Abnormal driving behavior recognition method and device and vehicle

Abstract

The invention discloses an abnormal driving behavior recognition method and device, which can be applied to a new energy automobile or an intelligent automobile, can realize the detection of an abnormal driving behavior without adding a sensor, and also can avoid the limitation of the sensor in the process of detecting the abnormal driving behavior. The method comprises: acquiring a first manipulation parameter, a first motion state parameter, and first environment information (S410), the first manipulation parameter being a manipulation parameter for controlling a vehicle to travel at a first time, the first motion state parameter being used for indicating a motion state of the vehicle at the first time, and the first environment information being used for indicating the first environment information; the first environment information is used for indicating the surrounding environment of the vehicle at the first moment; determining a first expected manipulation parameter based on a driver model according to the first motion state parameter and the first environmental information (S430); and determining whether an abnormal driving behavior exists according to the first manipulation parameter and the first expected manipulation parameter (S450).

Description

Method, device and vehicle for identifying abnormal driving behavior Technical Field

[0001] The embodiments of the present application relate to the field of safety, and more specifically, to a method, device, and vehicle for identifying abnormal driving behavior. Background Art

[0002] In recent years, with the rapid development of society and the large-scale construction of urban roads, the number of vehicles has increased rapidly, and the incidence of traffic accidents has continued to rise. For vehicles controlled by drivers, the driver controls the vehicle based on their own driving experience while perceiving the surrounding environment. Their behavior has a crucial impact on driving safety. However, during driving, drivers are inevitably prone to driving fatigue, distraction, emotional driving, and sudden illness. As a result, abnormal operation of the vehicle, such as the steering wheel, accelerator pedal, and brake pedal, can directly affect driving safety. Therefore, monitoring the driver's driving behavior and identifying abnormal driving behavior is of great significance for improving the driver's driving ability, reducing their driving workload, and ultimately reducing the occurrence of traffic accidents.

[0003] Existing methods for detecting abnormal driving behavior typically rely on cameras capturing images of the user's body or face, extracting features, and then analyzing the user's abnormal driving behavior. This approach cannot detect abnormal driving behavior in poor lighting or when the user is obscured. Other detection methods based on wearable sensors utilize physiological data to analyze and identify abnormal driving behavior, but these methods are limited by the need to collect user vital data. Furthermore, these methods often require the additional cost of sensors. Therefore, independently and accurately detecting abnormal driving behavior has become a pressing issue.

[0004] Summary of the Invention

[0005] The embodiments of the present application provide a method and device for identifying abnormal driving behavior, which can independently and accurately identify a user's abnormal driving behavior without adding additional sensors, thereby reducing the cost of detecting abnormal driving behavior.

[0006] In a first aspect, a method for identifying abnormal driving behavior is provided, comprising: obtaining a first manipulation parameter, a first motion state parameter, and first environmental information, the first manipulation parameter being a manipulation parameter for controlling the travel of a vehicle at a first moment, the first motion state parameter being used to indicate the motion state of the vehicle at the first moment, and the first environmental information being used to indicate the surrounding environment in which the vehicle is located at the first moment; determining a first expected manipulation parameter based on a driver model according to the first motion state parameter and the first environmental information; and determining whether there is abnormal driving behavior according to the first manipulation parameter and the first expected manipulation parameter.

[0007] In this embodiment of the present application, the presence of abnormal driving behavior for a vehicle is determined based on the user's input control parameters and the driver model's predicted control parameters. This allows for detection of abnormal driving behavior without the need for additional sensors. Furthermore, because this detection method is directly based on the user's control parameters for the vehicle, it avoids the limitations of sensors in detecting abnormal driving behavior and reduces the delay in detecting abnormal driving behavior.

[0008] In combination with the first aspect, in certain implementations of the first aspect, the method also includes: obtaining a historical manipulation parameter sequence, the historical manipulation parameter sequence including at least one manipulation parameter for controlling the driving of the vehicle within a first historical time period before the first moment; and determining the driver model based on the historical manipulation parameter sequence.

[0009] In combination with the first aspect, in certain implementations of the first aspect, the method further includes: obtaining a second manipulation parameter, a second motion state parameter and second environmental information, the second manipulation parameter being a manipulation parameter for controlling the travel of the vehicle at the second moment, the second motion state parameter being used to indicate the motion state of the vehicle at the second moment, and the second environmental information being used to indicate the surrounding environment in which the vehicle is located at the second moment; determining a second expected manipulation parameter based on the driver model according to the second motion state parameter and the second environmental information; and determining whether there is abnormal driving behavior according to the first manipulation parameter and the first expected manipulation parameter, including: determining whether there is abnormal driving behavior according to a first manipulation parameter sequence and a first expected manipulation parameter sequence, wherein the first manipulation parameter sequence includes the first manipulation parameter and the second manipulation parameter, and the first expected manipulation parameter sequence includes the first expected manipulation parameter and the second expected manipulation parameter.

[0010] Since the user's manipulation of the vehicle is usually consistent, in the embodiment of the present application, the abnormal driving behavior is identified based on the first manipulation parameter sequence and the first expected manipulation parameter sequence, which helps to improve the accuracy of identifying abnormal driving behavior.

[0011] In combination with the first aspect, in certain implementations of the first aspect, the method also includes: determining a first confidence level based on the first manipulation parameter sequence and the first expected manipulation parameter sequence, the first confidence level being used to characterize the degree of similarity between the first manipulation parameter sequence and the first expected manipulation parameter sequence; and determining whether there is abnormal driving behavior based on the first manipulation parameter sequence and the first expected manipulation parameter sequence, including: when the first confidence level is greater than or equal to a first threshold, determining whether there is abnormal driving behavior based on the first manipulation parameter sequence and the first expected manipulation parameter sequence.

[0012] In an embodiment of the present application, when the first confidence level is greater than or equal to the first threshold, determining whether there is abnormal driving behavior is based on the first manipulation parameter sequence and the first expected manipulation parameter sequence. This can avoid abnormal detection caused by environmental information, motion state parameters, and driver model operation errors, and can improve the accuracy of abnormal driving behavior detection.

[0013] In combination with the first aspect, in certain implementations of the first aspect, the method also includes: when the first confidence is greater than or equal to a second threshold, optimizing the first prediction model based on the first manipulation parameter sequence and the first expected manipulation parameter sequence, and the second threshold is less than the first threshold.

[0014] In an embodiment of the present application, when the first confidence level is greater than or equal to the second threshold, the driver model is optimized according to the first manipulation parameter sequence, so that the manipulation parameters output by the driver model are more in line with the user's driving habits, thereby improving the accuracy of abnormal driving behavior detection.

[0015] In combination with the first aspect, in some implementations of the first aspect, the method further includes: obtaining first information, where the first information is used to indicate the user's manipulation preference; and determining the driver model based on the first information.

[0016] In the embodiment of the present application, since the driver model is determined according to the user's operating preferences, a differentiated configuration of the driver model can be achieved, and the accuracy of the method for detecting abnormal driving behavior can be improved.

[0017] In combination with the first aspect, in certain implementations of the first aspect, the method also includes: determining an evaluation function based on the first information; determining whether abnormal driving behavior exists based on the first manipulation parameter sequence and the first expected manipulation parameter sequence, including: determining a value of the evaluation function based on the first manipulation parameter sequence and the first expected manipulation parameter sequence; when the value of the evaluation function is greater than or equal to a third threshold, determining that abnormal driving behavior exists.

[0018] In the embodiment of the present application, an evaluation function is determined based on the first information, so that evaluation criteria for abnormal driving behaviors can be set according to different user manipulation preferences, thereby reducing the possibility of misjudgment and improving the accuracy of identifying abnormal driving behaviors.

[0019] In combination with the first aspect, in certain implementations of the first aspect, the method further includes: determining the operating condition type of the vehicle based on the first motion state parameter and / or the first state parameter; and determining the evaluation function based on the operating condition type of the vehicle.

[0020] In an embodiment of the present application, by determining the operating condition type of the vehicle and thereby determining the evaluation function, it is possible to set corresponding evaluation criteria for abnormal driving behavior based on various scenarios in the user's actual driving, thereby reducing the possibility of misjudgment and improving the accuracy of identifying abnormal driving behavior.

[0021] In combination with the first aspect, in certain implementations of the first aspect, the method further includes: when it is determined that abnormal driving behavior exists, prompting the user that abnormal driving behavior exists.

[0022] In an embodiment of the present application, when abnormal driving behavior is determined, the user can be prompted to focus his or her attention, thereby reducing the occurrence of traffic accidents.

[0023] In combination with the first aspect, in certain implementations of the first aspect, the method further includes: when it is determined that abnormal driving behavior exists, controlling the vehicle to be in an automatic driving mode.

[0024] In combination with the first aspect, in certain implementations of the first aspect, the transportation tool is a vehicle, and the manipulation parameter includes at least one of a steering wheel angle, an accelerator pedal travel, and a brake pedal travel.

[0025] In the second aspect, a device for identifying abnormal driving behavior is provided, which includes: an acquisition module, which can be used to acquire a first manipulation parameter, a first motion state parameter and first environmental information, the first manipulation parameter being a manipulation parameter for controlling the travel of a vehicle at a first moment, the first motion state parameter being used to indicate the motion state of the vehicle at the first moment, and the first environmental information being used to indicate the surrounding environment in which the vehicle is located at the first moment; a processing module, which can be used to determine a first expected manipulation parameter based on a driver model according to the first motion state parameter and the first environmental information; and determine whether there is abnormal driving behavior according to the first manipulation parameter and the first expected manipulation parameter.

[0026] In combination with the first aspect, in certain implementations of the first aspect, the method also includes: obtaining a historical manipulation parameter sequence, the historical manipulation parameter sequence including at least one manipulation parameter for controlling the driving of the vehicle within a first historical time period before the first moment; and determining the driver model based on the historical manipulation parameter sequence.

[0027] In combination with the second aspect, in certain implementations of the second aspect, the acquisition module can also be used to: acquire a second manipulation parameter, a second motion state parameter and second environmental information, wherein the second manipulation parameter is a manipulation parameter for controlling the travel of the vehicle at the second moment, the second motion state parameter is used to indicate the motion state of the vehicle at the second moment, and the second environmental information is used to indicate the surrounding environment in which the vehicle is located at the second moment; the processing module can also be used to: determine a second expected manipulation parameter based on the driver model according to the second motion state parameter and the second environmental information; the processing module can specifically be used to: determine whether there is abnormal driving behavior based on the first manipulation parameter sequence and the first expected manipulation parameter sequence, wherein the first manipulation parameter sequence includes the first manipulation parameter and the second manipulation parameter, and the first expected manipulation parameter sequence includes the first expected manipulation parameter and the second expected manipulation parameter.

