signal; after the turn signal is automatically activated, the probability of going straight of the vehicle is continued to be detected in real time or periodically to see whether it is greater than the third threshold. When it is detected that the probability of going straight of the vehicle is greater than the third threshold, it indicates that the user has completed the lane change (corresponding to time T4: the time when the lane change is completed). If the user has not turned on the signal at this time, the vehicle automatically turns on the signal.

[0117] In the embodiment of the present application, whether the vehicle will change lanes is predicted based on multiple effective features, the vehicle's lane change behavior can be accurately predicted, and the user can be reminded to turn on the turn signal before executing the lane change. The user can be guided to execute the lane change after turning on the turn signal for a specified period of time. It can not only guide the user's positive driving behavior, but also send lane change expectations to vehicles in the lane and the adjacent lane in advance, thereby reducing the frequency of safety accidents; the method can also automatically turn on the light when it is detected that the user does not turn on the light after the lane change is completed, thereby avoiding confusion for vehicles in the lane and the adjacent lane, and thus avoiding a decrease in traffic efficiency.

[0118] In order to more clearly understand step S401 in the embodiment shown in FIG4 , the following, by way of example, in combination with FIG5 , introduces a method for obtaining a vehicle's straight-moving probability, left lane-changing probability, and right lane-changing probability provided in an embodiment of the present application.

[0119] As shown in FIG5 , the method 500 includes the following steps:

[0120] S501: When the vehicle is in a driving state, periodically acquire one or more characteristic signals for characterizing the driving state of the vehicle.

[0121] In some examples, the one or more characteristic signals (also understood as basic characteristic signals) include one or more of the following characteristic signals:

[0122] (1) Lane width.

[0123] Lane width refers to the distance between the centerline of the left lane and the centerline of the right lane of the lane where the vehicle is located before changing lanes and crossing the line. The specific explanation will be introduced in the form of schematic diagrams in subsequent embodiments.

[0124] Optionally, the lane width can be measured by a distance measuring sensor.

[0125] (2) The distance between the vehicle’s front-view camera and the lane line.

[0126] In some examples, the distance between the vehicle's front-view camera and the lane marking is any one or more of the following distances: the distance between the vehicle's front-view camera and the left lane marking, the distance between the vehicle's front-view camera and the right lane marking.

[0127] In some examples, the vehicle's forward-looking camera can be located on the vehicle's central axis. The left lane line refers to the left lane line of the lane in which the vehicle is located before changing lanes and crossing the line. Similarly, the right lane line refers to the right lane line of the lane in which the vehicle is located before changing lanes and crossing the line. The specific explanation will be introduced in the form of schematic diagrams in subsequent embodiments.

[0128] In some examples, the distance between the vehicle's front-view camera and the lane line may refer to the distance between the vehicle's front-view camera and the lane line in a direction perpendicular to the vehicle's central axis; it may also refer to the distance between the vehicle's front-view camera and the lane line in a direction perpendicular to the lane line.

[0129] Optionally, the distance between the vehicle's forward-looking camera and the lane line can be measured by the vehicle's forward-looking camera itself.

[0130] (3) The distance between the vehicle wheels and the lane markings.

[0131] In particular, this application does not limit the number of wheels of the vehicle. The vehicle can be, for example, a two-wheeled vehicle, a three-wheeled vehicle, a four-wheeled vehicle, a six-wheeled vehicle, etc. The distance between the vehicle wheels and the lane line can include the distance between any one or more wheels of the vehicle and the lane line.

[0132] Optionally, for example, it can be the distance between the two front wheels of the vehicle and the lane line. The distance between the wheels and the lane line can include any one or more of the following distances: the distance between the left front wheel of the vehicle and the left lane line; the distance between the left front wheel of the vehicle and the right lane line; the distance between the right front wheel of the vehicle and the left lane line; the distance between the right front wheel of the vehicle and the right lane line.

[0133] Optionally, the distance between the vehicle wheels and the lane line can be measured by a distance measuring sensor.

[0134] (4) The distance between the center of mass of the vehicle and the lane line.

[0135] In some examples, the distance between the center of mass of the vehicle and the lane line includes any one or more of the following distances: the distance between the center of mass of the vehicle and the left lane line; the distance between the center of mass of the vehicle and the right lane line. The specific explanation will be introduced in the form of schematic diagrams in subsequent embodiments.

[0136] Optionally, the distance between the center of mass of the vehicle and the lane line can be measured by a ranging sensor.

[0137] (5) The distance between the vehicle's centroid and the lane line.

[0138] In some examples, the distance between the centroid of the vehicle and the lane line includes any one or more of the following distances: the distance between the centroid of the vehicle and the left lane line; the distance between the centroid of the vehicle and the right lane line. The specific explanation will be introduced in the form of schematic diagrams in subsequent embodiments.

[0139] Optionally, the distance between the centroid of the vehicle and the lane line can be measured by a ranging sensor.

[0140] (6) The angle between the central axis of the vehicle's front-view camera and the lane line.

[0141] In some examples, the angle between the central axis of the vehicle's front-view camera and the lane line includes any one or more of the following angles: the angle between the central axis of the vehicle's front-view camera and the left lane line; the angle between the central axis of the vehicle's front-view camera and the right lane line. The specific explanation will be introduced in the form of schematic diagrams in subsequent embodiments.

