Lane departure warning method and device, electronic equipment and storage medium

Abstract

The application relates to a lane departure warning method and device, electronic equipment, a storage medium and a computer program product. The method comprises the following steps: acquiring lane line parameters, vehicle driving parameters and a warning function state signal in a vehicle driving process; the warning function state signal is a signal for identifying the running state of a lane departure warning function; in the case that a current vehicle mode is in a warning waiting mode, a judgment result of whether a current driving state meets a lane departure warning condition is acquired, the current vehicle mode being determined according to the warning function state signal; in the case that the judgment result is that the lane departure warning condition is met, a warning starting mode is entered; in the warning starting mode, in the case that the vehicle is in a non-lane changing process, a warning result is determined based on the lane line parameters and the vehicle driving parameters. The method can improve the accuracy of lane departure warning.

Description

Technical Field

[0001] This application relates to the field of automotive technology, and in particular to a lane departure warning method, device, electronic device, storage medium, and computer program product. Background Technology

[0002] With the development of automotive technology, lane departure warning systems can monitor vehicle driving information in real time and promptly remind the driver to concentrate when there is a risk of the vehicle deviating from its lane, ensuring normal and safe driving. This is especially suitable for large commercial vehicles in long-distance driving industries such as express delivery and logistics, improving driving safety.

[0003] However, most existing lane departure warning systems assess the risk of lane departure by comparing the lane line fitting parameters identified by the forward-facing camera with a calibrated critical distance when the real-time detection system is activated. If the distance exceeds the critical distance, a lane departure warning is triggered, which is prone to false alarms and results in low accuracy. Summary of the Invention

[0004] Therefore, it is necessary to provide a lane departure warning method, device, electronic device, computer-readable storage medium, and computer program product that can improve the accuracy of lane departure warning in response to the above-mentioned technical problems.

[0005] Firstly, this application provides a lane departure warning method. The method includes:

[0006] The system acquires lane line parameters, vehicle driving parameters, and warning function status signals during vehicle operation; the warning function status signals are signals used to identify the operating status of the lane departure warning function.

[0007] When the current vehicle mode is in the warning waiting mode, the system obtains the judgment result of whether the current driving status meets the lane departure warning conditions. The current vehicle mode is determined based on the warning function status signal.

[0008] If the judgment result is that the lane departure warning conditions are met, the warning activation mode is entered;

[0009] In the warning activation mode, when the vehicle is not in a lane-changing process, the warning result is determined based on the lane line parameters and the vehicle driving parameters.

[0010] In one embodiment, the determination result of whether the current driving state meets the lane departure warning conditions includes:

[0011] If the lane line parameters meet the lane line quality conditions and the vehicle driving parameters meet the driving parameter conditions, the judgment result is determined to meet the lane departure warning conditions.

[0012] In one embodiment, the lane line parameters include lane line quality parameters, and the vehicle driving parameters include vehicle speed, pedal parameters, and turn signal; the determination process for each of the lane line parameters and the vehicle driving parameters includes:

[0013] When the lane line quality parameter is a preset parameter type, it is determined that the lane line parameter meets the lane line quality condition, and the preset parameter type is a parameter type suitable for lane departure warning judgment;

[0014] If the vehicle speed is within the preset speed range, the pedal parameters have not reached the set pedal parameter threshold, and the turn signal indicates that the turn signal is not activated, then the vehicle driving parameters are determined to meet the vehicle driving parameter conditions.

[0015] In one embodiment, the vehicle driving parameters include steering wheel angle; the process for determining whether the vehicle is in a non-lane-changing process includes:

[0016] The steering wheel angle is compared with a preset standard angle to obtain the angle difference between the steering wheel angle and the preset standard angle;

[0017] If the turning angle difference reaches the turning angle difference threshold, it is determined that the vehicle is in a non-lane-changing process.

[0018] In one embodiment, the lane line parameters further include lane line fitting parameters, which include lateral distance, vehicle orientation angle, road curvature, and road curvature change rate. The vehicle driving parameters also include vehicle yaw rate and vehicle lateral acceleration.

[0019] The determination of the warning result based on the lane line parameters and the vehicle driving parameters includes:

[0020] The lateral distance, vehicle heading angle, road curvature, rate of change of road curvature, vehicle yaw rate, and vehicle lateral acceleration are input into a pre-trained warning decision tree model. The warning decision tree model then assesses each of these parameters to determine whether a vehicle departure warning should be issued. In one embodiment, the training process of the warning decision tree model includes:

[0021] Obtain a sample training set; the sample training set includes sample lateral distance, sample vehicle orientation angle, sample road curvature, sample road curvature change rate, sample vehicle yaw rate, and sample vehicle lateral acceleration.

[0022] The initial decision tree model is trained based on the sample training set to determine the information gain corresponding to at least two decision modes constructed from the sample lateral distance, the sample vehicle orientation angle, the sample road curvature, the sample road curvature change rate, the sample vehicle yaw rate, and the sample vehicle lateral acceleration.

[0023] Based on each information gain, a decision tree model for early warning is determined.

[0024] Secondly, this application also provides a lane departure warning device. The device includes:

[0025] The data acquisition module is used to acquire lane line parameters, vehicle driving parameters, and warning function status signals during vehicle operation; the warning function status signals are signals used to identify the operating status of the lane departure warning function.

[0026] The judgment module is used to obtain a judgment result on whether the current driving state meets the lane departure warning conditions when the current vehicle mode is in the warning waiting mode. The current vehicle mode is determined based on the warning function status signal.

[0027] The control module is used to enter the warning activation mode when the judgment result is that the lane departure warning conditions are met;

[0028] The warning module is used to determine the warning result based on the lane line parameters and the vehicle driving parameters when the vehicle is not in a lane-changing process in the warning activation mode.