[0028] In combination with the second aspect, in certain implementations of the second aspect, the processing module can also be used to: determine a first confidence level based on the first manipulation parameter sequence and the first expected manipulation parameter sequence, and the first confidence level is used to characterize the degree of similarity between the first manipulation parameter sequence and the first expected manipulation parameter sequence; the processing module is specifically used to: determine whether there is abnormal driving behavior based on the first manipulation parameter sequence and the first expected manipulation parameter sequence when the first confidence level is greater than or equal to a first threshold.

[0029] In combination with the second aspect, in certain implementations of the second aspect, the processing module can also be used to: optimize the first prediction model according to the first manipulation parameter sequence when the first confidence is greater than or equal to a second threshold, and the second threshold is less than the first threshold.

[0030] In combination with the second aspect, in certain implementations of the second aspect, the acquisition module can also be used to obtain first information, where the first information is used to indicate the user's manipulation preference; the processing module can also be used to determine the driver model based on the first information.

[0031] In combination with the second aspect, in certain implementations of the second aspect, the processing module can also be used to: determine an evaluation function based on the first information; the processing module can specifically be used to: determine the value of the evaluation function based on the first manipulation parameter sequence and the first expected manipulation parameter sequence; when the value of the evaluation function is greater than or equal to a third threshold, determine that abnormal driving behavior exists.

[0032] In combination with the second aspect, in certain implementations of the second aspect, the processing module can also be used to: determine the operating condition type of the vehicle based on the first motion state parameter and / or the first state parameter; and determine the evaluation function based on the operating condition type of the vehicle.

[0033] In combination with the second aspect, in certain implementations of the second aspect, the processing module may further be configured to: when determining that abnormal driving behavior exists, prompt the user that abnormal driving behavior exists.

[0034] In combination with the second aspect, in certain implementations of the second aspect, the processing module may also be used to: when it is determined that abnormal driving behavior exists, control the vehicle to be in an automatic driving mode.

[0035] In combination with the second aspect, in certain implementations of the second aspect, the transportation tool is a vehicle, and the manipulation parameter includes at least one of a steering wheel angle, an accelerator pedal travel, and a brake pedal travel.

[0036] In a third aspect, a device is provided, comprising a processor and a memory, wherein the memory is used to store program instructions, and the processor is used to call the program instructions so that the device executes any possible method in the first aspect.

[0037] In a fourth aspect, a vehicle is provided, comprising the device of any one of the second aspect or the third aspect.

[0038] Exemplarily, the vehicle may be a smart car or a new energy vehicle, etc.

[0039] In a fifth aspect, a computer program product is provided, comprising: a computer program code, which, when executed on a computer, enables the computer to execute the method in the first aspect.

[0040] In a sixth aspect, a computer-readable medium is provided, wherein the computer-readable medium stores a program code, and when the computer program code is run on a computer, the computer executes the method in the first aspect.

[0041] In a seventh aspect, a chip is provided, which includes a processor and a data interface. The processor reads instructions stored in a memory through the data interface, so that the computer executes the method in the first aspect above. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] FIG1 is a functional block diagram of a vehicle provided in an embodiment of the present application.

[0043] FIG2 is a schematic diagram of a driving scenario provided in an embodiment of the present application.

[0044] FIG3 is a schematic flowchart of a method for training a driver model provided in an embodiment of the present application.

[0045] FIG4 is a schematic flowchart of a method for identifying abnormal driving behavior provided in an embodiment of the present application.

[0046] FIG5 is a schematic flowchart of another method for identifying abnormal driving behavior provided in an embodiment of the present application.

[0047] FIG6 is a schematic structural diagram of a device for identifying abnormal driving behavior provided in an embodiment of the present application.

[0048] FIG7 is a schematic diagram of a system flow for identifying abnormal driving behavior provided in an embodiment of the present application.

[0049] FIG8 is a flow chart of optimizing a driver model provided in an embodiment of the present application.

[0050] FIG9 is a schematic diagram of another process for identifying abnormal driving behavior provided in an embodiment of the present application.

[0051] FIG10 is a structural diagram illustrating an apparatus according to an embodiment of the present application. DETAILED DESCRIPTION

[0052] The technical solutions in the embodiments of the present application will be described below with reference to the accompanying drawings.

[0053] The technical solutions in the embodiments of the present application will be described below with reference to the accompanying drawings.

[0054] In this application, a vehicle may include one or more different types of 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, a vehicle may include a car, bicycle, motorcycle, train, subway, airplane, ship, aircraft, robot, or other types of vehicles or movable objects. The following briefly describes the functions that a vehicle may have, taking a vehicle as an example.

[0055] For example, FIG1 is a functional block diagram of a vehicle 100 provided in an embodiment of the present application. The vehicle 100 may include a perception system 120, a display device 130, and a computing platform 150. 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 a global positioning system (GPS), a Beidou system or other positioning system, an inertial measurement unit (IMU), a lidar, a millimeter-wave radar, an ultrasonic radar, and a camera.

[0056] 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 a field programmable gate array (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, which can be used to store instructions. Some or all of the processors 151 to 15n can call the instructions in the memory and execute the instructions to implement the corresponding functions. The memory can also be used to store data, such as road maps, route information, the vehicle's position, direction, speed and other such vehicle data, as well as other information. This information can be used by the vehicle 100 and the computing platform 150 during operation of the vehicle 100 in autonomous, semi-autonomous and / or manual modes.

[0057] Vehicle 100 may include an advanced driving assistant system (ADAS). ADAS utilizes a variety of sensors on the vehicle (including but not limited to: lidar, millimeter-wave radar, camera, ultrasonic sensor, global positioning system, inertial measurement unit) to obtain information from the surroundings of the vehicle, and analyzes and processes the obtained information to implement functions such as obstacle perception, target recognition, vehicle positioning, path planning, driver monitoring / reminders, etc., thereby improving the safety, automation and comfort of vehicle driving.

[0058] It should be understood that the structure and function of the vehicle in FIG1 should not be construed as limiting the embodiments of the present application.

[0059] Optionally, other vehicles may have similar functions to the vehicle, such as also including a perception system, including a computing platform, etc. For the sake of brevity, no further description is given here. It should be understood that the embodiments of the present application do not limit this.

[0060] For example, FIG2 is a schematic diagram of a driving scenario provided by an embodiment of the present application. The driver can manipulate the vehicle by perceiving environmental information. For example, the vehicle may include a body system, a powertrain system, a chassis system, etc., thereby controlling the acceleration, braking, steering, etc. of the vehicle. For another example, the vehicle may be the vehicle 100 shown in FIG1 . The vehicle can obtain information about its surrounding environment through its perception system. For example, it can obtain information about traffic facilities such as roads, road fences, and intersections in the environment; it can obtain information about road conditions such as lane lines, potholes, and road slopes used to divide lanes; it can obtain information about traffic participants such as the location, size, and speed of surrounding pedestrians and vehicles; it can obtain information about weather conditions such as cloudy, sunny, rainy, and snowy days, etc. For the sake of brevity, it will not be described in detail here.

[0061] For example, FIG3 is a schematic flowchart of a method for training a driver model provided by an embodiment of the present application. The method 200 may include:

[0062] S210 : Acquire a first manipulation parameter, where the first manipulation parameter is a manipulation parameter of the vehicle in a manual driving mode.

[0063] Exemplarily, the first manipulation parameter may be a manipulation parameter for controlling the travel of a vehicle, that is, the first manipulation parameter may be a manipulation parameter output by a user to control the travel of the vehicle. For example, taking the vehicle as an example, the first manipulation parameter may be a manipulation parameter for controlling the vehicle by a user when the vehicle is in manual driving mode, such as the steering wheel, accelerator pedal, brake pedal, and other controllable tools, and the corresponding steering wheel angle, accelerator pedal stroke, brake pedal stroke, and other information output. For traditional fuel-powered vehicles, the first manipulation parameter may also include, for example, clutch pedal stroke. It should be understood that the above method regarding the first manipulation parameter is merely an example and is not limited in this embodiment of the present application.

[0064] S220: Acquire motion state information of a vehicle. The motion state information can be used to represent the motion state of the vehicle.

[0065] Exemplarily, the motion state information can be used to characterize the motion state of the vehicle. For example, the motion state information may include the speed, acceleration, angular velocity, angular acceleration, etc. of the vehicle. Exemplarily, the motion state information of the vehicle can be determined based on the data collected by the sensors of the perception system. For example, taking the vehicle as a vehicle, the perception system of the vehicle may include an IMU, and based on the collection results of the IMU, the acceleration and / or angular acceleration of the vehicle can be determined; for another example, the vehicle may include a speed sensor, and based on the data collected by the speed sensor, the speed of the vehicle can be determined; for another example, the vehicle may also include a wheel speed sensor, and based on the data collected by the wheel speed sensor, the speed of the vehicle can be determined. It should be understood that the above method of obtaining the motion state information of the vehicle is only an example for the sake of explanation, and the embodiments of the present application do not limit this.

[0066] S230: Acquire environmental information of the environment in which the vehicle is located.

[0067] Exemplarily, the environmental information can be used to describe the environmental information of the vehicle. For example, taking the vehicle as an example, the information of other vehicles, pedestrians and other traffic elements around the vehicle can be determined based on the data collected by the perception system of the vehicle. Based on the information of the traffic elements, the obstacles that may exist in the driving process of the vehicle can be determined. For example, the area where other vehicles are located, the heading angle and distance of the surrounding vehicles relative to the vehicle can be obtained, so that the driving path of the vehicle can be planned; the information of the road where the vehicle is located can be determined based on the vehicle's sensors and / or vehicle network data. The road information can be information such as the slope, inclination, and width of the road. It should be understood that the environmental constraints to which the vehicle is subject during driving can be determined based on the environmental information. It should be understood that the above description of environmental information is only an example for the sake of explanation, and the embodiments of the present application are not limited to this.

[0068] It should be understood that some or all of steps S210 to S230 may be performed simultaneously, or step S210 may be performed first, or S220 may be performed first, or S230 may be performed first. This is not limited in the embodiments of the present application.

[0069] S240: Perform model training based on the first manipulation parameter, motion state information, and environmental information to obtain a driver model. The driver model can be used to predict the manipulation parameters for controlling the vehicle.

[0070] Exemplarily, the first manipulation parameter corresponds to the motion state parameter and the environmental information. For example, the first manipulation parameter may be a manipulation parameter for controlling the travel of the vehicle at any moment, the first motion state information may indicate the motion state of the vehicle at that moment, and the environmental information may indicate the surrounding environment in which the vehicle is located at that moment. That is, at that moment, the user's control of the vehicle based on the motion state and surrounding environment of the vehicle may be reflected as the first manipulation parameter. Therefore, model training may be performed based on the first manipulation parameter, the corresponding motion state parameter, and the environmental information, thereby obtaining a driver model for predicting the manipulation parameters for controlling the vehicle. In other words, based on the trained driver model, manipulation parameters for simulating autonomous driving may be output based on the vehicle's perception of the real-time environment and its own motion state.