[0142] Optionally, the angle between the central axis of the vehicle's front-view camera and the lane line can be directly measured by a sensor or measured during a calibration process.

[0143] (7) Driver's head angle.

[0144] In some examples, the driver's head rotation angle includes any one or more of the following angles: the rotation angle of the driver's head in the XOY plane in a spatial rectangular coordinate system established with the driver's facial center as the coordinate origin, the driver's forward viewing direction as the Z-axis direction, and the horizontal direction as the X-axis direction; the rotation angle of the driver's head in the XOZ plane; the rotation angle of the driver's head in the YOZ plane. The specific explanation will be introduced in the form of schematic diagrams in subsequent embodiments.

[0145] (8) The difference in wheel speed between the wheels of a vehicle.

[0146] In some examples, this application does not limit the number of wheels on a vehicle. The vehicle may be, for example, a two-wheeled vehicle, a three-wheeled vehicle, a four-wheeled vehicle, a six-wheeled vehicle, etc. The wheel speed difference between the wheels of the vehicle may include, for example, the wheel speed difference between any two wheels of the vehicle. In one example, the wheel speed difference between the wheels of the vehicle includes any one or more of the following wheel speed differences:

[0147] The wheel speed difference between the left front wheel and the right front wheel of the vehicle; the wheel speed difference between the left rear wheel and the right rear wheel of the vehicle; the wheel speed difference between the left front wheel and the left rear wheel of the vehicle; the wheel speed difference between the right front wheel and the right rear wheel of the vehicle; the wheel speed difference between the left front wheel and the right rear wheel of the vehicle; the wheel speed difference between the left rear wheel and the right front wheel of the vehicle.

[0148] In yet another example, when the number of wheels of the vehicle is 3, the wheel speed differences between the wheels of the vehicle include any one or more of the following wheel speed differences:

[0149] The wheel speed difference between the first wheel and the second wheel of the vehicle; the wheel speed difference between the first wheel and the third wheel of the vehicle; the wheel speed difference between the second wheel and the third wheel of the vehicle.

[0150] In another example, when the vehicle is a tracked vehicle, the tracks of the vehicle can be regarded as the wheels described in the embodiments of the present application. In this case, the wheel speed difference between the wheels of the vehicle can be expressed by the transmission speed difference between the tracks of the vehicle. For example, when the vehicle has two tracks, the wheel speed difference of the wheels of the vehicle refers to the speed difference between the two tracks.

[0151] In some embodiments, the vehicle may also be a combination of tracks and wheels.

[0152] Alternatively, wheel speed differences between the wheels of the vehicle may be a signal for vehicle dynamics estimation.

[0153] In some embodiments, when the vehicle is in a driving state, in response to a first operation of the user, one or more characteristic signals for characterizing the driving state of the vehicle are periodically acquired.

[0154] In some examples, the user's first operation may be an operation in which the user presses a first control. In addition, the first operation may also be an operation in which the user issues a voice command or a gesture command, etc., which is not limited in this application.

[0155] S502: Determine the vehicle's current left lane change probability, right lane change probability, and straight driving probability based on the one or more characteristic signals.

[0156] In some embodiments, based on the one or more characteristic signals described above, the time domain characteristics corresponding to the one or more characteristics (one or more of the following: mean, standard deviation, median, maximum, root mean square value, crest factor, shape factor, impact factor, and margin factor) are further determined based on the one or more characteristic signals. Furthermore, the vehicle's current left lane change probability, right lane change probability, and straight driving probability are determined based on the time domain characteristics corresponding to the one or more characteristic signals. In one example, these characteristics are detailed in Table 1 below.

[0157] Table 1 Characteristic signal table

[0158] That is to say, the above-mentioned determination of the vehicle's current left lane change probability, right lane change probability and straight going probability based on the one or more characteristic signals can also be understood as: the vehicle's current left lane change probability, right lane change probability and straight going probability can be determined based on one or more of the 133 characteristic signals in Table 1 above.

[0159] It should be understood that the characteristic signals listed in Table 1 above are only for illustrative purposes. In the embodiment of the present application, the characteristic signals used to determine the vehicle's current left lane change probability, right lane change probability, and straight driving probability may also include other characteristic signals that are similar or close in nature to the above characteristic signals, and the present application does not limit this.

[0160] The characteristic signals listed in Table 1 are applicable to, for example, four-wheel vehicles. When the number of wheels on a vehicle changes, the corresponding characteristic signals may also change. For example, characteristic signals such as the speed difference between the vehicle's wheels and the distance between the vehicle's wheels and lane markings will also have different numbers of corresponding characteristic signals when the number of wheels changes. Furthermore, characteristic signals may include the distance between any one or more of the vehicle's wheels and the lane markings. The distance between the vehicle's front wheels and the lane markings described in Table 1 is for illustrative purposes only and does not limit the present application.

[0161] In some embodiments, the vehicle's current left lane change probability, right lane change probability and straight driving probability can also be determined in combination with the steering wheel angle, steering wheel angular velocity, hand torque, accelerator pedal opening, vehicle speed, vehicle lateral acceleration, vehicle longitudinal acceleration, yaw angular velocity, accelerator pedal opening rate, etc.