[0029] Thirdly, this application also provides an electronic device. The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the lane departure warning method described above.

[0030] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the steps of the lane departure warning method described above.

[0031] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the steps of the lane departure warning method described above.

[0032] The aforementioned lane departure warning method, device, electronic equipment, storage medium, and computer program product acquire lane line parameters, vehicle driving parameters, and warning function status signals during vehicle operation. The warning function status signals identify the operational status of the lane departure warning function. When the current vehicle mode is in warning waiting mode, the system obtains a judgment result on whether the current driving state meets the lane departure warning conditions; the current vehicle mode is determined based on the warning function status signals. If the judgment result indicates that the lane departure warning conditions are met, the system enters a warning activation mode. In the warning activation mode, when the vehicle is not in a lane-changing process, the warning result is determined based on the lane line parameters and vehicle driving parameters. During the lane departure warning process, by acquiring lane line parameters, vehicle driving parameters, and warning function status signals during driving, and after determining entry into the warning activation mode based on the warning function status signals, and then determining that the vehicle is not in a lane-changing process, the warning result is determined based on the lane line parameters and vehicle driving parameters. By determining the warning result only when entering the warning activation mode and being in a non-lane-changing process, a higher performance lane departure warning can be achieved, thereby improving the accuracy of the lane departure warning. Attached Figure Description

[0033] Figure 1 This is an application environment diagram of the lane departure warning method in one embodiment;

[0034] Figure 2 This is a flowchart illustrating a lane departure warning method in one embodiment;

[0035] Figure 3 This is a structural block diagram of a lane departure warning device in one embodiment;

[0036] Figure 4 This is a diagram of the internal structure of an electronic device in one embodiment. Detailed Implementation

[0037] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0038] The lane departure warning method provided in this application embodiment can be applied to, for example, Figure 1 The application environment shown. Among them, Figure 1The lane departure warning system shown may include a camera unit, a data acquisition unit, an alarm unit, and an instrument display unit, as well as electronic devices connected to these units. The camera unit can be a camera mounted at an appropriate height on the front of the vehicle or the windshield, used to detect the left and right lane line data in real time and send it to the electronic devices. The data acquisition unit can be a unit used to collect vehicle driving parameters, including accelerator pedal opening signals, brake pedal opening signals, vehicle speed signals, turn signal signals, steering wheel angle signals, vehicle yaw rate signals, and lateral acceleration signals. The alarm unit can be a buzzer located in the driver's cabin, which can sound an alarm when needed. The instrument display unit can be the vehicle's instrument panel, or a large in-vehicle screen when used in conjunction with advanced driver-assistance systems. The instrument display unit can display the lane departure warning system status, lane line recognition status, and lane departure warning information, showing the warning results.

[0039] The electronic device can be a controller on the vehicle, integrating a signal detection module, a system status judgment module, a lane departure risk judgment module, and a lane departure warning signal transmission module. The signal detection module checks the input signals; if a signal input value is missing or exceeds the normal range, it is considered abnormal. The system status judgment module determines the operating status of the lane departure warning system, which can include five states: off, waiting, on, alarm, and fault. The lane departure risk judgment module assesses the risk of lane departure and provides a warning. The lane departure warning signal transmission module sends system status information, lane detection information, and lane departure alarm information to the instrument display unit and alarm unit, controlling them to perform corresponding actions.

[0040] In one embodiment, the electronic device acquires lane line parameters, vehicle driving parameters, and warning function status signals during vehicle operation; the warning function status signals are signals used to identify the operating status of the lane departure warning function; when the current vehicle mode is in warning waiting mode, the device acquires a judgment result on whether the current driving state meets the lane departure warning conditions, the current vehicle mode being determined based on the warning function status signals; if the judgment result indicates that the lane departure warning conditions are met, the device enters the warning activation mode; in the warning activation mode, when the vehicle is not in a lane-changing process, the warning result is determined based on the lane line parameters and vehicle driving parameters.

[0041] In one embodiment, such as Figure 2As shown, a lane departure warning method is provided, which is applied to... Figure 1 Taking an electronic device as an example, the explanation includes the following steps:

[0042] Step 202: Obtain lane line parameters, vehicle driving parameters, and warning function status signals during vehicle operation; the warning function status signals are signals used to identify the operating status of the lane departure warning function.

[0043] Lane line parameters can be real-time lane line data of the vehicle measured by a camera during vehicle operation. This real-time lane line data can include left and right lane line data. Vehicle driving parameters can be parameters involved in vehicle operation, specifically obtained by acquiring signals such as accelerator pedal opening, brake pedal opening, vehicle speed, turn signal, steering wheel angle, vehicle yaw rate, and lateral acceleration.

[0044] The warning function status signal can be a signal used to identify the operating status of the lane departure warning function. The warning function status signal can be represented by a status value, which can be any identifier with identifying function, such as numbers, letters, or feature codes. For example, numbers can be used as status values; 0 can represent the function being off and 1 can represent the function being on. Alternatively, 1 can represent the function being off and 0 can represent the function being on. In this embodiment, 0 represents the function being off and 1 represents the function being on. The operating status of the lane departure warning function can include five states: off, waiting, on, alarm, and fault. The lane departure warning function can be equipped with a corresponding function switch. When the function switch is activated, if the driver turns on the function switch, the system enters the waiting state; if the driver turns off the function switch, the system returns to the off state.

[0045] Step 204: When the current vehicle mode is in the warning waiting mode, obtain the judgment result of whether the current driving status meets the lane departure warning conditions. The current vehicle mode is determined based on the warning function status signal.