[0071] For example, in practice, since the sampling frequencies of various sensors may differ, the acquired motion state parameters and environmental information may not, strictly speaking, indicate the motion state and surrounding environment of the vehicle at the moment when the first manipulation parameter exists. When the time difference between the moment corresponding to the motion state parameter, the moment corresponding to the environmental information, and the moment corresponding to the first manipulation parameter is within a preset threshold, the moment corresponding to the motion state parameter, the moment corresponding to the environmental information, and the moment corresponding to the first manipulation parameter can be considered to be the same moment. For example, the purpose is to obtain the first manipulation parameter, motion state parameter and environmental information at the 10th second (s). Due to the influence of factors such as sampling frequency in actual work, the first manipulation parameter obtained may be the manipulation parameter for controlling the travel of the vehicle at 10.05s, the motion state parameter obtained may be the motion state of the vehicle at 9.91s, and the corresponding environmental information obtained may be the surrounding environment of the vehicle at 10.03s. Since the time difference between the corresponding moments is less than or equal to a preset time difference threshold (e.g., 0.1s), it can be considered that the obtained motion state parameter and environmental information indicate the motion state and surrounding environment of the vehicle at the moment corresponding to the first manipulation parameter. That is, the moments corresponding to the first manipulation parameter, motion state parameter and surrounding environment are the same moment (e.g., the 10th second). That is, it can be considered that the obtained motion state parameter indicates the motion state of the vehicle at the 10th second, and the environmental information indicates the surrounding environment of the vehicle at the 10th second. It can be considered that the first manipulation parameter is the manipulation parameter for controlling the travel of the vehicle at the 10th second.

[0072] For example, the driver model can be obtained by performing model training based on at least one set of first manipulation parameters and at least one corresponding set of motion state information and environmental information. For example, determining the driver model based on multiple sets of first manipulation parameters and corresponding motion state information and environmental information can be based on an optimization algorithm, such as a tabu algorithm, genetic algorithm, particle swarm optimization algorithm, or ant colony algorithm, thereby obtaining a driver model that matches the user's actual manipulation behavior. For the sake of brevity, this is not further described here.

[0073] For example, users with different driving preferences may have different manipulations of the vehicle in the same driving scenario. For example, taking the vehicle as an example, a user with an aggressive driving preference and a user with a conservative driving preference, or the users corresponding to the first information "aggressive" and "conservative", may have significant differences in their manipulation of the steering wheel, accelerator pedal, brake pedal, gear, etc. when facing the same driving scenario, such as overtaking, changing lanes, etc., so that in the same scenario, the vehicles driven by the two may have different motion states (such as different speeds, accelerations, etc.), which can be reflected in the fact that the vehicles driven by the two may have different driving trajectories, etc.

[0074] For example, users with different driving preferences may have significant differences in their manipulation of vehicles in the same driving scenario. That is, under the same motion state parameters and environmental information, vehicles controlled by users with different driving preferences may have significantly different first manipulation parameters for controlling the travel of each vehicle. Therefore, different driver models can be obtained through model training for users with different driving preferences, and the driving behavior of users with different driving preferences can be predicted based on these different driver models to improve prediction accuracy. For example, taking the vehicle as an example, for users with conservative driving preferences, a preview driver model can be used, while for users with aggressive driving preferences, a driver model obtained through neural network training can be used. In other words, different types of driver models can be used for users with different driving styles. For another example, for users with conservative driving preferences, a driver model obtained through neural network training can be used, while for users with aggressive driving preferences, a driver model obtained through neural network training can also be used. The two driver models can have different feature parameters. In other words, for users with different driving styles, the same type of driver model with different feature parameters can be used.

[0075] For example, during the model training process and after determining the driving eye model, the driver model may be optimized and adjusted based on the first manipulation parameter, motion state information, and environmental information. The optimization and adjustment of the driver model may include multiple implementations. In some implementations, the process may include multiple iterations. One iteration may include the following steps:

[0076] S310 , inputting motion state information and environmental information into the driver model, processing the motion state information and environmental information through the driver model, and outputting a prediction result, that is, outputting the predicted operating parameters of the vehicle.

[0077] S320. Based on the predicted manipulation parameter and the first manipulation parameter, a first loss value can be calculated through a loss function. The first loss value can represent the deviation between the predicted result and the first manipulation parameter. The greater the deviation between the predicted result and the first manipulation parameter, the greater the first loss value.

[0078] S330: Adjust the driver model according to the first loss value.

[0079] The above shows an iterative process in optimizing the driver model. After one iteration, it can be detected whether the optimization termination condition has been met. When the termination condition is not met, the next iterative process is carried out. When the optimization termination condition is met, the driver model used in this iterative process can be output as the driver model finally used to predict the driver's operating parameters.

[0080] The optimization termination condition can be when the number of iterations reaches a target number or when the loss function meets a preset condition, or when the performance does not improve within a certain number of iterations. The target number can be a pre-set number of iterations used to determine when to end the driver model optimization process to avoid wasting training resources. The termination condition can be when the loss function value remains unchanged or does not decrease over a period of time during the iteration process, such as when the absolute value of the difference between two or more consecutive loss functions is less than the target value, indicating that the adjustment of the driver model has achieved the desired effect.

[0081] It should be understood that the above method of optimizing the driver model is only an example and is not limited to this embodiment of the present application.

[0082] For example, the driver model can be determined by a cloud server, and the cloud server can also optimize the driver model, and the vehicle can obtain the driver model determined / optimized by the cloud server; or after the cloud server determines the driver model, the vehicle obtains the driver model, and can subsequently optimize the driver model; or the vehicle determines the driver model and optimizes the driver model. The embodiments of the present application are not limited to this.

[0083] In the embodiment of the present application, by determining the driver model, the user's operation parameters of the vehicle can be predicted based on the driver model, so that abnormal driving behavior can be detected based on the predicted operation parameters.

[0084] For example, FIG4 is a schematic flow chart of a method for identifying abnormal driving behavior provided in an embodiment of the present application. The method 400 includes the following steps:

[0085] S410: Acquire a first manipulation parameter, a first motion state parameter, and first environmental information.

[0086] Exemplarily, the first moment can be any moment, and the control parameters for controlling the travel of the vehicle at that moment, namely the first control parameters, can be obtained. That is to say, the first control parameters can characterize the user's control of the vehicle at the first moment, the first motion state parameter can be used to indicate the motion state of the vehicle at the first moment, and the first environmental information can be used to indicate the surrounding environment of the vehicle at the first moment.

[0087] For ease of understanding and explanation, the manipulation parameters for controlling the travel of a vehicle, or the manipulation parameters representing the user's control over the vehicle, may be referred to as active manipulation parameters.

[0088] For example, because a user can continuously drive the vehicle for a certain period of time, it is possible to continuously detect abnormal driving behavior. For example, active control parameters within a preset time period, along with corresponding motion state information and environmental information within that time period, can be obtained to determine whether abnormal driving behavior occurs within that time period. For another example, to conserve resources, multiple sets of active control parameters and corresponding motion state information and environmental information within the preset time period can be obtained based on a certain sampling frequency. These multiple sets of control parameters can be sequenced in chronological order.

[0089] For ease of understanding and explanation, the embodiments of the present application distinguish between active manipulation parameters at multiple moments. For example, the active manipulation parameter at the first moment can be defined as the first manipulation parameter. Accordingly, the motion state parameter used to indicate the motion state of the vehicle at that moment can be defined as the first motion state parameter. The environmental information used to indicate the surrounding environment of the vehicle at that moment can be defined as the first environmental information. The active manipulation parameter at the second moment can be defined as the second manipulation parameter. Accordingly, the motion state parameter and environmental information at that moment can be defined as the second motion state parameter and the second environmental information. The manipulation parameters used to control the vehicle's travel within the first historical time length before the first moment can be referred to as historical manipulation parameters. The sequence composed of at least one historical manipulation parameter within the first historical time length can be referred to as a historical manipulation parameter sequence. The historical manipulation parameter can be the manipulation parameter used to control the vehicle's travel during this travel, or it can be the manipulation parameter used to control the vehicle's travel during the user's last driving. The first historical time length can be any length, such as 5 minutes, 10 minutes, etc. For the sake of brevity, it will not be described in detail here.

[0090] Exemplarily, active manipulation parameters at multiple moments, as well as motion state parameters and environmental information corresponding to the multiple moments can be obtained. Exemplarily, a first manipulation parameter sequence, a motion state parameter sequence, and an environmental information sequence can be obtained. The first manipulation parameter sequence can include active manipulation parameters at multiple moments, such as a first manipulation parameter and a second manipulation parameter, or active manipulation parameters at other moments. The motion state parameter sequence can include motion state parameters corresponding to the multiple moments, and the environmental information sequence can include environmental information corresponding to the multiple moments. For example, multiple active manipulation parameters within the time window can be obtained based on a time window. The multiple active manipulation parameters can constitute a first manipulation parameter sequence in a certain order (such as a time order, etc.). Accordingly, the motion state parameter sequence and the environmental information sequence within the time window can be determined. The duration of the time window can be a fixed duration, or a duration determined according to the driving scenario, etc., and the embodiments of the present application do not limit this.

[0091] It should be understood that the above description of obtaining the first manipulation parameter, the first motion state parameter and the first environmental information is only an example, and the embodiments of the present application do not limit this.

[0092] In some possible implementations, first information may be obtained, and the first information may be used to indicate a user manipulation preference.

[0093] Exemplarily, the first information can be determined based on a user identifier. For example, the user identifier can be obtained, for example, by obtaining the user's voiceprint information, facial features, or account information used by the user. Based on the user identifier, the user's manipulation preferences associated with the user identifier can be determined, for example, by querying a server or a local database, thereby determining the first information based on the user identifier. It should be understood that methods for obtaining user identifiers can refer to existing technologies and are not further described here for the sake of brevity.

[0094] Exemplarily, the first information can be determined based on historical manipulation data. For example, taking a vehicle as an example, the user's driving style can be determined based on the user's manipulation of the vehicle over a period of time. For example, the expected driving paths, manipulation data, etc. of users with different driving styles in different scenarios can be preset. The user's manipulation data of the vehicle over a certain period of time (such as 5 minutes) can be matched with the preset expected manipulation data of different driving styles to determine the first information. For another example, the same user's driving preferences in different scenarios may also vary. After the working conditions of the vehicle change, the first information can be determined based on the user's manipulation of the vehicle over a certain period of time. When the first information changes, the driver model can be determined accordingly, thereby ensuring the accuracy of the abnormal behavior detection method.

[0095] Exemplarily, the first information can be determined based on user information. For example, taking a vehicle as a means of transportation, when a user first drives the vehicle, they can set their driving preferences through a mobile phone, tablet computer, or other terminal, or a cockpit screen such as the vehicle's central control screen. For another example, while driving the vehicle, the user can input user information such as gender, age, and years of driving experience through a mobile phone, tablet computer, or other terminal, or a cockpit screen such as the vehicle's central control screen. Based on this user information, matching first information can be determined based on a user characteristic table stored locally or on a server. This user characteristic table can be used to determine a mapping relationship between the user information and the first information. For example, the user characteristic table can be determined based on an analysis of the operating preferences of different users of different ages, genders, years of driving experience, and educational levels. For the sake of brevity, this is not further described here.