[0162] In one implementation, during the execution of the method of an embodiment of the present application, if the user performs active intervention, the user's active intervention operation shall prevail. For example, the vehicle determines that the current scenario needs to remind the user to turn according to one or more of the above-mentioned characteristics. When the user has completed the turn before the system issues a reminder, the system will not issue a reminder message.

[0163] In another implementation, the vehicle system can comprehensively determine subsequent operations based on the user's intervention operation and the left lane change probability, right lane change probability and straight driving probability determined by the system. For example, when the left lane change probability determined by the system is greater than 90%, and the user's intervention operation is to turn on the right turn signal, the system determines that the user's operation is an error and automatically turns on the left turn signal. For another example: when the vehicle is currently in the turn signal activation state and the vehicle's straight driving probability is not greater than the third threshold, if the user actively performs the turn signal return operation, the user's turn signal return operation is invalid at this time.

[0164] In one implementation, when a vehicle is in motion, the values ​​of the raw signals collected by the vehicle are stored by the on-board computer, thereby obtaining N basic characteristic signals (. Then, based on the rolling window sampling method, N data samples corresponding to the N basic characteristic signals can be further generated based on these N basic characteristic signals. For each data sample, characteristic indicators such as its sample mean and variance can be calculated, thereby obtaining multiple characteristic signals similar to those listed in Table 1 above.

[0165] In this link, for the recognition model, the dimensionality of the input space is actually changed (the input dimension increases (that is, the input feature signal is diversified), which enriches the amount of effective information received by the model and enhances the generalization ability of the model).

[0166] In one implementation, the model's input can be N basic features, and its output can be the result value of "lane change" or "non-lane change", with "1" representing lane change and "0" representing non-lane change. Based on the sampling data of each rolling window (i.e., the N basic features corresponding to the current sampling period), the model will output 1 current corresponding result. Then, for a period of time, the model will have multiple consecutive results (for example, [0,0,1,1,0,…,1]), which constitute 1 result sample. Furthermore, combined with the verified output result distribution, the vehicle's current left lane change probability, right lane change probability, and straight-ahead probability can be determined based on the result.

[0167] In an embodiment of the present application, during vehicle driving, the vehicle's left lane change probability, right lane change probability, and straight-ahead probability can be determined based on any one or more characteristic signals among the lane width, the distance between the vehicle's front-view camera and the lane line, the distance between the vehicle's wheels and the lane line, the distance between the vehicle's center of mass / centroid and the lane line, the angle between the central axis of the vehicle's front-view camera and the lane line, the driver's head angle, and the wheel speed difference. Since these signal characteristics are characteristics that show statistical differences in lane changing and non-lane changing scenarios, the determined left lane change probability, right lane change probability, and straight-ahead probability can be made more accurate, thereby significantly improving the accuracy of lane change prediction.

[0168] To more clearly understand the multiple characteristic signals listed in Table 1 above, the following is a detailed introduction to the time domain characteristics: mean, standard deviation, median, maximum value, root mean square value, crest factor, shape factor, impact factor, and margin factor.

[0169] 1. Mean

[0170] The mean, or arithmetic average, is the sum of all the data in a set divided by the number of data points. It is an indicator of the trend in a data set.

[0171] n values ​​x1,x2,x3,…,x n The formula for calculating the mean is:

[0172] 2. Standard deviation

[0173] The standard deviation is the most commonly used measure of statistical dispersion in probability statistics. It is defined as the square root of the arithmetic mean of the squared deviations of the standard values ​​of each unit in the population from their mean. It reflects the degree of dispersion among individuals within a group.

[0174] n values ​​x1,x2,x3,…,x n The standard deviation of is calculated as:

[0175] 3. Median

[0176] The median, also known as the median, is the number in the middle of a sequentially arranged set of data. It represents a value in a sample, population, or probability distribution, and divides the set into two equal parts. For a finite set of numbers, the median can be found by sorting all the observations from highest to lowest and finding the one in the middle. If there is an even number of observations, the median is usually the average of the two middle values.

[0177] 4. Maximum

[0178] refers to n values ​​x1,x2,x3,…,x n The maximum value in .

[0179] 5. Root mean square value

[0180] The root mean square value, also called the RMS value or effective value, is calculated by squaring, averaging, and then taking the square root.

[0181] n values ​​x1,x2,x3,…,x n The formula for calculating the root mean square value of is:

[0182] 6. Crest Factor

[0183] n values ​​x1,x2,x3,…,x n The crest factor is:

[0184] 7. Form Factor

[0185] n values ​​x1,x2,x3,…,x n The shape factor is:

[0186] 8. Impact Factor

[0187] n values ​​x1,x2,x3,…,x n The impact factor is:

[0188] 9. Margin Factor

[0189] n values ​​x1,x2,x3,…,x n The margin factor is:

[0190] In an example, the distances between the front camera and the left lane line obtained in the current cycle are 1.0m, 1.01m, 0.92m, 0.90m, and 0.85m respectively. Then the time domain feature signals corresponding to the basic feature signal "the distance between the front camera and the left lane line" are:

[0191] Mean:

[0192] Standard Deviation:

[0193] Median: 0.92; Maximum: 1.01;

[0194] RMS value:

[0195] Crest Factor: Form Factor:

[0196] Impact Factor: Margin Factor:

[0197] In some examples, when the vehicle is driving smoothly within the lane line, the lane line distance variance (square of the standard deviation) is small, while during the lane change process, the lane line distance variance is large; during the lane change process, the crest factor amplitude increases, while the increase in the root mean square value may be affected by non-lane change data and the increase is smaller; the impact factor amplitude increases, while the increase in the mean value may be affected by non-lane change data and the increase is smaller; the increasing trend of the shape factor amplitude during the lane change process will be easier to observe; the margin factor amplitude increases, while the increase in the root mean square value may be affected by non-lane change data and the increase is smaller.