[0046] The current vehicle mode can be any operating mode involved in vehicle operation. This mode can include the vehicle's lane departure warning mode, which refers to the mode used to determine whether to issue a lane departure warning. The lane departure warning mode can include various modes such as warning off mode, warning waiting mode, and warning activation mode. Warning off mode indicates that a lane departure warning judgment has not yet been initiated, meaning no lane departure warning judgment is needed or has not been performed. Warning waiting mode indicates a standby phase, which can be a waiting state before the lane departure warning judgment process. In warning waiting mode, as long as a command is received or triggered, the warning activation mode can be entered. Warning activation mode can be a mode that is in the process of lane departure warning judgment.

[0047] The current driving status can be the vehicle's real-time driving status, which determines whether the vehicle is eligible for lane departure warning. The vehicle's driving status can be determined by one of the following parameters: lane line parameters or vehicle driving parameters, or it can be determined by both. Lane departure warning conditions can be set to determine whether the current driving status meets the criteria for entering the lane departure warning process. These conditions can be set based on actual vehicle configuration parameters and operating conditions.

[0048] Specifically, the electronic device can determine whether the current vehicle mode is in a warning waiting mode based on the warning function status signal. For example, the electronic device can judge based on the status value of the warning function status signal. If the electronic device determines that the status value is 0, it means that the function is off, that is, the current vehicle mode is not in a warning waiting mode, but in a warning off mode. If the electronic device determines that the status value is 1, it means that the function is on, that is, the current vehicle mode is in a warning waiting mode. When the current vehicle mode is in a warning waiting mode, the electronic device can further obtain the judgment result of whether the current driving state meets the lane departure warning conditions, and determine whether to enter the vehicle departure warning judgment process based on the judgment result.

[0049] Step 206: If the judgment result is that the lane departure warning conditions are met, enter the warning activation mode.

[0050] The judgment result meeting the lane departure warning conditions can mean that the current driving state meets the lane departure warning conditions. Specifically, when the lane line recognition quality is good, the vehicle speed is within the set speed range, and the brake pedal opening and accelerator pedal opening have not reached the set opening limits, the warning activation mode can generally be entered.

[0051] Step 208: In the warning activation mode, when the vehicle is not in a lane-changing process, the warning result is determined based on the lane line parameters and vehicle driving parameters.

[0052] The "non-lane-changing process" refers to the process where the vehicle is not changing lanes from one lane to another. The opposite of the "non-lane-changing process" is the "lane-changing process," which refers to the process where the vehicle is changing lanes from one lane to another. When the electronic equipment determines that the vehicle is in a non-lane-changing process, it uses lane line parameters and vehicle driving parameters to determine whether to issue a lane departure warning. By considering the special circumstances of the lane-changing process and taking into account driving intentions, the user-friendliness of the human-machine interface is improved.

[0053] In one embodiment, if the electronic device determines that a lane departure warning is required during a non-lane-changing scenario, it can enter an alarm state and control the alarm unit to issue an alarm. Specifically, the duration of the alarm can be set, such as 2 seconds. If, after 2 seconds of alarming, another 2 seconds have passed and the device still determines that there is a risk of lane departure, the alarm will continue to be issued.

[0054] In the aforementioned lane departure warning method, lane line parameters, vehicle driving parameters, and warning function status signals are acquired during vehicle operation. The warning function status signals identify the operational status of the lane departure warning function. When the current vehicle mode is in warning waiting mode, a judgment result is obtained regarding whether the current driving state meets the lane departure warning conditions; the current vehicle mode is determined based on the warning function status signals. If the judgment result indicates that the lane departure warning conditions are met, the warning activation mode is entered. In the warning activation mode, if the vehicle is not in a lane-changing process, the warning result is determined based on the lane line parameters and vehicle driving parameters. During the lane departure warning process, by acquiring lane line parameters, vehicle driving parameters, and warning function status signals during driving, and after determining entry into the warning activation mode based on the warning function status signals, and then determining if the vehicle is not in a lane-changing process, the warning result is determined based on the lane line parameters and vehicle driving parameters. By determining the warning result only when entering the warning activation mode and being in a non-lane-changing process, a higher performance lane departure warning can be achieved, thereby improving the accuracy of the lane departure warning.

[0055] In one embodiment, obtaining the judgment result of whether the current driving state meets the lane departure warning conditions includes: if the lane line parameters meet the lane line quality conditions and the vehicle driving parameters meet the driving parameter conditions, determining that the judgment result meets the lane departure warning conditions.

[0056] The lane line quality condition can be a set condition for judging the quality of lane line parameters, specifically judging the recognition quality of lane line parameters captured by the camera. The driving parameter condition can be a set condition for judging whether the values ​​of vehicle driving parameters are within the range suitable for lane departure warning judgment. When the electronic device determines that the lane line parameters meet the lane line quality requirements and the vehicle driving parameters meet the driving parameter conditions, it determines that the judgment result is that the lane departure warning condition is met.

[0057] In this embodiment, by setting lane line quality conditions and driving parameter conditions, the electronic device determines that the lane departure warning condition is met only when both lane line quality conditions and driving parameter conditions are met simultaneously, thereby improving the accuracy of lane departure warning.

[0058] In one embodiment, lane line parameters include lane line quality parameters, and vehicle driving parameters include pedal parameters and turn signal; the determination process for each of the lane line parameters and vehicle driving parameters includes:

[0059] When the lane line quality parameters are of the preset parameter type, it is determined that the lane line parameters meet the lane line quality conditions. The preset parameter type is a parameter type suitable for lane departure warning judgment.

[0060] If the pedal parameter does not reach the set pedal parameter threshold and the turn signal indicates that the turn signal is not activated, it is determined that the vehicle driving parameter meets the vehicle driving parameter conditions.