[0096] It should be understood that the above method of obtaining the first information is only an example for the convenience of explanation, and the embodiments of the present application do not limit this.

[0097] It should be understood that the first information can be in any data format, such as numbers, letters, or strings. For example, different labels can be used to represent the driving preference. For example, labels such as "aggressive," "conservative," and "neutral" can be used to represent the first information. The label "aggressive" can indicate that the user's driving style is more aggressive. For example, when driving a vehicle to a traffic light and the green light is short, a user with an aggressive driving style may increase their speed to pass the intersection. Correspondingly, a user with a conservative driving style may reduce their speed and wait for the next traffic light. For example, numbers, letters, and characters can be used to represent different driving styles. For example, the numbers "1," "2," and "3" can be used to represent users with different driving styles. For example, the number "1" represents a user with an aggressive driving preference, while the number "5" represents a user with a more conservative driving style. It should be understood that the above description of driving styles is merely an example for ease of explanation. Different driving preferences can also be distinguished in other ways, and the embodiments of the present application are not limited to this.

[0098] It should be understood that the acquisition, storage, use, and processing of user information involved in this application are all compliant with local laws and regulations. For example, the acquisition, storage, use, and processing of user information involved in this patent are operations that have been performed with the consent of the individual.

[0099] In some possible implementations, the driver model may be determined based on the first information.

[0100] For example, since the first information can be used to indicate the user's manipulation preference, a matching driver model can be determined based on the first information. For example, a matching driver model can be determined based on the first information, such as by querying, etc. For the sake of brevity, this is not further described here.

[0101] Since different users may have different operating preferences, and the same user may have different operating preferences under different working conditions, in an embodiment of the present application, the first information can be obtained, and then the driver model can be determined based on the first information, thereby improving the accuracy of abnormal driving behavior detection.

[0102] S430: According to the first motion state parameter and the first environmental information, based on the driver model, a first expected manipulation parameter may be determined.

[0103] For example, based on the motion state parameters of the vehicle and the corresponding environmental information, based on the driver model, the user's control parameters for the vehicle can be predicted in the motion state and surrounding environment, that is, the expected control parameters. The expected control parameters can refer to the control parameters determined based on the driver model. For example, based on the driver model determined by method 200, the expected control parameters at the first moment can be determined according to the surrounding environment and motion state of the vehicle. For example, the target speed and / or target path of the vehicle in this scenario can be planned based on the motion state parameters and environmental information. Based on the target speed and / or target path, the driver model can output the expected control parameters to predict the user's control of the vehicle when the vehicle is in the motion state and the environment, that is, the first expected control parameters. It should be understood that the above method for determining the first expected control parameters is only an example for the sake of explanation, and the embodiments of the present application do not limit this.

[0104] For example, when obtaining user-input control parameters at multiple moments, along with corresponding motion state parameters and environmental information, a plurality of corresponding expected control parameters can be determined based on the driver model. For example, when obtaining a first control parameter sequence, a first expected control parameter sequence can be determined based on the corresponding motion state parameter sequence and environmental information sequence, and based on the driver model. This first expected control parameter sequence can include a plurality of expected control parameters. For the sake of brevity, this will not be further described here.

[0105] In some possible implementations, the driver model can be determined based on a historical sequence of operating parameters. For example, while the vehicle is in motion, active operating parameters can be acquired and recorded in real time. A matching driver model can be determined based on the user's vehicle operation over a specific period of time. The driver model can also be optimized and adjusted based on the active operating parameters over that period of time, thereby ensuring the accuracy of the abnormal behavior detection method.

[0106] S450: Determine whether there is abnormal driving behavior based on the first manipulation parameter and the first expected manipulation parameter.

[0107] For example, when there is a large deviation between the first manipulation parameter and the first expected manipulation parameter, it can be determined that there is abnormal driving behavior. For example, taking the vehicle as the vehicle and the manipulation parameters as the accelerator pedal stroke, the brake pedal stroke and the steering wheel angle, among the first manipulation parameters obtained, the accelerator pedal stroke is 15 centimeters (cm), the brake pedal stroke is 0, and the steering wheel angle is 0, that is, the user did not control the vehicle to decelerate in this scenario, while among the expected manipulation parameters determined by the driver model, the accelerator pedal stroke is 0 centimeters, the brake pedal stroke is 15 centimeters, and the steering wheel angle is 0, that is, the driver model predicts that the user needs to control the vehicle to decelerate in this scenario, thus, it can be considered that there is abnormal driving behavior; for another example, the threshold of the difference between the first manipulation parameter and the first expected manipulation parameter can be determined by looking up a table. When the difference between the two is greater than or equal to the threshold, it can be determined that there is abnormal driving behavior. It should be understood that the above method of determining abnormal driving behavior is only an example for the sake of explanation, and the embodiments of the present application are not limited to this.

[0108] For example, when acquiring multiple active steering parameters, the presence of abnormal driving behavior can be determined based on the multiple active steering parameters and the corresponding multiple expected steering parameters determined based on the driver model. For example, the deviation between each active steering parameter and the corresponding expected steering parameter can be determined based on the multiple active steering parameters and the multiple expected steering parameters. When the cumulative value of the deviations between the multiple active steering parameters and the multiple expected steering parameters is greater than or equal to a set threshold, abnormal driving behavior can be considered to exist. The set threshold can be preset or calculated based on the operating conditions of the vehicle, and this application is not limited to this. For another example, the value of an evaluation function can be determined based on the multiple active steering parameters and the expected steering parameters, thereby determining the presence of abnormal driving behavior.

[0109] For example, taking a vehicle as an example, the first manipulation parameter may include a steering wheel angle, accelerator pedal travel, brake pedal travel, etc. input by the user, and accordingly, the first expected manipulation parameter may also include a steering wheel angle, accelerator pedal travel, brake pedal travel, etc. determined by the driver model. For ease of explanation, in the following embodiments of the present application, (δ, α, F b ) represents the first manipulation parameter, wherein δ can be used to represent the steering wheel angle input by the user, α can be used to represent the accelerator pedal stroke output by the user, and F b It can be used to represent the brake pedal stroke input by the user. The accelerator pedal stroke and the brake pedal stroke can be actual pedal strokes or normalized pedal strokes (for example, normalizing the pedal stroke input by the user with the total pedal stroke). Accordingly, (δ m , αm 、F bm ) represents the first expected control parameter, where δ m Can be used to represent the steering wheel angle determined by the driver model, α m It can be used to represent the accelerator pedal travel determined by the driver model, F bm It can be used to represent the brake pedal travel determined by the driver model. It should be understood that the first operating parameter and the first expected operating parameter can also be other operating parameters such as the vehicle gear position. For the sake of brevity, no examples are given here one by one. It should be understood that the above representation methods of the first operating parameter and the first expected operating parameter are only examples for the convenience of description and are not limited to this embodiment of the present application.

[0110] For example, based on the first manipulation parameter and the first expected manipulation parameter, a first confidence level can be determined, and the first confidence level can be used to indicate the degree of similarity between the first manipulation parameter and the first expected manipulation parameter. For example, taking a vehicle as an example, based on the first manipulation parameter (δ, α, F b ) and the first expected manipulation parameter (δ m , α m 、F bm ), the residual vector between the two can be determined in, ε1=(δ-δ m ), ε2=(α-α m ),ε3=(F b -F bm ), that is, the elements in the residual vector can be used to represent the difference between the multiple control parameters output by the user and the corresponding multiple control parameters predicted by the driver model. Accordingly, the length of the residual vector is It can be used to represent the degree of difference between the first manipulation parameter and the first expected manipulation parameter. The larger the length of the residual vector, the greater the difference between the first manipulation parameter and the first expected manipulation parameter. In some possible implementations, for ease of characterization and statistics, the length of the residual vector can be mapped to the interval [0, 1] to represent the degree of similarity between the first manipulation parameter and the first expected manipulation parameter. In other words, the first confidence level can be a function of the length of the residual vector. For example, γ is used to represent the first confidence level. The larger the first confidence level γ is, the higher the similarity between the first manipulation parameter and the first expected manipulation parameter is.

[0111] It should be understood that the above method for determining the first confidence level is merely an example for ease of explanation, and other methods may be used to determine the first confidence level. For example, different weights may be assigned to multiple control parameters, such as steering wheel angle and accelerator pedal travel, based on their impact on driving safety, and the first confidence level may be determined based on these weights. It should be understood that the present embodiment does not limit the method for determining the first confidence level.

[0112] For example, when the first confidence level is greater than or equal to a first threshold, it may be determined whether abnormal driving behavior exists based on the first manipulation parameter and the first expected manipulation parameter.

[0113] In an embodiment of the present application, when the first confidence level is greater than or equal to the first threshold value, the compliance between the first manipulation parameter and the first expected manipulation parameter is good, and it can be considered that the driver model is in a normal operating state. Therefore, whether there is abnormal driving behavior is determined based on the first manipulation parameter and the first expected manipulation parameter, which can reduce the misjudgment of abnormal driving behavior.

[0114] For example, whether abnormal driving behavior exists can be determined based on multiple active steering parameters and multiple expected steering parameters within a first time period. The first time period can be the duration of a sampling window for the active steering parameters or the duration of a time window used to detect abnormal driving behavior. The multiple first steering parameters and the multiple first expected steering parameters have a one-to-one correspondence.

[0115] In this embodiment of the present application, abnormal driving behavior within a first time period can be detected based on multiple active control parameters and corresponding expected control parameters within that time period. Because a user's control of a vehicle is typically consistent, detecting and identifying abnormal driving behavior based on multiple control parameters input by the user within the first time period helps improve the accuracy of detecting and identifying abnormal driving behavior.

[0116] Exemplarily, during the user's operation of a vehicle, abnormal driving behavior can be detected based on a time window. The duration of the time window can be the first duration. Within the first duration, the user can output multiple sets of active control parameters. Accordingly, the driver model can also output multiple sets of expected control parameters. Exemplarily, the multiple sets of active control parameters can constitute a first control parameter sequence in a preset sorting manner. Accordingly, the multiple sets of corresponding predicted control parameters can constitute a first expected control parameter sequence. The first control parameter sequence and the first expected control parameter sequence can have the same sorting manner, that is, the first control parameter sequence can include active control parameters at N moments. The first expected control parameter sequence can include expected control parameters at N moments. The active control parameters in the first control parameter sequence can have a one-to-one correspondence with the expected control parameters of the same sorting in the first expected control parameter sequence. N is a positive integer greater than or equal to 2. For example, taking the vehicle as an example, for example, (δ i , α i 、F bi ) represents the i-th group of active manipulation parameters in the first manipulation parameter sequence (for example, the active manipulation parameters at the first moment), and accordingly, (δ mi , α mi 、F bmi ) represents the expected manipulation parameters of the i-th group in the first expected manipulation parameter sequence (for example, the expected manipulation parameters at the first moment predicted based on the driver model). Accordingly, The residual vector between the i-th group of active steering parameters and its corresponding expected steering parameters in the first steering parameter sequence is represented. A sequence of multiple residual vectors composed of the order of multiple groups of active steering parameters can be simply referred to as a residual sequence, where i is a positive integer less than or equal to N.