[0198] The following describes in detail the one or more characteristic signals described in S501 in conjunction with FIG. 6 to FIG. 8 .

[0199] For example, Figure 6 shows a schematic representation of one or more characteristic signals in a coordinate system based on the vehicle's posture, taking into account the vehicle's yaw angle. The vehicle's yaw angle is the angle between the vehicle's front direction and the lane centerline, i.e., angle α in Figure 6.

[0200] As shown in Figure 6, taking a vehicle changing lanes to the left as an example, the lane to the left of the vehicle is the target lane. The lane marking to the left of the vehicle's current lane is the left lane marking, which is also the target lane marking during the left lane change. The lane marking to the right of the vehicle's current lane is the right lane marking. Point A is the reference position of the vehicle's forward-facing camera, point B is the vehicle's centroid, point C is the vehicle's center of mass, D1, D2, D3, and D4 are the reference positions of the vehicle's left front wheel, left rear wheel, right front wheel, and right rear wheel, respectively. The lane width is EF. The distance between the vehicle's forward-facing camera and the left lane marking is AO1, which is perpendicular to the vehicle's central axis L. The distance between the vehicle's left front wheel and the left lane marking is D1O2, which is perpendicular to the vehicle's central axis L. The distance between the vehicle's centroid and the left lane marking is BO4, which is perpendicular to the vehicle's central axis L. The distance between the vehicle's center of mass and the left lane marking is CO3, which is perpendicular to the vehicle's central axis L. The angle between the central axis of the vehicle's forward-facing camera and the left lane marking is β.

[0201] Exemplarily, FIG7 shows a schematic diagram representing one or more characteristic signals in a coordinate system established based on the lane centerline without considering the yaw angle of the vehicle.

[0202] As shown in Figure 7, taking a vehicle changing lanes left as an example, the lane to the left of the vehicle is the target lane. The lane marking to the left of the vehicle's current lane is the left lane marking, which is also the target lane marking during the left lane change. The lane marking to the right of the vehicle's current lane is the right lane marking. Point A is the reference position of the vehicle's forward-facing camera, point B is the vehicle's centroid, point C is the vehicle's center of mass, D1, D2, D3, and D4 are the reference positions of the vehicle's left front wheel, left rear wheel, right front wheel, and right rear wheel, respectively. The lane width is EF. The distance between the vehicle's forward-facing camera and the left lane marking is AO5, which is perpendicular to the lane centerline. The distance between the vehicle's left front wheel and the left lane marking is D1O6, which is perpendicular to the lane centerline. The distance between the vehicle's centroid and the left lane marking is BO8, which is perpendicular to the lane centerline. The distance between the vehicle's center of mass and the left lane marking is CO7, which is perpendicular to the lane centerline. The angle between the vehicle's forward-facing camera's central axis and the left lane marking is β.

[0203] For example, Figure 8 shows a schematic diagram of a head rotation angle provided by an embodiment of the present application. As shown in Figure 8, a spatial rectangular coordinate system is established with the center position of the driver's facial coronal plane as the origin, the driver's forward viewing direction as the Z-axis direction, and the horizontal direction as the X-axis direction. The driver's head rotation angle is the rotation angle of the driver's head in the XOY plane; the rotation angle of the driver's head in the XOZ plane; and the rotation angle of the driver's head in the YOZ plane. As shown in Figure 8, the driver has a maximum measurable angle α1 in the XOZ plane. Similarly, the driver also has maximum measurable angles in the XOY plane and the YOZ plane (not shown in the figure).

[0204] For example, in conjunction with FIG9 , a human-computer interaction situation corresponding to the lane change prediction method provided in an embodiment of the present application is schematically illustrated.

[0205] As vehicles become increasingly intelligent, the current architectural evolution trend of vehicle steering wheels includes removing the levers (turn signal switch, dimmer switch, main light switch, fog light switch) and assigning the functions corresponding to the levers to the controls on the steering wheel surface, allowing users to achieve human-vehicle interaction through the controls on the steering wheel surface.

[0206] FIG9 (a) shows a schematic diagram of the layout of a currently available steering wheel. As shown in FIG9 (a), steering wheel 910 includes control 911 and control 912. Control 911 is used by the user to turn on the right turn signal, and control 912 is used by the user to turn on the left turn signal. However, a turn signal control method such as steering wheel 910 may cause some vehicle-vehicle interaction problems. For example, when a user needs to turn on the turn signal, because the user needs (and should) always pay attention to the surrounding conditions while driving the vehicle, the user can only rely on subjective perception to turn the steering wheel blindly. It is difficult for the user to achieve accurate blind operation and it is easy to press the wrong button, which causes some unnecessary trouble to the user.