[0061] The lane line quality parameter refers to parameters used to characterize the quality of lane line recognition. This parameter can include left lane line quality parameters and right lane line command parameters. The lane line recognition quality can be set with a value range, determined by the forward-looking perception camera algorithm. The value range is generally an integer between 0 and 2, with higher values ​​indicating better lane line recognition quality. The preset parameter type can be a parameter type suitable for lane departure warning judgment, i.e., a type with better lane line quality. This preset parameter type can correspond to specific lane line quality values, such as 1 or 2.

[0062] Vehicle driving parameters can include pedal parameters and turn signal signals. Pedal parameters can refer to the accelerator pedal opening and brake pedal opening. The accelerator pedal opening value can range from 0 to 100%, the brake pedal opening value can range from 0 to 100%, the vehicle speed value can range from 0 to 120 km / h, and the turn signal value ranges from 0 to 2 (integer values). 0 represents no turn signal, 1 represents the left turn signal being on, and 2 represents the right turn signal being on.

[0063] The preset vehicle speed range can be set according to the speed range used for the function; generally, the preset speed range can be 50-120 km / h. For accelerator pedal opening, an accelerator pedal opening threshold is set. This threshold can be a set limit reached by the accelerator pedal opening. Considering that a driver's large-scale pressing of the accelerator pedal indicates an urgent need for power, and that rapid acceleration on highways is likely for overtaking, a condition is set to ensure that the driver's accelerator pedal opening does not exceed the threshold to avoid affecting normal driving. The accelerator pedal opening threshold can be obtained through vehicle calibration and driver surveys. Similarly, for brake pedal opening, a brake pedal opening threshold is set. This threshold can be a set limit reached by the brake pedal opening. Considering that a driver's large-scale pressing of the brake pedal indicates an urgent need for safety, and that rapid deceleration on highways is likely for obstacle avoidance and collision prevention, a condition is set to ensure that the driver's accelerator pedal opening does not exceed the brake pedal opening threshold to avoid affecting normal driving safety. The brake pedal opening threshold is obtained through vehicle calibration and driver surveys.

[0064] Specifically, the electronic device can judge the quality parameters of the left and right lane lines. If both the left and right lane line quality parameters are above 1 (e.g., the left lane line quality parameter is 1 and the right lane line quality parameter is 2), it indicates that the recognition quality of both lane lines is good, which meets the conditions for lane departure warning judgment. The electronic device can also determine whether the vehicle speed is within a preset speed range (e.g., 60 km / h indicates it is within the preset speed range). Regarding accelerator pedal opening, it determines whether the accelerator pedal opening has not reached the accelerator pedal opening threshold; regarding brake pedal opening, it determines whether the brake pedal opening has not reached the brake pedal opening threshold. Furthermore, based on the turn signal values, it determines whether the left and right turn signals are not activated.

[0065] In this embodiment, the electronic device judges the lane line quality parameters to determine whether they meet the lane line quality conditions. After judging the vehicle speed, pedal parameters and turn signal, it determines whether the vehicle driving parameter conditions are met, thereby improving the reliability of the lane departure warning function.

[0066] In one embodiment, vehicle driving parameters include steering wheel angle; the process for determining whether the vehicle is in a non-lane-changing process includes:

[0067] The steering wheel angle is compared with a preset standard angle to obtain the angle difference between the steering wheel angle and the preset standard angle;

[0068] If the turning angle difference reaches the turning angle difference threshold, it is determined that the vehicle is in a non-lane-changing process.

[0069] The steering wheel angle can be measured in real time at the current vehicle speed and lane parameters, corresponding to the steering wheel angle. The preset standard angle can be based on the standard angle involved in lane changing. The standard angle can be calculated by fitting the trajectory using a fifth-order polynomial. The lane changing time parameter used in the fitting can be obtained through calibration. The fifth-order polynomial can be a polynomial obtained by solving the vehicle's lane changing trajectory model.

[0070] Specifically, the electronic equipment will determine whether the vehicle is in the process of changing lanes. Specifically, it can compare the steering wheel angle of the vehicle at the current speed and lane parameters with the standard steering angle during the lane change operation to determine the angle difference between the two. If the angle difference between the two reaches the angle difference threshold, it is determined that it is not a lane change. If it does not reach the angle difference threshold, it is determined that it is a lane change.

[0071] In this embodiment, the electronic device compares the angle difference between the steering wheel angle and the standard steering angle with a preset angle difference threshold to accurately determine whether the vehicle is in a non-lane-changing process, thereby improving the accuracy of lane departure warning.

[0072] In one embodiment, the lane line parameters also include lane line fitting parameters, which include lateral distance, vehicle orientation angle, road curvature, and road curvature change rate. The vehicle driving parameters also include vehicle yaw rate and vehicle lateral acceleration.

[0073] Based on lane line parameters and vehicle driving parameters, the warning result is determined, including:

[0074] The lateral distance, vehicle orientation angle, road curvature, road curvature change rate, vehicle yaw rate, and vehicle lateral acceleration are input into a pre-trained warning decision tree model. The warning decision tree model then judges the lateral distance, vehicle orientation angle, road curvature, road curvature change rate, vehicle yaw rate, and vehicle lateral acceleration to determine whether to issue a vehicle deviation warning.

[0075] The warning decision tree model can be a pre-trained model. Electronic devices can use this model to assess the risk of left and right deviation from vehicle safety. Specifically, the electronic device can input lateral distance, vehicle heading angle, road curvature, rate of change of road curvature, vehicle yaw rate, and vehicle lateral acceleration into the pre-trained warning decision tree model. The warning decision tree model can have multiple levels, with each level making judgments on a specific type of parameter until all parameters (lateral distance, vehicle heading angle, road curvature, rate of change of road curvature, vehicle yaw rate, and vehicle lateral acceleration) have been assessed. The final warning result can then be determined, which may include either issuing a vehicle deviation warning or not issuing one.