[0117] For example, the first confidence level can also be used to indicate the similarity between the multiple sets of active control parameters and the multiple sets of expected control parameters. For example, taking the vehicle as an example, the first time period may include n sets of active control parameters (δ i , α i 、F bi ) and the expected control parameters (δ mi , α mi 、F bmi ), accordingly, the residual vector can be determined Wherein, n is a positive integer greater than or equal to 2, and i is a positive integer less than or equal to n. For example, the similarity between the multiple sets of active control parameters and the multiple sets of expected control parameters can be determined based on the Euclidean distance. For ease of explanation, the Euclidean distance is represented by d, where The first confidence level γ can be a function of the Euclidean distance d. For example, to facilitate characterization and statistics, the Euclidean distance can be mapped to the interval [0, 1] to represent the degree of similarity between the multiple sets of active control parameters and the multiple sets of expected control parameters, i.e., to determine the first confidence level. For the sake of brevity, this is not further described here.

[0118] It should be understood that the above description of the first manipulation parameter sequence and the method of the first confidence level are merely examples for the sake of illustration, and the embodiments of the present application do not limit this.

[0119] For example, when the first confidence level is greater than or equal to a first threshold, it may be determined whether abnormal driving behavior exists based on multiple sets of first manipulation parameters and multiple sets of second parameters within the first time period.

[0120] In an embodiment of the present application, when the first confidence level is greater than or equal to the first threshold, determining whether there is abnormal driving behavior is based on the first manipulation parameter sequence and the first expected manipulation parameter sequence. This can avoid abnormal detection caused by environmental information, motion state parameters, and driver model operation errors, and can improve the accuracy of abnormal driving behavior detection.

[0121] In some possible implementations, when the first confidence level is greater than or equal to a second threshold, the driver model may be optimized based on the first manipulation parameter sequence, and the second threshold may be less than or equal to the first threshold. For example, when it is determined that there is no abnormal driving behavior and the first confidence level is greater than or equal to the second threshold, the driver model may be optimized based on the first manipulation parameter sequence. This may make the expected manipulation parameters determined by the driver model more consistent with the user's driving behavior, thereby improving the accuracy of abnormal driving behavior detection.

[0122] For example, determining whether there is abnormal driving behavior based on the first manipulation parameter sequence and the first expected manipulation parameter sequence may be based on the residual sequence and the evaluation function to determine whether there is abnormal driving behavior. For example, taking the vehicle as an example, the active manipulation parameter and the expected manipulation parameter may be functions that change over time, such as the active manipulation parameter (δ(t), α(t), F b (t)), expected control parameters (δ m (t), α m (t), F bm (t)), the start and end time of the time window are represented by t1 and t2 respectively. The residual vector can also be a function that changes with time. It can be determined whether there is abnormal driving behavior based on the evaluation function. For example, the evaluation function can be Where k1, k2, and k3 are weight coefficients of the control parameters, J is the value of the evaluation function, or the first evaluation value, which can be used to evaluate whether there is abnormal driving behavior, and ε1(t) can be used to represent the time-varying difference between the steering wheel angle output by the user and the steering wheel angle output by the driver model. In other words, ε1(t) = δ(t) - δ m (t), when the calculation result of the evaluation function, or the first evaluation value, is greater than or equal to the third threshold, it can be determined that the vehicle has abnormal driving behavior; for example, when determining multiple active control parameters within a time window according to a certain sampling frequency, the evaluation function can be calculated in a discrete manner, which will not be repeated here for the sake of brevity.

[0123] Exemplarily, the working condition of the vehicle can be determined based on at least one of the motion state information and environmental information. For example, taking the vehicle as an example, the working condition of the vehicle can be determined based on the working condition recognition model through processing methods such as fragment division and feature extraction according to the vehicle's motion state (such as the vehicle's speed, acceleration, angular velocity, etc.), environmental information (such as road slope, expected speed on the road, etc.) and the first control parameter (such as the steering wheel angle input by the user, accelerator pedal stroke, brake pedal stroke, etc.). For example, it can be determined based on the above parameters that the vehicle is in starting, accelerating, decelerating, turning, high-speed cruising, lane change to avoid collision, parking, etc., or other working condition types. For the sake of brevity, examples will not be given one by one. Furthermore, an evaluation function can be determined based on the determined working condition to determine whether there is abnormal driving behavior. For example, when the evaluation function sets corresponding multiple weight coefficients for multiple control parameters, for example, the evaluation function is J=k1ε1 2 +k2ε2 2 +k3ε3 2 , one or more weight coefficients, such as at least one of k1, k2, and k3, can be determined based on the operating conditions to determine the evaluation function. For another example, because user driving behavior may vary under different operating conditions, the evaluation function under different operating conditions can take different functional forms. For example, the evaluation function can be a linear function, a quadratic function, a power function, an exponential function, etc. For the sake of brevity, each example is not provided here. It should be understood that the above method of determining the evaluation function based on the operating condition of the vehicle is merely an example and is not limited to this embodiment of the present application.

[0124] For example, an evaluation function may be determined based on the first information. For example, since a driver model may be determined based on the user's manipulation preference, an evaluation function may be determined based on the first information. For example, the evaluation function is J=k1ε1 2 +k2ε2 2 +k3ε3 2According to the first information, a corresponding set of weight coefficients k1, k2, and k3 can be determined, so that the evaluation criteria for abnormal driving behaviors can be set according to different user operation preferences, thereby reducing the possibility of misjudgment and improving the accuracy of identifying abnormal driving behaviors.

[0125] For example, the evaluation function can be determined in combination with the operating condition and the first information. For example, after determining the operating condition and the first information of the vehicle, the corresponding evaluation function can be determined by, for example, a two-dimensional table lookup method, etc., which will not be described here for brevity.

[0126] For example, it is possible to determine whether there is abnormal driving behavior based on the evaluation function according to the multiple sets of first and second manipulation parameters obtained within the first time period. For example, taking the vehicle as the vehicle, the duration of the sampling time window as the first time period, and its start and end times as t1 and t2 respectively, the evaluation function can be The presence of abnormal driving behavior can be determined based on the residuals between multiple sets of first and second manipulation parameters within the time window and the evaluation function. For example, abnormal driving behavior can be determined when the calculation result of the evaluation function is greater than or equal to a third threshold. For another example, abnormal driving behavior can be determined when the calculation result of the evaluation function within a second time period is greater than or equal to the third threshold, where the second time period may include one or more sampling time windows. It should be understood that the above method for determining abnormal driving behavior is merely an example and is not limited in this embodiment of the present application.

[0127] It should be understood that the above description of the evaluation function is only an example for ease of explanation, and the embodiments of the present application do not limit this.

[0128] In an embodiment of the present application, the presence of abnormal driving behavior for the vehicle is determined based on the manipulation parameters input by the user and the manipulation parameters of the driver model, making it possible to detect abnormal driving behavior without adding new sensors. Moreover, because this detection method is directly based on the user's manipulation parameters for the vehicle, it can avoid the limitations of sensors in detecting abnormal driving behavior and reduce the delay in detecting abnormal driving behavior. Furthermore, because the driver model is determined based on the user's manipulation preferences, differentiated configuration of the driver model can be achieved, which can improve the accuracy of the abnormal driving behavior detection method.

[0129] When detecting abnormal driving behavior, obtaining the control parameters output by the user when normally operating the vehicle by querying the database may require a large database, and obtaining the database may also require a large amount of engineering work. Moreover, the database located in the cloud may be affected by the communication quality. Due to the limitation of storage capacity, the database stored locally may not be able to obtain accurate control parameters. Therefore, in the embodiment of the present application, the expected control parameters can be regarded as the control parameters output by the user when normally operating the vehicle. The expected control parameters are obtained based on the driver model, which can avoid the limitations of storage space, communication quality, database acquisition, etc. However, during the user's operation of the vehicle, due to the influence of factors such as the surrounding environment, the vehicle's motion state, and the user's own operating habits, the use of a fixed driver model will result in a low accuracy in detecting abnormal driving behavior. In the embodiment of the present application, the driver model is determined by the historical operation sequence, which can improve the accuracy of identifying abnormal driving behavior. In addition, the driver model is determined based on the user's operation preferences, which can also achieve differentiated settings for the driver model and improve the accuracy of identifying abnormal driving behavior.

[0130] For example, taking a vehicle as an example, FIG5 is a schematic flow chart of a method for identifying abnormal driving behavior provided by an embodiment of the present application. The method 500 includes some or all of the following steps:

[0131] S510: Acquire a first manipulation parameter sequence.

[0132] For example, when a user controls the vehicle based on their perception of the environment, active manipulation parameters for controlling the vehicle's movement can be obtained. For example, in the case of a vehicle, the first manipulation parameters may include steering wheel angle, accelerator pedal travel, brake pedal travel, etc. The vehicle can adjust its motion state, such as speed, acceleration, angular velocity, etc., based on the first manipulation parameters.

[0133] For example, multiple active manipulation parameters input by the user within a first duration may be obtained. Further, the multiple active manipulation parameters within the first duration may constitute a first manipulation parameter sequence. For the sake of brevity, further details are omitted here.

[0134] For example, the number of active manipulation parameters included in the first manipulation parameter sequence can be greater than or equal to a preset threshold. For example, taking the preset threshold as 6, the first manipulation parameter sequence can be composed of 6 or more active manipulation parameters. For the sake of brevity, this is not further described here. The preset threshold can also be other values ​​(such as 3, 10, etc.), which is not limited in this embodiment of the present application.

[0135] S513: Acquire the vehicle's motion state information and environment information.

[0136] For example, the description of the vehicle's motion state information and environmental information can refer to steps S220 and S230, which will not be repeated here for the sake of brevity.

[0137] For example, when a plurality of active manipulation parameters are acquired, a corresponding plurality of motion state parameters and environmental information may be acquired accordingly.

[0138] It should be understood that the active control parameters, the vehicle's motion state information and environmental information can be obtained simultaneously, or the active control parameters can be obtained first, or the motion state information can be obtained first, or the environmental information can be obtained first. The embodiments of the present application do not limit this.

[0139] S516: Acquire first information, and determine a driver model based on the first information.

[0140] Exemplarily, the first information can be determined based on historical operating parameters. For example, the historical operating data can be active operating parameters previously stored in a database, such as driving data retained from a user's previous driving of the vehicle. The historical operating parameters can also be operating parameters generated during the current driving process. For example, during the current driving process, the user has been operating the vehicle for 15 minutes, and the active operating parameters within 10 minutes of the current driving process can be obtained. For example, the user's operating preferences can be determined through comparison, matching, etc., thereby determining the first information. For the sake of brevity, this is not further described here.