[0207] FIG9(a) shows a schematic diagram of a steering wheel layout according to an embodiment of the present application, which can be applied to the solution of the present application. As shown in FIG9(b), steering wheel 920 includes controls 921, which are used by the user to turn the right turn signal, turn the left turn signal, and return the turn signal. The specific operation to be performed is determined in accordance with the solution of the embodiment of the present application.

[0208] In some examples, when the solution of the embodiments of the present application is applied to a vehicle having a steering wheel 920 as shown in (b) of Figure 9, the user can trigger the operation of periodically determining the vehicle's left lane change probability, right lane change probability and straight driving probability based on one or more characteristic signals by clicking the control 921. The control 921 can, for example, be located at a first position on the steering wheel 920, which refers to a position that can be easily touched when the user is holding the steering wheel.

[0209] For example, the following human-vehicle interaction examples may be included:

[0210] In one example, when a user is driving a vehicle and intends to change lanes to the left, the user can click control 921. At this time, the vehicle periodically determines the vehicle's left lane change probability, right lane change probability, and straight-going probability based on one or more characteristic signals. When it is determined that the left lane change probability is greater than a second threshold or greater than a first threshold, the vehicle can automatically turn on the left turn signal. Similarly, when the user intends to change lanes to the right, the user can click control 921. At this time, the vehicle periodically determines the vehicle's left lane change probability, right lane change probability, and straight-going probability based on one or more characteristic signals. When it is determined that the right lane change probability is greater than a second threshold or greater than the first threshold, the vehicle can automatically turn on the right turn signal.

[0211] In another example, while the user is driving the vehicle, the vehicle periodically determines the vehicle's left lane change probability, right lane change probability and straight-going probability based on one or more characteristic signals. When it is determined that the left lane change probability is less than a first threshold and greater than a second threshold, the vehicle can remind the user to turn on the left turn signal. When the user clicks the control 921, the vehicle turns on the left turn signal. Similarly, while the user is driving the vehicle, the vehicle periodically determines the vehicle's left lane change probability, right lane change probability and straight-going probability based on one or more characteristic signals. When it is determined that the right lane change probability is less than a first threshold and greater than a second threshold, the vehicle can remind the user to turn on the right turn signal. When the user clicks the control 921, the vehicle turns on the right turn signal.

[0212] In another example, while the user is driving the vehicle, the user can click control 921. At this time, the vehicle periodically determines the vehicle's left lane change probability, right lane change probability and straight-going probability based on one or more characteristic signals. When it is determined that the left lane change probability is greater than a first threshold and the vehicle's turn signal is in an inactive state, the vehicle turns on the left turn signal; similarly, when it is determined that the right lane change probability is greater than the first threshold and the vehicle's turn signal is in an inactive state, the vehicle turns on the right turn signal.

[0213] In another example, after the user completes the lane change, the user can click control 921. At this time, the vehicle periodically determines the vehicle's left lane change probability, right lane change probability and straight-going probability based on one or more characteristic signals. When it is determined that the vehicle's straight-going probability is greater than a third threshold and the vehicle's turn signal is in an activated state, the vehicle turns off the turn signal (i.e., automatically returns the light).

[0214] In an embodiment of the present application, combined with the lane change prediction method provided in an embodiment of the present application, the user can realize various steering interactions between the user and the vehicle (including: turning on the left turn signal, turning on the right turn signal, and turning the return signal) through a button on the steering wheel, which can avoid the user pressing the wrong steering control and improve the user's operating convenience.

[0215] For example, FIG10 shows a schematic flow chart of a lane change prediction device 1000 provided in an embodiment of the present application. As shown in FIG10 , the device 1000 includes:

[0216] The acquisition module 1010 is used to acquire the vehicle's straight-moving probability, left lane-changing probability, and right lane-changing probability when the vehicle is in a driving state.

[0217] In some examples, when the vehicle is in a driving state, the acquisition module 1010 periodically acquires one or more characteristic signals used to characterize the driving state of the vehicle; and further acquires the vehicle's straight-moving probability, left lane change probability, and right lane change probability based on the one or more characteristic signals.

[0218] The specific process of the acquisition module acquiring the vehicle's straight-moving probability, left lane-changing probability, and right lane-changing probability is described in detail in the embodiments shown in FIG. 4 and FIG. 5 , and will not be repeated here for the sake of brevity.

[0219] The decision module 1020 is used to remind the user to activate the turn signal when it is determined that the probability of changing lanes to the left or the probability of changing lanes to the right is less than the first threshold and greater than the second threshold, and the turn signal of the vehicle is in an inactive state (i.e., the user has not turned on the turn signal).

[0220] In some examples, the decision module 1020 may send an instruction message to the execution module 1030 of the vehicle computer. After receiving the instruction message, the execution module 1030 may remind the user to activate the turn signal through a voice reminder. The voice reminder may be, for example: It is detected that you intend to change lanes, please turn on the turn signal after 3 seconds; for example, it may also be: It is detected that you are about to change lanes, please turn on the turn signal in advance; for example, it may also be other voice prompts, such as music reminders; for example, it may be a user-defined voice, which is not limited in this application.