[0076] In one embodiment, to further enhance the safety of the lane departure warning system, a lane departure warning safety limit is set in addition to the warning decision tree model. When the front wheels of the vehicle are detected to be crossing the outer edge of the lane line, even if the warning result determined by the warning decision tree model is not to issue a lane departure warning, the electronic device will forcefully determine to issue a lane departure warning to ensure safety.

[0077] In one embodiment, when the electronic device determines that the vehicle is not in a lane-changing process and the warning result is to issue a vehicle departure warning, a warning signal is output according to a preset warning method. For example, when it is determined that the vehicle is not in a lane-changing process and the decision result is to issue a vehicle departure warning, a 2-second high-frequency warning signal will be issued. Furthermore, if it is determined that the vehicle is changing lanes but the driver has not used the turn signal, a 2-second low-frequency reminder signal will be issued.

[0078] In this embodiment, the electronic device combines signals such as lateral distance, vehicle orientation angle, road curvature, road curvature change rate, vehicle yaw rate, and vehicle lateral acceleration to judge the risk of lane departure warning, which can improve the reliability of the lane departure warning function.

[0079] In one embodiment, the training process of the early warning decision tree model includes:

[0080] Obtain the sample training set; the sample training set includes sample lateral distance, sample vehicle orientation angle, sample road curvature, sample road curvature change rate, sample vehicle yaw rate, and sample vehicle lateral acceleration.

[0081] The initial decision tree model is trained based on the sample training set, and the information gain corresponding to at least two decision modes is determined by the sample lateral distance, sample vehicle orientation angle, sample road curvature, sample road curvature change rate, sample vehicle yaw rate, and sample vehicle lateral acceleration.

[0082] Based on each information gain, a decision tree model for early warning is determined.

[0083] In the training process of the early warning decision tree model, the training sample set of the decision tree can be determined based on the driver's style. That is, in the sample training set, whether to issue an alarm in each sample scenario is decided by the driver. Finally, a sample training set of alarms expected by the driver is obtained, making the alarm timing of the lane departure warning system more in line with the driver's personalized needs. The decision mode can be determined by the sample lateral distance, sample vehicle orientation angle, sample road curvature, sample road curvature change rate, sample vehicle yaw rate, and sample vehicle lateral acceleration. Different layers of the decision tree can be composed of different attributes. Therefore, six attributes can construct multiple decision modes, and the corresponding information gain can be calculated for each decision mode.

[0084] Since the risk assessment logic for deviations to the left or right is consistent, only the deviation to the left will be used as an example for explanation. The leaf nodes of the decision tree can be designated as "alarm" or "no alarm." Considered attributes include the vehicle's front-view camera data, specifically the lateral distance of the vehicle's front center position from the left lane line, the vehicle's orientation angle, the road curvature, the rate of change of road curvature, the vehicle's yaw rate, and the vehicle's lateral acceleration. During decision tree training, information gain can be used to evaluate the purity improvement achieved by using each attribute for partitioning. Purity refers to the proportion of samples belonging to the same category within a decision tree's branch node. Information gain is used to select the partitioning attributes for the decision tree, considering that all attributes are continuous variables, and processing these continuous variables based on a dichotomy method.

[0085] Specifically, the electronic device acquires a sample training set. Each attribute in the training set—namely, sample lateral distance, sample vehicle orientation angle, sample road curvature, sample road curvature change rate, sample vehicle yaw rate, and sample vehicle lateral acceleration—is a continuous variable. The electronic device can discretize each attribute, uniformly selecting 10 values ​​for each attribute within its range. For example, the sample lateral distance attribute, with a range of -5 to 5 meters, would have ten values ​​uniformly selected from -5 to 5 meters. Furthermore, the electronic device can employ an orthogonal experimental design. For all the obtained test groups, drivers evaluate whether each group should issue a lane departure warning, resulting in two categories of samples: "warning" and "no warning." These samples are used to train a decision tree, determining the information gain. The decision model with the highest information gain is selected as the warning decision tree model. The decision tree training process is recursive, and the resulting decision tree can be used for lane departure warning decision-making.

[0086] In this embodiment, training a decision tree makes the lane departure judgment process more aligned with the driver's individual driving style and habits. Compared to methods such as neural networks, this approach requires less controller computing power, development time, and training difficulty. Furthermore, the decision tree model is easier to interpret and understand, with clear logic. This improves the efficiency and accuracy of lane departure warning.

[0087] In one embodiment, the complete steps of a lane departure warning method in a specific implementation are as follows:

[0088] The lane departure warning method proposed in this application can be applied to a lane departure warning system, which mainly includes a forward-facing camera, a controller, an instrument display system, alarm elements, and switches. The forward-facing camera is used to detect left and right lane line data in real time. It can output cubic polynomial fitting parameters for the left and right lane lines and the corresponding recognition quality. The cubic polynomial fitting parameters may include lateral distance, vehicle orientation angle, road curvature, and the rate of change of road curvature. Depending on the vehicle model, the camera is installed at an appropriate height on the front of the vehicle or the windshield.