[0141] Exemplarily, a driver model may be determined based on the first information.

[0142] It should be understood that for the description of the first information and the driver model, reference may be made to steps S410 to S430 , etc., which will not be repeated here for the sake of brevity.

[0143] S520 : Determine a first expected manipulation parameter sequence according to the vehicle's motion state information and environment information and based on a driver model.

[0144] For example, the description of determining the expected manipulation parameters can be referred to in step S430 , which will not be repeated here for the sake of brevity.

[0145] For example, when acquiring multiple active steering parameters within a first duration, multiple expected steering parameters can be determined based on multiple sets of environmental information and vehicle motion state information, based on a driver model. These multiple sets of expected steering parameters can constitute a first expected steering parameter sequence. For the sake of brevity, this is not further described here.

[0146] S525 : Determine a residual sequence according to the first manipulation parameter sequence and the first expected manipulation parameter sequence.

[0147] Exemplarily, based on multiple active manipulation parameters and corresponding multiple expected manipulation parameters, multiple residual vectors can be determined, and the residual vectors can be used to represent the difference between the set of first manipulation parameters and the corresponding second manipulation parameters. Accordingly, the multiple residual vectors can be composed into a residual sequence based on the sorting of the first manipulation parameter sequence. For the sake of brevity, they will not be repeated here.

[0148] S530: Determine a first confidence level.

[0149] For example, the first confidence level can be determined based on the residual sequence. i , α i 、F bi ) and the expected control parameters (δ mi , α mi 、F bmi ) can be expressed as Representation, for example, can be based on the Euclidean distance between the residual vector and the zero point This represents the degree of similarity between the active control parameters and the expected control parameters. Furthermore, for statistical purposes, the Euclidean distance can be mapped to the interval [0, 1], with the mapping result serving as the first confidence factor γ. For example, γ = 1 - 2*arctan(d) / π. It should be understood that the above method for determining the first confidence factor is merely an example for illustrative purposes and is not intended to be limiting in this embodiment of the present application.

[0150] S535, determine whether the first confidence level is greater than or equal to a first threshold. If the first confidence level is greater than the first threshold, jump to step S545.

[0151] For example, the description of the first threshold value may refer to step S450 , which will not be repeated here for the sake of brevity.

[0152] Optionally, at S540, the vehicle's operating condition is determined based on the first maneuvering parameter, the environmental information, and the vehicle's motion state information. For example, the vehicle's speed and acceleration can be used to determine whether the vehicle is accelerating; the distribution of obstacles around the vehicle can be determined based on the environmental information, and combined with the vehicle's motion state information, it can be determined whether the vehicle is in a lane-changing obstacle avoidance condition, etc. It should be understood that descriptions of vehicle operating conditions and operating condition types can refer to related technologies, and for the sake of brevity, examples are not provided here.

[0153] S545, determining the evaluation function.

[0154] For example, the evaluation function can be determined based on the operating conditions. For example, the form of the evaluation function can be determined by table lookup, or the weights of multiple control parameters in the evaluation function can be determined based on the operating conditions. For the sake of brevity, examples are not given here one by one.

[0155] Exemplarily, the evaluation function may be determined based on the first information, or may be determined based on the first information combined with the working conditions. For the sake of brevity, details are not given here.

[0156] For example, the description of the evaluation function may refer to step S450 , which will not be repeated here for the sake of brevity.

[0157] S550, determine whether there is abnormal driving behavior. When there is abnormal driving behavior, you can jump to step S555.

[0158] For example, the first evaluation value can be determined by the evaluation function according to the residual sequence. For example, the duration of the sampling time window is the first duration, and its start and end times are t1 and t2 respectively, and the residual sequence is The evaluation function is For example, at least one of the weight coefficients k1, k2, and k3 can be determined based on the operating conditions and / or the first information, thereby determining the evaluation function. A first evaluation value can be determined by the evaluation function based on the residual sequence. When the first evaluation value is greater than or equal to the third threshold, it can be determined that abnormal driving behavior exists within the sampling time window. For another example, the second time period can include multiple sampling time windows of the first time period (for example, including 5 sampling time windows). The corresponding first evaluation value can be determined based on the residual sequence of each sampling time window of the first time period. When the number of the multiple first evaluation values ​​greater than or equal to the third threshold is not less than a preset number (for example, the preset number is 3), for example, 3 of the first evaluation values ​​of the 5 sampling time windows are greater than or equal to the third threshold, it can be determined that abnormal driving behavior exists within the second time period. For another example, when multiple sampling time windows with first evaluation values ​​greater than or equal to the third threshold are continuous within the second time period and the number is greater than a preset number (for example, the preset number is 2), it can be determined that abnormal driving behavior exists. It should be understood that the above method for determining whether abnormal driving behavior exists is only an example for ease of explanation and is not limited in the embodiments of the present application.

[0159] Optionally, S555 prompts the user that there is abnormal driving behavior.

[0160] For example, when abnormal driving behavior is determined, the user can be reminded. For example, taking a vehicle as an example, when abnormal driving behavior is determined, text, images, etc. can be displayed on the vehicle's screen to remind the user. For example, the words "Please pay attention to driving safety" can be displayed on the vehicle's central control screen, and the words "Please drive safely" can be displayed on the vehicle's front windshield through a heads-up display. The user can also be reminded through voice, such as the vehicle using a buzzer alarm or voice broadcast "Please pay attention to driving safety", etc. For the sake of brevity, examples will not be given one by one. It should be understood that the above methods of reminding the user are only examples for ease of explanation and are not limited to these in the embodiments of the present application.

[0161] Optionally, when abnormal driving behavior is determined, a first function can be triggered. For example, the first function can be an autonomous driving function, an advanced driver assistance function, or the like. For example, during manual driving, a user's manipulation of the vehicle may be identified as abnormal driving behavior due to limitations such as blind spots. However, since the vehicle's perception system can better perceive the surrounding environment, when abnormal driving behavior is determined, the autonomous driving function or advanced driver assistance function can be directly triggered. Alternatively, a voice prompt or central control screen can be used to prompt the user whether to enable the autonomous driving function or advanced driver assistance function, such as a voice prompt such as "Suggest you switch to advanced driver assistance." For another example, when driving on a highway, after abnormal driving behavior is detected, if the abnormal driving behavior continues to be detected for a preset time, an automatic parking function can be triggered to control the vehicle to automatically park on a safe road section. For another example, when the chip or abnormal driving behavior recognition device determines that abnormal driving behavior has occurred, the chip or abnormal driving behavior recognition device can directly trigger the first function, or send a message to another device to instruct the other device to trigger the first function. For the sake of brevity, this will not be described in detail here.

[0162] Optionally, in S560 , it is determined whether the first confidence level is greater than or equal to a second threshold. If the first confidence level is greater than or equal to the second threshold, the process may jump to step S565 .

[0163] For example, when the first confidence level is greater than or equal to a second threshold, the driver model may be optimized according to the first manipulation parameter sequence.

[0164] For example, the description of the second threshold value may refer to step S440 , which will not be repeated here for the sake of brevity.

[0165] It should be understood that step S560 and step S535 can be performed simultaneously, or step S535 can be performed first, or step S560 can be performed first. This embodiment of the present application does not limit this.

[0166] Optionally, S565, driver model optimization.

[0167] For example, when the first confidence level is greater than or equal to the second threshold, the driver model can be optimized based on the residual sequence using an optimization algorithm, such as a taboo algorithm, a genetic algorithm, a particle swarm optimization algorithm, an ant colony algorithm, etc., which are not listed here. For example, using the particle swarm optimization algorithm as an example, the positions of the particles in the N-dimensional space can represent the control parameters that need to be optimized, and the model predictive control method can be used to establish / optimize the driver model, such as X i =(X1,X2,X3,···,X n )=(N p ,N c ,Q,R,Δv,Δy···), where N p ,N c , Q, R, Δv, Δy, etc. can represent the prediction time domain, control time domain, control matrix parameters, state matrix parameters, expected speed correction and expected lateral displacement correction, that is, the control parameters that need to be optimized. For example, according to V id =ωV id +C1radm(0,1)(P id -X id )+C1radm(0,1)(P gd -X id ), X id =X id +V id Update the speed and position of each particle, where C1 and C2 can be acceleration factors, radm(0,1) can be a random number between the interval (0,1), and V id Can be the velocity vector of the i-th particle in the d-th dimension, X id It can represent the position vector of the i-th particle in the d-th dimension, ω is the inertia factor, its value is non-negative, P id is the position vector of the d-dimensional variable of the optimal solution of the i-th particle in the iterative process, P gd is the position vector of the d-th dimension variable of the optimal solution among all particles, where d is a positive integer. For the sake of brevity, this is not described in detail here. It should be understood that the description of the particle swarm optimization algorithm can refer to the relevant technology, and for the sake of brevity, this is not described in detail here.

[0168] For example, when the first confidence level is greater than or equal to the second threshold, the driver model may be optimized using a loss function based on the first manipulation parameter sequence and the first expected manipulation parameter. For example, a residual sequence may be determined based on the first manipulation parameter sequence and the manipulation parameter predicted by the driver model after optimization by the particle swarm algorithm, and a first loss value may be determined using a loss function. For example, the loss function may be: Here, cost can be the first loss value, and f can be a functional relationship mapping the residual to the first loss value, i.e., a loss function, thereby adjusting the driver model based on the first loss value. It should be understood that the description of the loss function can be referred to in related art and is not repeated here for the sake of brevity.

[0169] Exemplarily, the optimization of the driver model can be achieved by a cloud server or by the vehicle. For example, after determining that the first confidence level is greater than or equal to a second threshold, the first manipulation parameter sequence, as well as the corresponding multiple environmental information and motion state parameters, can be sent to the cloud server, so that the cloud server can optimize the driver model. Accordingly, the optimized driver model can be configured to the vehicle, so that the vehicle can determine whether abnormal driving behavior exists based on the optimized driver model. For another example, after determining that the first confidence level is greater than or equal to the second threshold, the vehicle can optimize the driver model based on the first manipulation parameter sequence, as well as the corresponding multiple environmental information and motion state parameters, so that the driver model can better fit the user's manipulation preferences and improve detection accuracy.

[0170] It should be understood that the above method of optimizing the driver model is only an example for ease of explanation and is not limited to this embodiment of the present application.

[0171] In actual driving, since the user's manipulation of the vehicle is continuous, obtaining a first manipulation parameter sequence within a period of time and detecting abnormal driving behavior based on it can help improve the accuracy of detecting abnormal driving behavior.