[0221] In some other examples, after the execution module 1030 receives the indication information, it can remind the user to turn on the turn signal through the vehicle display screen. For example, a pop-up box can be displayed on the vehicle display screen: It is detected that you are about to change lanes, please turn on the turn signal in advance. The content displayed in the pop-up box can be in any form, and this application does not limit this.

[0222] In some other examples, after receiving the indication information, the execution module 1030 can remind the user to turn on the turn signal through voice reminders and vehicle display screen reminders. For example, a voice prompt can be issued to the user: It is detected that you are about to change lanes, please turn on the turn signal in advance, and a pop-up box can be displayed on the vehicle display screen: It is detected that you are about to change lanes, please turn on the turn signal in advance. In addition, the user can be reminded to activate the turn signal in any other possible way (for example: by vibration, emitting special smells, lights, etc.), and this application does not limit this.

[0223] The decision module 1020 is also used to instruct the execution module 1030 to automatically activate the turn signal when it is determined that the probability of changing lanes to the left or the probability of changing lanes to the right is greater than the first threshold and the vehicle's turn signal is in an inactive state (that is, the user still does not turn on the turn signal after being reminded).

[0224] In some examples, the second threshold may be any value that is smaller than the first threshold and greater than or equal to 50%, for example, any value between 50% and 80%.

[0225] The decision module 1020 is further configured to instruct the execution module 1030 to automatically turn on the turn signal when it is determined that the straight driving probability is greater than a third threshold and the turn signal is in an activated state (ie, the user has not turned on the turn signal).

[0226] In the embodiment of the present application, whether the vehicle will change lanes is predicted based on multiple effective features, the vehicle's lane change behavior can be accurately predicted, and the user can be reminded to turn on the turn signal before executing the lane change. The user can be guided to execute the lane change after turning on the turn signal for a specified period of time. It can not only guide the user's positive driving behavior, but also send lane change expectations to vehicles in the lane and the adjacent lane in advance, thereby reducing the frequency of safety accidents; the method can also automatically turn on the light when it is detected that the user does not turn on the light after the lane change is completed, thereby avoiding confusion for vehicles in the lane and the adjacent lane, and thus avoiding a decrease in traffic efficiency.

[0227] It should be understood that the embodiments of the present application can be applied to vehicles, for example, can be applied to the vehicle's vehicle-mounted system, such as the Hongmeng vehicle-mounted system.

[0228] An embodiment of the present application further provides a computer-readable medium storing a program code. When the computer program code is executed on a computer, the computer executes any one of the methods in FIG. 4 or FIG. 5 .

[0229] The above embodiments can be used alone or in combination with each other to achieve different technical effects.

[0230] In the embodiments provided in the present application above, the methods provided in the embodiments of the present application are introduced from the perspective of an electronic device (e.g., a vehicle-mounted system) as the executing entity. In order to implement the various functions in the methods provided in the embodiments of the present application above, the electronic device may include a hardware structure and / or a software module to implement the above functions in the form of a hardware structure, a software module, or a hardware structure plus a software module. Whether one of the above functions is executed in the form of a hardware structure, a software module, or a hardware structure plus a software module depends on the specific application and design constraints of the technical solution.

[0231] This technical solution is based on model theory, and the model can complete the classification work of the required identified data points through hyperplanes.

[0232] In one example, when the model input is 2D data (in which case the sample points can be visualized as points on a plane), the hyperplane degenerates into a straight line on the plane (represented by a linear equation). If the lane-changing data point group and the non-lane-changing data point group are mixed together, it is impossible to completely separate the two point groups with a straight line, and the classifier performance is low. If the two data point groups are separated from each other by a large distance, then they can be easily distinguished by a straight line (hyperplane), and the classifier performance will be relatively improved.

[0233] In light of current practical problems, model input features must be clearly distinguishable in two different scenarios (corresponding to two modalities): "lane change" and "non-lane change." This distinguishability can be demonstrated by the statistical difference in sample means between the two modalities (e.g., testing for statistical significance based on p-values), the linear relationship between features across modalities (strong linear relationships mean that features can be linearly represented, resulting in high feature information redundancy), or the distance between data points across modalities (e.g., Euclidean distance).

[0234] Taking the characteristic signals listed in Table 1 as an example, the 133 features listed in Table 1 can be divided into five types, namely: (1) lane attributes (i.e., lane width); (2) distance between vehicle and lane line (i.e., the distance between the vehicle's front-view camera and the lane line, the distance between the vehicle's wheels and the lane line, the distance between the vehicle's center of mass and the lane line, and the distance between the vehicle's centroid and the lane line); (3) posture of vehicle and lane line (i.e., the angle between the central axis of the vehicle's front-view camera and the lane line); (4) driver's head angle; and (5) wheel speed difference.

[0235] Combined with actual data, further explanation is as follows:

[0236] (1) Lane attributes: Lane width varies in actual driving environments (according to my country's road design standards, it ranges from 2.3 to 3.75 meters). Therefore, the distance between the vehicle and the lane line needs to be calculated or verified with the help of the real lane width measured in real time.