[0089] The instrument cluster display system is typically the vehicle's dashboard, but can also function as a large infotainment screen when used in conjunction with advanced driver-assistance systems (ADAS). The system displays the lane departure warning system status, lane line recognition status, and lane departure warning information. The lane departure warning system status includes five states: off, waiting, on, alarm, and fault. Different states are indicated by different colored and graphic icons. Lane line recognition status includes three states: lane line detected (good quality), lane line detected (poor quality), and lane line not detected. These states are distinguished by the presence or absence of lane lines and their color. Lane departure warning information includes two states: left lane departure warning and right lane departure warning. When an alarm is triggered, the driver is alerted via both the instrument cluster display and an audible alarm. The audible alarm is triggered by a buzzer located in the driver's cabin. A switch is conveniently located in the driver's cabin, allowing the driver to easily turn the lane departure warning system on and off.

[0090] The controller mainly includes a signal inspection module, a system status judgment module, a lane departure risk judgment module, and a lane departure warning signal transmission module. The controller is used to perform various calculations in the lane departure warning method. It receives the cubic polynomial fitting parameters and corresponding recognition quality from the forward-looking camera and the vehicle driving signal from the vehicle chassis, and transmits the calculated results to the instrument display system and alarm elements, and issues an alarm when necessary.

[0091] The input signals to the controller may include function switch signals, lane line recognition quality and lateral distance on both sides, vehicle orientation angle, road curvature and road curvature change rate, accelerator pedal opening signal, brake pedal opening signal, vehicle speed signal, turn signal, steering wheel angle signal, vehicle yaw rate signal and lateral acceleration signal, etc. The output signals of the controller may include lane departure warning system status, lane line recognition status, left lane departure warning signal and right lane departure warning signal, etc.

[0092] The controller's signal inspection module can check the input signals. If a signal input value is missing or exceeds the normal range, it will be determined as an abnormal input signal. The normal ranges for each parameter are as follows: the function switch signal is a 0 or 1 state value, where 0 represents the function is off and 1 represents the function is on; the lane line recognition quality range is determined by different forward-looking perception camera algorithms, generally an integer value between 0 and 2, with larger values ​​indicating better quality; the absolute value range of lane line fitting parameter C0 is 0 to 5m, the absolute value range of lane line fitting parameter C1 is 0 to π / 2 rad, and the absolute value range of lane line fitting parameter C2 is 0 to 0.2 / m. The absolute value of C3 ranges from 0 to 0.002 / m^2; the accelerator pedal opening signal ranges from 0 to 100%; the brake pedal opening signal ranges from 0 to 100%; the vehicle speed signal ranges from 0 to 120 km / h; the turn signal ranges from 0 to 2 (integer values), where 0 represents no turn signal, 1 represents the left turn signal is on, and 2 represents the right turn signal is on; the absolute value of the steering wheel angle signal ranges from 0 to 1000 degrees; the absolute value of the vehicle yaw rate ranges from 0 to 3 rad / s; and the absolute value of the vehicle lateral acceleration ranges from 0 to 15 m / s^2. The actual range of these parameters depends on the vehicle model and the operating environment.

[0093] The controller's system status judgment module can switch and judge the working status of the lane departure warning system. The status includes five states: off, waiting, on, alarm, and fault. If the driver activates the function switch, it enters the waiting state; if the driver deactivates the function switch, it returns to the off state. When the function switch is activated, it will transition from waiting to on state under the following conditions: the lane line recognition quality on both sides is good (e.g., lane line recognition quality reaches 1 or 2); the vehicle speed meets the operating speed requirements, generally 50–120 km / h; and the driver's accelerator pedal deflection does not exceed the limit (this condition considers that a large deflection of the accelerator pedal indicates a significant increase in power). The urgent need for speed is considered in highway driving. During rapid acceleration, it's highly likely the driver is trying to overtake. To avoid affecting normal driving, a condition is set to ensure the driver's accelerator pedal opening does not exceed a certain limit. This limit was obtained through vehicle calibration and driver surveys. Similarly, the driver's brake pedal opening does not exceed a certain limit. This condition considers that a large braking force indicates a strong need for safety. During rapid deceleration on highways, it's highly likely the driver is trying to avoid obstacles or collisions. To avoid affecting normal driving safety, a condition is set to ensure the driver's accelerator pedal opening does not exceed a certain limit. This limit was obtained through vehicle calibration and driver surveys. Finally, the driver did not use their turn signal.

[0094] The above activation conditions can be flexibly calibrated and adjusted according to the actual vehicle conditions. If the above conditions are not met, the vehicle will remain in a waiting state.

[0095] After entering the active state, if a risk of lane departure to the left or right is detected, an alarm will be triggered for 2 seconds (which can be calibrated). If a lane departure risk is still detected after 2 seconds of alarm, the alarm will continue. If some conditions for entering the active state from the waiting state are not met during the alarm, such as the driver activating the turn signal, the alarm will stop, the system will return to the waiting state, and the current system status will be displayed on the instrument panel via a status icon. The system will re-enter the active state once the conditions are met.

[0096] During the operation of the lane departure warning system, if the signal inspection module determines that any abnormal signal lasts for more than 2 seconds or a hardware failure such as a controller occurs, it will enter a fault state and issue a fault warning through the instrument panel. At this time, the lane departure warning function will be downgraded and will no longer be able to perform the warning function. The instrument fault warning will be automatically canceled after the abnormality disappears, and the system will automatically enter the off state. The buzzer will sound a warning sound, the function switch will be reset, and the driver needs to restart the function switch to turn on the function.

[0097] The lane departure risk assessment module takes into account the special nature of lane-changing scenarios. Some drivers do not turn on their turn signals when changing lanes, which is a relatively dangerous and bad driving habit. The controller can first determine whether the vehicle is in the process of changing lanes. The controller compares the steering wheel angle under the current vehicle speed and lane line perception parameters with the standard steering angle during the lane-changing operation. The standard steering angle is calculated based on the fifth-order polynomial fitting trajectory of the lane-changing process. The lane-changing time parameter used in the fitting is obtained by calibration. If the difference between the two steering angles exceeds a threshold, it is determined that it is not a lane change. If it does not exceed the threshold, it is determined that it is a lane change.