[0172] For example, taking a vehicle as an example, when the vehicle includes a driver monitoring system (DMS), when the user is driving the vehicle, the DMS system may not be able to identify the user's fatigue driving behavior due to limitations such as the user's clothing, camera shooting angle, blind spots, etc. However, since the user's manipulation of the vehicle can be reflected in the manipulation parameters, the abnormal driving behavior can be identified based on the active manipulation parameters and the driver model, so that the user's abnormal driving behavior can be identified even when the DMS system cannot work normally, which can help improve driving safety and reduce the occurrence of traffic accidents.

[0173] It should be understood that the above method 400 can be combined with the method 500, and the embodiment of the present application does not limit this.

[0174] The present application also provides an apparatus for implementing any of the above methods. For example, an apparatus includes units for implementing each step performed by a chip, a vehicle, an abnormal driving behavior identification device, and the like in any of the above methods. For example, please refer to Figure 6, which is a schematic diagram of the structure of an apparatus for abnormal driving behavior identification provided in an embodiment of the present application. The apparatus 700 may include an acquisition module 710 and a processing module 720.

[0175] Among them, the acquisition module 710 can be used to obtain a first manipulation parameter, a first motion state parameter and a first environmental information, wherein the first manipulation parameter is a manipulation parameter for controlling the driving of a vehicle at a first moment, the first motion state parameter is used to indicate the motion state of the vehicle at the first moment, and the first environmental information is used to indicate the surrounding environment in which the vehicle is located at the first moment; the processing module 720 can be used to determine a first expected manipulation parameter based on the driver model according to the first motion state parameter and the first environmental information; and determine whether there is abnormal driving behavior according to the first manipulation parameter and the first expected manipulation parameter.

[0176] For example, the description of the first manipulation parameter, the first motion state parameter and the first environmental information can refer to step S410, which will not be repeated here for the sake of brevity.

[0177] Exemplarily, the acquisition module 710 may also be used to acquire a historical manipulation parameter sequence, and the processing module 720 may also be used to determine the driver model based on the historical manipulation parameter sequence.

[0178] Exemplarily, the acquisition module 710 can also be used to: acquire a second manipulation parameter, a second motion state parameter and second environmental information, wherein the second manipulation parameter is a manipulation parameter for controlling the travel of the vehicle at the second moment, the second motion state parameter is used to indicate the motion state of the vehicle at the second moment, and the second environmental information is used to indicate the surrounding environment in which the vehicle is located at the second moment; the processing module 720 can also be used to: determine a second expected manipulation parameter based on the driver model according to the second motion state parameter and the second environmental information; the processing module 720 can specifically be used to: determine whether there is abnormal driving behavior based on the first manipulation parameter sequence and the first expected manipulation parameter sequence, wherein the first manipulation parameter sequence includes the first manipulation parameter and the second manipulation parameter, and the first expected manipulation parameter sequence includes the first expected manipulation parameter and the second expected manipulation parameter.

[0179] For example, the description of the first manipulation parameter sequence and the first expected manipulation parameter sequence may refer to steps S410 to S450 , which will not be repeated here for the sake of brevity.

[0180] Exemplarily, the processing module 720 can also be used to: determine a first confidence level based on the first manipulation parameter sequence and the first expected manipulation parameter sequence, wherein the first confidence level is used to characterize the degree of similarity between the first manipulation parameter sequence and the first expected manipulation parameter sequence; the processing module 720 is specifically used to: determine whether there is abnormal driving behavior based on the first manipulation parameter sequence and the first expected manipulation parameter sequence when the first confidence level is greater than or equal to a first threshold value.

[0181] For example, the description of the first confidence level may refer to step S450 , which will not be repeated here for the sake of brevity.

[0182] Exemplarily, the processing module 720 can also be used to: optimize the first prediction model according to the first manipulation parameter sequence and the first expected manipulation parameter sequence when the first confidence is greater than or equal to a second threshold, and the second threshold is less than the first threshold.

[0183] In some possible implementations, the acquisition module 710 may also be used to acquire first information, where the first information is used to indicate a user's manipulation preference; and the processing module 720 may also be used to determine a driver model based on the first information.

[0184] Exemplarily, the processing module 720 can also be used to: determine an evaluation function based on the first information; the processing module 720 can specifically be used to: determine the value of the evaluation function based on the first manipulation parameter sequence and the first expected manipulation parameter sequence; when the value of the evaluation function is greater than or equal to a third threshold, determine that abnormal driving behavior exists.

[0185] Exemplarily, the processing module 720 may also be used to: determine the operating condition type of the vehicle according to the first motion state parameter and / or the first state parameter; and determine the evaluation function according to the operating condition type of the vehicle.

[0186] Exemplarily, the processing module 720 may also be configured to: when determining that abnormal driving behavior exists, prompt the user that abnormal driving behavior exists.

[0187] In some possible implementations, the processing module 720 may also be used to control the vehicle to be in an automatic driving mode when abnormal driving behavior is determined to exist.

[0188] Exemplarily, the transportation tool is a vehicle, and the maneuvering parameter includes at least one of a steering wheel angle, an accelerator pedal travel, and a brake pedal travel.

[0189] It should be understood that the device for identifying abnormal driving behavior shown in Figure 6 can be used to implement the above-mentioned method 400, and the device shown in Figure 6 can also be used to implement the method for identifying abnormal driving behavior described in method 500. The specific steps can refer to the above description of Figures 3 to 5. For the sake of brevity, this embodiment of the present application will not be repeated.

[0190] For example, Figure 7 shows a schematic flow chart of a system for identifying abnormal driving behavior. As shown in Figure 7, the system for identifying abnormal driving behavior may include a model determination unit, a residual calculation unit, an abnormal driving behavior identification unit, an abnormal driving behavior reminder unit, etc.

[0191] In some possible implementations, the model determination unit can be used to determine the parameters of an appropriate driver model based on the provided historical driving data. For example, when the vehicle is powered on, after identifying the driver's identity, the parameters of the corresponding driver model can be determined based on the driver's historical operating parameters. For another example, the parameters of the corresponding driver model can be determined based on the operating preferences selected by the driver. For the sake of brevity, this will not be described in detail here. The model determination unit can be located in the vehicle or in a cloud server. When the model determination unit is located in the cloud server, the vehicle can obtain the parameters of the driver model through network devices, etc. It should be understood that the embodiments of the present application are not limited to this.

[0192] For example, the residual calculation unit can be used to determine a residual vector based on the active steering parameters and expected steering parameters input by the actual driver, thereby determining the degree of similarity between the active steering parameters and the expected steering parameters. For example, the steering parameters may include steering wheel angle, throttle opening, brake pedal travel, etc. For the sake of brevity, these parameters are not further described here.

[0193] For example, the abnormal driving behavior identification unit can be used to identify abnormal driving behavior. For example, to improve the accuracy of abnormal driving behavior identification, the abnormal driving behavior identification unit can also implement functions such as confidence calculation, evaluation function calculation, and operating condition identification. These functions can be combined to achieve abnormal driving behavior identification. For the sake of brevity, these functions will not be further described here.

[0194] For example, the abnormal driving behavior reminder unit can be used to remind users of abnormal driving behavior. For example, it can control related devices to prompt the user of abnormal driving behavior through text, sound, light, etc. For the sake of brevity, this will not be described in detail here.

[0195] It should be understood that the division of the above-mentioned model determination unit, residual calculation unit, abnormal driving behavior detection unit, abnormal driving behavior reminder unit, etc. is merely a logical division. For example, the above-mentioned processing module 720 may include the residual calculation unit, and may also include the abnormal driving behavior detection unit, etc. For another example, the above-mentioned device 700 may include part or all of the model determination unit, driver model, residual calculation unit, abnormal driving behavior detection unit and abnormal driving behavior reminder unit, etc. For the sake of brevity, they will not be repeated here.

[0196] In some possible implementations, the driver manipulates the vehicle based on environmental information, and the driver model can output expected manipulation parameters based on the vehicle's motion state and environmental information. The residual calculation unit can then obtain the actual manipulation parameters output by the driver and the manipulation parameters output by the model, thereby determining a residual sequence. The abnormal driving behavior detection unit can determine whether abnormal driving behavior exists based on the residual sequence, such as by identifying abnormal driving behavior when a first confidence level meets a preset condition. Upon determining the presence of abnormal driving behavior, the abnormal driving behavior alert unit can indicate the presence of abnormal driving behavior.

[0197] For the convenience of explanation and description, the control parameters include steering wheel angle δ, accelerator pedal travel α, brake pedal travel F b For example, the embodiment of this application takes Indicates the steering wheel angle change rate, Indicates the rate of change of the driving pedal stroke. Other symbols are similar and will not be given examples here for the sake of brevity.

[0198] For example, Figure 8 shows a schematic flow chart of optimizing a driver model. For example, as shown in Figure 8, the driver model may include a planning module and a control module. The planning module may be used to plan the vehicle's target speed, target path, etc., and the control module may be used to determine expected maneuvering parameters.

[0199] For example, as shown in FIG8 , the driver model and abnormal driving behavior identification unit can obtain the vehicle's motion state information, such as vehicle speed v, acceleration a, angular acceleration γ, etc. The residual calculation unit can determine the residual vector based on the manipulation parameters output by the user and the manipulation parameters output by the driver model. Thus, the abnormal driving behavior identification unit can identify abnormal driving behavior based on the residual vector. Furthermore, the driver model and abnormal driving behavior identification unit can also include a learning unit, which can be used to optimize or adjust the driver model so that the driver model can more accurately identify the driver's abnormal driving behavior. For example, the learning module can be based on the objective function Based on an optimization algorithm (such as a particle swarm optimization algorithm), the driver model is optimized. When the driver model meets the termination condition (for example, the number of iterations n reaches a preset threshold, and the change between the objective functions corresponding to two consecutive iterations is less than an error threshold), the optimization of the planning module and / or the control module of the driver model can be terminated. For another example, as shown in FIG8 , when no abnormal driving behavior is detected, or the duration of the abnormal driving behavior is less than or equal to the time threshold T lim ′, the manipulation parameters corresponding to the time period can be provided to the learning module, so that the learning module can optimize the driver model so that the output of the driver model is closer to the driver's driving behavior.

[0200] For example, FIG9 is a flowchart of another method for identifying abnormal driving behavior provided in an embodiment of the present application.

[0201] For example, as shown in FIG9 , the working condition of the vehicle can be determined based on environmental information, such as obstacle information, and then the values ​​of the weight coefficients (e.g., at least one of k1, k2, and k3) in the evaluation function under the working condition can be determined based on the working condition of the vehicle, which can also be used to determine the threshold J of the evaluation function. lim and time threshold T lim , thereby determining whether there is abnormal driving behavior. Accordingly, the online learning unit can optimize and adjust the driver model based on the abnormal driving behavior identification result and the corresponding first manipulation parameter, so that the expected manipulation parameter output by the driver model unit can be closer to the driver's driving behavior.

[0202] It should be understood that the descriptions of Figures 7 to 9 are merely examples for ease of explanation and may also be combined with the contents of Figures 2 to 6 . The embodiments of the present application do not limit this.