[0237] (2) Distance between the vehicle and the lane line: In non-lane-changing scenarios, the instantaneous value or average of this distance value should be greater than a certain threshold (for example, greater than 0.5 meters. That is, when the distance value is less than 0.5 meters, the outer contour of the vehicle may have overlapped with the target lane line, and the vehicle may be in the process of changing lanes). In addition, regardless of the curvature of the road, when the driver drives the vehicle stably within the lane, the variation of this distance value is generally small and the sample value range is small, which is very different from the data modality of this type of characteristic signal in the lane-changing scenario;

[0238] (3) Vehicle and lane line posture: Regardless of the curvature of the road, vehicles in a lane usually need to follow (or be equidistantly parallel to) the center line of the lane. Therefore, during the non-lane change process, the angle between the vehicle's front and the lane line is relatively small. On the contrary, during high-speed lane changes, although the vehicle posture changes slightly, high-speed lane changes usually correspond to highway environments (with small road curvature), so the angle changes are significant. At the same time, during low-speed lane changes, there are many scenarios where the vehicle posture changes significantly (such as changing lanes or ramp entrances and exits), so the angle changes are also significant.

[0239] (4) Head movement angle: During the lane change process, the driver will constantly check the rearview mirror for self-protection. That is, the driver's body behavior patterns and attention distribution are different during the lane change and non-lane change process, so the amplitude and frequency of head movement will also be different;

[0240] (5) Wheel speed difference: During straight driving without lane change, the wheel speed difference of the four wheels is usually the same (within the measurement accuracy) or very close, otherwise the yaw moment acting on the vehicle body will cause it to deviate from the straight driving state; during cornering without lane change, due to the differential and Ackerman angle constraints, the wheel speed difference of the four wheels is different. However, at this time, the vehicle is driving steadily in the lane, the lateral acceleration of the vehicle fluctuates less, and the degree of change in the wheel speed difference is low and the extreme difference is small. In addition, the average curvature formed by the lane change trajectory is usually greater than the curvature of the curve. Therefore, the wheel speed difference in the straight driving scenario and the wheel speed difference in the lane change scenario will have different data modes.

[0241] It can be found that the 133 features listed in Table 1 all have differences in data modality. The degree of change and trend of the values ​​of these features in lane-changing and non-lane-changing scenarios are different or opposite, that is, they have good distinguishability between different modalities, which is conducive to improving the classification effect of the linear classifier.

[0242] Furthermore, the calculation formulas based on all the calculation features can be divided into:

[0243] (1) Numerical characteristics of sample distribution

[0244] About the central tendency of a distribution: mean and median;

[0245] Regarding the degree of dispersion of the distribution: standard deviation.

[0246] (2) Waveform characteristics of the sample curve (the same distribution digital characteristics do not mean the same distribution, so the characteristics of the data line shape need to be considered)

[0247] Root mean square value (all signals used are first processed into positive real numbers);

[0248] Fractions with the mean as the denominator: crest factor and shape factor;

[0249] Use the root mean square value or its square value as the denominator of the fraction: impact factor and margin factor.

[0250] As mentioned above, the values ​​of the aforementioned features typically differ between the two modes (lane change and non-lane change), and the magnitude and direction of the changes differ. This means that the "mean" and "standard deviation" of the original signal involved are typically different under lane change and non-lane change conditions. When the data volume is large enough, the sample distribution often exhibits a normal distribution or an approximately normal distribution, and the median will approach the mean, which means that the differences in the modal means may also be reflected in the median. Furthermore, the crest factor and shape factor, which use the mean as the denominator of the fraction, will capture this modal difference.

[0251] On the other hand, when the values ​​of the original signals are all positive real numbers, the differences in the mean may also be reflected in the RMS value. Then, the impact factor and margin factor with the RMS value or its square value as the denominator of the fraction may also represent the modal differences between the lane-changing and non-lane-changing data.

[0252] In summary, combining the model principles used in practice, the model theory's requirements for data modality, the driver's actual behavior, and the data change pattern in the real vehicle environment, it can support the rationality of using the 133 features listed in Table 1 for lane change prediction and improving the accuracy of lane change prediction.

[0253] Similarly, characteristic signals with properties similar to or similar to the characteristic signals listed above can also be effectively used for vehicle lane change prediction.

[0254] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or 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 can 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.

[0255] In this embodiment, the electronic device can be divided into functional modules according to the above-mentioned method example. For example, each functional module can be divided according to each function, or two or more functions can be integrated into a single processing module. The above-mentioned integrated modules can be implemented in the form of hardware. It should be noted that the module division in this embodiment is illustrative and is only a logical functional division. In actual implementation, other division methods may be used.

[0256] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0257] In the several embodiments provided in this 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 merely schematic. For example, the division of the unit is only a logical function division. In actual implementation, there may be other division methods, such as 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.

[0258] 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 units may be selected to achieve the purpose of the solution of this embodiment according to actual needs.

[0259] 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.

[0260] If this 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. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method of each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program code.

[0261] The above are only specific embodiments of the present application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A method for predicting lane change, characterized in that: The method comprises: When the vehicle is in a driving state, periodically determining a left lane change probability, a right lane change probability, and a straight driving probability of the vehicle according to one or more characteristic signals, wherein the one or more characteristic signals are characteristic signals for characterizing the driving state of the vehicle; When the left lane change probability or the right lane change probability is less than a first threshold and greater than a second threshold, and the turn signal of the vehicle is in an inactivated state, a first reminder is issued to the user, wherein the first reminder is used to remind the user to activate the turn signal, wherein the second threshold is less than the first threshold.