[0098] Secondly, a decision tree is used to assess the risk of left and right lane departure. The training sample set for the decision tree is determined based on the driver's style, making the alarm timing of the lane departure warning system more aligned with the driver's individual needs. Since the logic for judging the risk of left and right lane departure is the same, only left lane departure will be used as an example for explanation. The leaf nodes of the decision tree are "alarm" or "no alarm". The attributes considered include the lateral distance |C0| between the vehicle's front-view camera (i.e., the center position of the front of the vehicle) and the left lane line, the vehicle's heading angle |C1|, the road curvature |C2|, the rate of change of road curvature |C3|, the vehicle's yaw rate, and the vehicle's lateral acceleration.

[0099] In decision tree training, information gain is used to evaluate the "purity improvement" of splitting using each attribute. Purity refers to the proportion of samples belonging to the same class in the branch nodes of the decision tree. Information gain is used to select the splitting attribute for the decision tree. For discrete attributes, the information gain formula is as follows:

[0100]

[0101] In the formula, Gain(D,a) represents information gain, D is the current sample set, a is a discrete attribute, Ent(D) represents information entropy, V is the number of possible values ​​for the discrete attribute, and Dv indicates that the v-th branch node contains the set of samples in D for all attributes a = av. The formula for information entropy is as follows:

[0102]

[0103] In the formula, pk is the proportion of the k-th class of samples in the current sample set D, and σ is the number of sample classes.

[0104] Considering that all attributes in this application are continuous variables, and processing continuous variables based on the bisection method, let the split point be t. The actual information gain formula used is as follows:

[0105]

[0106] Among them, D γt γ represents the two sub-sample sets after the sample set D is partitioned based on the partition point t, and γ is the number of the two sub-sample sets after a similar partition.

[0107] To train the decision tree, sample data must first be collected. Since all five attributes are continuous variables, they need to be discretized. Each attribute is assigned 10 values ​​evenly within its range. Using an orthogonal experimental design, drivers evaluate whether a lane departure warning is issued for each experimental group, resulting in two categories: "warning" and "no warning." The decision tree is then trained using these samples in a recursive manner. The resulting decision tree can be used for lane departure warning decisions. To further enhance the safety of the lane departure warning system, a lane departure warning safety limit is set outside the decision tree. When the system detects that the vehicle's front wheels are crossing the outer edge of the lane line, even if the driver's evaluation is "no warning," the system will force a "warning" to ensure safety. When the system determines that the lane departure warning is not a lane change and the decision result is "warning," a 2-second high-frequency warning signal is issued. If the system determines that the lane departure warning is a lane change but the driver has not activated the turn signal, a 2-second low-frequency warning signal is issued. The lane departure warning signal transmission module sends system status information, lane line perception information, and lane departure warning information to the instrument panel and alarm components, controlling the instrument panel and alarm components to perform corresponding actions.

[0108] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0109] Based on the same inventive concept, this application also provides a lane departure warning device for implementing the lane departure warning method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations of one or more lane departure warning device embodiments provided below can be found in the limitations of the lane departure warning method described above, and will not be repeated here.

[0110] In one embodiment, such as Figure 3As shown, a lane departure warning device 300 is provided, including: a data acquisition module 302, a judgment module 304, a control module 306, and a warning module 308, wherein:

[0111] The data acquisition module 302 is used to acquire lane line parameters, vehicle driving parameters, and warning function status signals during vehicle operation; the warning function status signals are signals used to identify the operating status of the lane departure warning function.

[0112] The judgment module 304 is used to obtain the judgment result of whether the current driving state meets the lane departure warning conditions when the current vehicle mode is in the warning waiting mode. The current vehicle mode is determined according to the warning function status signal.

[0113] The control module 306 is used to enter the warning activation mode when the judgment result is that the lane departure warning conditions are met.

[0114] The warning module 308 is used to determine the warning result based on lane line parameters and vehicle driving parameters when the vehicle is not in a lane-changing process in the warning activation mode.

[0115] In one embodiment, the judgment module 304 is further configured to determine that the judgment result meets the lane departure warning condition when the lane line parameters meet the lane line quality condition and the vehicle driving parameters meet the driving parameter condition.

[0116] In one embodiment, lane line parameters include lane line quality parameters, and vehicle driving parameters include vehicle speed, pedal parameters, and turn signal. The judgment module 304 is further configured to determine that the lane line parameters meet the lane line quality conditions when the lane line quality parameters are of a preset parameter type, wherein the preset parameter type is a parameter type suitable for lane departure warning judgment; and to determine that the vehicle driving parameters meet the vehicle driving parameter conditions when the vehicle speed is within a preset speed range, the pedal parameters have not reached the set pedal parameter threshold, and the turn signal indicates that the turn signal is not activated.

[0117] In one embodiment, the warning module 308 is further configured to compare the steering wheel angle with a preset standard angle to obtain the angle difference between the steering wheel angle and the preset standard angle; and determine that the vehicle is in a non-lane-changing process when the angle difference reaches the angle difference threshold.

[0118] In one embodiment, the lane line parameters further include lane line fitting parameters, which include lateral distance, vehicle orientation angle, road curvature, and road curvature change rate. The vehicle driving parameters also include vehicle yaw rate and vehicle lateral acceleration. The warning module 308 is further used to input the lateral distance, vehicle orientation angle, road curvature, road curvature change rate, vehicle yaw rate, and vehicle lateral acceleration into a pre-trained warning decision tree model. The warning decision tree model judges the lateral distance, vehicle orientation angle, road curvature, road curvature change rate, vehicle yaw rate, and vehicle lateral acceleration respectively to determine whether to issue a vehicle departure warning.

[0119] In one embodiment, the early warning module 308 is further configured to acquire a sample training set; the sample training set includes sample lateral distance, sample vehicle orientation angle, sample road curvature, sample road curvature change rate, sample vehicle yaw rate, and sample vehicle lateral acceleration; the initial decision tree model is trained based on the sample training set to determine the information gain corresponding to at least two decision modes constructed from the sample lateral distance, sample vehicle orientation angle, sample road curvature, sample road curvature change rate, sample vehicle yaw rate, and sample vehicle lateral acceleration; and the early warning decision tree model is determined based on each information gain.

[0120] The modules in the aforementioned lane departure warning device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of the electronic device in hardware form or independent of it, or stored in the memory of the electronic device in software form, so that the processor can call and execute the corresponding operations of each module.

[0121] In one embodiment, an electronic device is provided, which may be a controller on a vehicle, and its internal structure diagram may be as follows: Figure 4 As shown, the controller includes a processor, memory, and input / output interfaces. The memory is connected to the processor, and the processor is connected to the input / output interfaces. The processor provides computational and control capabilities. The controller's memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The processor's input / output interfaces are used for exchanging information between the processor and other controllers. When the computer program is executed by the processor, it implements a lane departure warning method.

[0122] Those skilled in the art will understand that Figure 4The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the electronic device to which the present application is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0123] In one embodiment, an electronic device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the lane departure warning method described above.

[0124] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the lane departure warning method described above.

[0125] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of the lane departure warning method described above.

[0126] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data shall comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0127] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0128] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0129] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A lane departure warning method, characterized in that, The method includes: The system acquires lane line parameters, vehicle driving parameters, and warning function status signals during vehicle operation; the warning function status signals are signals used to identify the operating status of the lane departure warning function. When the current vehicle mode is in the warning waiting mode, the system obtains the judgment result of whether the current driving status meets the lane departure warning conditions. The current vehicle mode is determined based on the warning function status signal. If the judgment result is that the lane departure warning conditions are met, the warning activation mode is entered; In the warning activation mode, when the vehicle is not in a lane-changing process, the warning result is determined based on the lane line parameters and the vehicle driving parameters.

2. The method according to claim 1, characterized in that, The determination result of whether the current driving status meets the lane departure warning conditions includes: If the lane line parameters meet the lane line quality conditions and the vehicle driving parameters meet the driving parameter conditions, the judgment result is determined to meet the lane departure warning conditions.

3. The method according to claim 2, characterized in that, The lane line parameters include lane line quality parameters, and the vehicle driving parameters include vehicle speed, pedal parameters, and turn signal. The determination process for the lane line parameters and the vehicle driving parameters includes: When the lane line quality parameter is a preset parameter type, it is determined that the lane line parameter meets the lane line quality condition, and the preset parameter type is a parameter type suitable for lane departure warning judgment; If the vehicle speed is within the preset speed range, the pedal parameters have not reached the set pedal parameter threshold, and the turn signal indicates that the turn signal is not activated, then the vehicle driving parameters are determined to meet the vehicle driving parameter conditions.

4. The method according to claim 1, characterized in that, The vehicle driving parameters include steering wheel angle; the process for determining whether the vehicle is in a non-lane-changing process includes: The steering wheel angle is compared with a preset standard angle to obtain the angle difference between the steering wheel angle and the preset standard angle; If the turning angle difference reaches the turning angle difference threshold, it is determined that the vehicle is in a non-lane-changing process.

5. The method according to claim 1, characterized in that, The lane line parameters also include lane line fitting parameters, which include lateral distance, vehicle orientation angle, road curvature and road curvature change rate. The vehicle driving parameters also include vehicle yaw rate and vehicle lateral acceleration. The determination of the warning result based on the lane line parameters and the vehicle driving parameters includes: The lateral distance, vehicle orientation angle, road curvature, road curvature change rate, vehicle yaw rate, and vehicle lateral acceleration are input into a pre-trained early warning decision tree model. The early warning decision tree model then judges the lateral distance, vehicle orientation angle, road curvature, road curvature change rate, vehicle yaw rate, and vehicle lateral acceleration to determine whether to issue a vehicle deviation warning.

6. The method according to claim 5, characterized in that, The training process of the early warning decision tree model includes: Obtain a sample training set; the sample training set includes sample lateral distance, sample vehicle orientation angle, sample road curvature, sample road curvature change rate, sample vehicle yaw rate, and sample vehicle lateral acceleration. The initial decision tree model is trained based on the sample training set to determine the information gain corresponding to at least two decision modes constructed from the sample lateral distance, the sample vehicle orientation angle, the sample road curvature, the sample road curvature change rate, the sample vehicle yaw rate, and the sample vehicle lateral acceleration. Based on each information gain, a decision tree model for early warning is determined.

7. A lane departure warning device, characterized in that, The device includes: The data acquisition module is used to acquire lane line parameters, vehicle driving parameters, and warning function status signals during vehicle operation; the warning function status signals are signals used to identify the operating status of the lane departure warning function. The judgment module is used to obtain a judgment result on whether the current driving state meets the lane departure warning conditions when the current vehicle mode is in the warning waiting mode. The current vehicle mode is determined based on the warning function status signal. The control module is used to enter the warning activation mode when the judgment result is that the lane departure warning conditions are met; The warning module is used to determine the warning result based on the lane line parameters and the vehicle driving parameters when the vehicle is not in a lane-changing process in the warning activation mode.

8. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images