2. The method according to claim 1, characterized in that: The method further comprises: When the left lane change probability or the right lane change probability is greater than the first threshold and the turn signal of the vehicle is in an inactivated state, activating the turn signal of the vehicle.

3. The method according to claim 1 or 2, characterized in that: The method further comprises: When the straight-moving probability of the vehicle is greater than a third threshold and the turn signal of the vehicle is in an activated state, the state of the turn signal of the vehicle is switched to an inactivated state.

4. The method according to any one of claims 1 to 3, characterized in that The method further comprises: In response to a user clicking on a first control of the vehicle, determining a left lane change probability, a right lane change probability, and a straight driving probability of the vehicle according to the one or more characteristic signals; When it is determined that the probability of the vehicle changing lanes to the left is greater than the second threshold or greater than the first threshold, activating the left turn signal of the vehicle; or, when it is determined that the probability of the vehicle changing lanes to the right is greater than the second threshold or greater than the first threshold, activating the right turn signal of the vehicle; and When it is determined that the straight-moving probability of the vehicle is greater than a third threshold, the left turn signal or the right turn signal is turned off.

5. The method according to any one of claims 1 to 4, characterized in that The method further comprises: The one or more characteristic signals are acquired periodically.

6. The method according to any one of claims 1 to 5, characterized in that The periodically determining the left lane change probability, the right lane change probability and the straight driving probability of the vehicle according to one or more characteristic signals includes: periodically determining one or more lane change results according to the one or more characteristic signals, the lane change results including left lane change, right lane change or straight driving; A left lane-changing probability, a right lane-changing probability, and a straight-going probability of the vehicle are determined according to the one or more lane-changing results.

7. The method according to any one of claims 1 to 6, characterized in that The one or more characteristic signals include a distance characteristic signal of the vehicle relative to an adjacent lane line, and / or an angle characteristic signal of the vehicle relative to the adjacent lane line.

8. The method according to claim 7, characterized in that The distance characteristic signal includes any one or more of the following characteristic signals: Lane width; The distance between the vehicle's forward-looking camera and an adjacent lane line; The distance between the wheels of the vehicle and the adjacent lane markings; The distance between the center of mass of the vehicle and an adjacent lane line; The distance between the centroid of the vehicle and the adjacent lane line.

9. The method according to claim 7 or 8, characterized in that: The angle characteristic signal includes any one or more of the following characteristic signals: The angle between the central axis of the vehicle's front-view camera and the adjacent lane line; the head angle of the driver of the vehicle; The wheel speed difference between the wheels of the vehicle.

10. The method according to claim 9, characterized in that The head angle of the driver driving the vehicle includes any one or more of the following angles: In a spatial rectangular coordinate system established with the center of the driver's facial projection as the origin O, the driver's forward viewing direction as the Z axis, and the horizontal direction as the X axis, the driver's turning angle on the XOY plane, the driver's turning angle on the XOZ plane, and the driver's turning angle on the YOZ plane.

11. The method according to claim 9 or 10, characterized in that: The wheel speed difference between the wheels of the vehicle includes any one or more of the following wheel speed differences: The wheel speed difference between the left front wheel and the right front wheel of the vehicle; the wheel speed difference between the left rear wheel and the right rear wheel of the vehicle; the wheel speed difference between the left front wheel and the left rear wheel of the vehicle; the wheel speed difference between the right front wheel and the right rear wheel of the vehicle; the wheel speed difference between the left front wheel and the right rear wheel of the vehicle; the wheel speed difference between the left rear wheel and the right front wheel of the vehicle.

12. The method according to any one of claims 1 to 11, characterized in that The determining of the left lane change probability, the right lane change probability and the straight driving probability of the vehicle according to one or more characteristic signals comprises: Determine one or more time domain feature sets according to the one or more characteristic signals, wherein the one or more time domain feature sets correspond to the one or more characteristic signals one by one, and the time domain features include one or more of a mean value, a standard deviation, a median, a maximum value, a root mean square value, a crest factor, a shape factor, an impact factor, and a margin factor; A left lane change probability, a right lane change probability, and a straight driving probability of the vehicle are determined according to the one or more time domain feature sets.

13. The method according to any one of claims 1 to 12, characterized in that The issuing of a first reminder to the user comprises: Sending the first reminder to the user by voice; Displaying the first reminder on a display screen of the vehicle; or The first reminder is sent to the user by vibration.

14. A vehicle-mounted electromechanical system, characterized in that: The vehicle-mounted electromechanical system comprises: one or more processors; one or more memories; and one or more computer programs, wherein the one or more computer programs are stored in the one or more memories, and the one or more computer programs include instructions, which, when executed by the one or more processors, cause the on-board electromechanical system to perform the method as described in any one of claims 1 to 13.

15. A vehicle, characterized in that: The vehicle comprises an on-board electromechanical system and a steering system, wherein the on-board electromechanical system is used to control the steering system to implement the method according to any one of claims 1 to 13.

16. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a computer, the computer is caused to implement the method according to any one of claims 1 to 13.

Images