[0203] It should be understood that the division of the various units or modules in the above devices is merely a division of logical functions. In actual implementation, they may be fully or partially integrated into a single physical entity, or they may be physically separated. Furthermore, the units or modules in the device may be implemented in the form of a processor calling software; for example, the device includes a processor connected to a memory storing instructions, and the processor calls the instructions stored in the memory to implement any of the above methods or the functions of the various units in the device. The processor may be a general-purpose processor, such as a CPU or a microprocessor, and the memory may be a memory within the device or external to the device. Alternatively, the units in the device may be implemented in the form of hardware circuits, and the functions of some or all of the units may be implemented through the design of the hardware circuits. The hardware circuits may be understood as one or more processors. For example, in one implementation, the hardware circuit is an ASIC, and the functions of some or all of the above units may be implemented through the design of the logical relationships between the components within the circuits. In another implementation, the hardware circuit may be implemented using a PLD, such as an FPGA, which may include a large number of logic gate circuits. The connections between the logic gate circuits are configured using a configuration file to implement the functions of some or all of the above units. All units of the above apparatus may be implemented entirely in the form of software called by a processor, or entirely in the form of hardware circuits, or partially in the form of software called by a processor and the rest in the form of hardware circuits.

[0204] In an embodiment of the present application, a processor is a circuit with the ability to process signals. In one implementation, the processor may be a circuit with the ability to read and execute instructions, such as a CPU, a microprocessor, a GPU, or a DSP. In another implementation, the processor may implement certain functions through the logical relationship of a hardware circuit, and the logical relationship of the hardware circuit 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 document to implement the configuration of the hardware circuit 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 an NPU, TPU, DPU, etc.

[0205] It can be seen that each unit in the above device can be one or more processors (or processing circuits) configured to implement the above method, such as: CPU, GPU, NPU, TPU, DPU, microprocessor, DSP, ASIC, FPGA, or a combination of at least two of these processor forms.

[0206] In addition, the various units in the above devices can be fully or partially integrated together, or can be implemented independently. In one implementation, these units are integrated together and implemented in the form of a system-on-a-chip (SOC). The SOC may include at least one processor for implementing any of the above methods or implementing the functions of the various units of the device. The type of the at least one processor can be different, for example, including a CPU and FPGA, a CPU and an artificial intelligence processor, a CPU and a GPU, etc.

[0207] For example, FIG10 is a structural diagram of an apparatus 1300 provided in an embodiment of the present application. Apparatus 1300 includes a processor 1302, a communication interface 1303, and a memory 1304. One example of apparatus 1300 is a chip. Another example of apparatus 1300 is a computing device.

[0208] The processor 1302, memory 1304, and communication interface 1303 can communicate via a bus. The memory 1304 stores executable code, and the processor 1302 reads the executable code in the memory 1304 to execute the corresponding method. The memory 1304 may also include other software modules required for running processes, such as an operating system.

[0209] For example, the executable code in the memory 1304 is used to implement the method shown in FIG. 3 to FIG. 5 , and the processor 1302 reads the executable code in the memory 1304 to execute the method shown in FIG. 3 to FIG. 5 .

[0210] The processor 1302 may be a CPU. The memory 1304 may include volatile memory (VM), such as random access memory (RAM). The memory 1304 may also include non-volatile memory (NVM), such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid state disk (SSD).

[0211] An embodiment of the present application also provides a vehicle, which may include the above-mentioned device 700, or the above-mentioned device 1300, or a device for implementing methods 200 to 500.

[0212] An embodiment of the present application further provides a computer program product, which includes: computer program code, which enables the computer to execute the above method when the computer program code is run on a computer.

[0213] An embodiment of the present application further provides a computer-readable storage medium, which stores a program code. When the computer program code runs on a computer, the computer executes the methods shown in Figures 3 to 5 above.

[0214] An embodiment of the present application also provides a chip, comprising: at least one processor and a memory, wherein the at least one processor is coupled to the memory and is configured to read and execute instructions in the memory to perform the methods of Figures 3 to 5 above.

[0215] The term "at least one" in this application means one or more, and the term "plurality" in this application means two or more.

[0216] In this application, the terms "first," "second," and so on are used to distinguish between identical or similar items with substantially the same purpose or function. It should be understood that "first," "second," and "nth" do not have a logical or temporal dependency, nor do they limit the quantity or order of execution. For example, "first manipulation parameter" and "second manipulation parameter" are used only to distinguish, and do not imply that the "first manipulation parameter" and "second manipulation parameter" have different priorities.

[0217] It should be understood that in the various embodiments of the present application, the size of the serial number of each process does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0218] It should be understood that determining B based on A does not mean determining B based solely on A. B can also be determined based on A and / or other information.

[0219] It should be understood that the term "and / or" in this document simply describes a relationship between related objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A exists alone, A and B exist simultaneously, or B exists alone. Furthermore, the character " / " in this document generally indicates that the related objects are in an "or" relationship.

[0220] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0221] As used in this specification, the terms "component," "module," "system," and the like are used to represent computer-related entities, hardware, firmware, a combination of hardware and software, software, or software in execution. For example, a component can be, but is not limited to, a process running on a processor, a processor, an object, an executable file, an execution thread, a program, and / or a computer. By way of illustration, both an application running on a computing device and a computing device can be a component. One or more components can reside in a process and / or an execution thread, and a component can be located on a computer and / or distributed between two or more computers. In addition, these components can be executed from various computer-readable media having various data structures stored thereon. Components can communicate, for example, via local and / or remote processes based on signals having one or more data packets (e.g., data from two components interacting with another component on a local system, a distributed system, and / or a network, such as the Internet interacting with other systems via signals).

[0222] 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.

[0223] 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.

[0224] 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 units is merely 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.

[0225] 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 these units may be selected to achieve the purpose of this embodiment according to actual circumstances.

[0226] 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.

[0227] If the functions are implemented in the form of software functional units and sold or used as independent products, they 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 described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory, a random access memory, a magnetic disk, or an optical disk.

[0228] The above description is merely a specific embodiment of the present application, but the scope of protection of the present 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 identifying abnormal driving behavior, characterized in that: include: Acquiring a first manipulation parameter, a first motion state parameter, and first environmental information, wherein the first manipulation parameter is a manipulation parameter for controlling the travel of a vehicle at a first moment, the first motion state parameter is used to indicate the motion state of the vehicle at the first moment, and the first environmental information is used to indicate the surrounding environment of the vehicle at the first moment; determining a first expected maneuvering parameter based on a driver model according to the first motion state parameter and the first environmental information; It is determined whether abnormal driving behavior exists based on the first manipulation parameter and the first expected manipulation parameter.

2. The method according to claim 1, wherein The method further comprises: Acquiring a historical manipulation parameter sequence, wherein the historical manipulation parameter sequence includes at least one manipulation parameter for controlling the driving of the vehicle within a first historical time period before the first moment; The driver model is determined according to the historical manipulation parameter sequence.

3. The method according to claim 1 or 2, wherein: The method further comprises: Acquiring a second manipulation parameter, a second motion state parameter, and second environmental information, wherein the second manipulation parameter is a manipulation parameter for controlling the travel of the vehicle at a second moment, the second motion state parameter is used to indicate the motion state of the vehicle at the second moment, and the second environmental information is used to indicate the surrounding environment of the vehicle at the second moment; determining a second expected maneuvering parameter based on the driver model according to the second motion state parameter and the second environmental information; The determining whether there is abnormal driving behavior according to the first manipulation parameter and the first expected manipulation parameter includes: Determine whether abnormal driving behavior exists based on a first manipulation parameter sequence and a first expected manipulation parameter sequence, wherein the first manipulation parameter sequence includes the first manipulation parameter and the second manipulation parameter, and the first expected manipulation parameter sequence includes the first expected manipulation parameter and the second expected manipulation parameter.

4. The method according to claim 3, wherein The method further comprises: determining a first confidence level according to the first manipulation parameter sequence and the first expected manipulation parameter sequence, wherein the first confidence level is used to represent a degree of similarity between the first manipulation parameter sequence and the first expected manipulation parameter sequence; The determining whether there is abnormal driving behavior according to the first manipulation parameter sequence and the first expected manipulation parameter sequence includes: When the first confidence level is greater than or equal to a first threshold, it is determined whether abnormal driving behavior exists according to the first operating parameter sequence and the first expected manipulation parameter sequence.

5. The method according to claim 4, wherein The method further comprises: The driver model is optimized according to the first manipulation parameter sequence when the first confidence level is greater than or equal to a second threshold value, and the second threshold value is less than the first threshold value.

6. The method according to any one of claims 1 to 5, characterized in that The method further comprises: Acquiring first information, where the first information is used to indicate a user manipulation preference; The driver model is determined according to the first information.

7. The method according to claim 6, wherein The method further comprises: determining an evaluation function according to the first information; The determining whether there is abnormal driving behavior according to the first manipulation parameter sequence and the first expected manipulation parameter sequence includes: determining a value of the evaluation function according to the first manipulation parameter sequence and the first expected manipulation parameter sequence; When the value of the evaluation function is greater than or equal to a third threshold, it is determined that abnormal driving behavior exists.

8. The method according to any one of claims 1 to 7, characterized in that The method further comprises: determining the operating condition type of the vehicle according to the first motion state parameter and / or the first motion state parameter; The evaluation function is determined according to the operating condition type of the vehicle.

9. The method according to any one of claims 1 to 8, characterized in that The method further comprises: When it is determined that abnormal driving behavior exists, the user is prompted that the abnormal driving behavior exists.

10. The method according to any one of claims 1 to 9, characterized in that The method further comprises: When it is determined that abnormal driving behavior exists, the vehicle is controlled to be in an automatic driving mode.

11. The method according to any one of claims 1 to 10, characterized in that The transportation tool is a vehicle, and the manipulation parameter includes at least one of a steering wheel angle, an accelerator pedal travel, and a brake pedal travel.

12. A device for identifying abnormal driving behavior, characterized in that: include: an acquisition module, configured to acquire a first manipulation parameter, a first motion state parameter, and first environmental information, wherein the first manipulation parameter is a manipulation parameter for controlling the travel of a vehicle at a first moment, the first motion state parameter is used to indicate the motion state of the vehicle at the first moment, and the first environmental information is used to indicate the surrounding environment of the vehicle at the first moment, and the vehicle is in a manual driving mode; a processing module, configured to determine a first expected maneuvering parameter based on a driver model according to the first motion state parameter and the first environmental information; It is determined whether abnormal driving behavior exists based on the first manipulation parameter and the first expected manipulation parameter.

13. The device according to claim 12, wherein The acquisition module is further configured to: Acquiring a historical operating parameter sequence, wherein the historical operating parameter sequence includes a plurality of sets of operating parameters for controlling the driving of the vehicle within a first historical time period before the first moment; The processing module is further configured to: The driver model is determined according to the historical manipulation parameter sequence.

14. The device according to claim 12 or 13, characterized in that The acquisition module is further configured to: Acquiring a second manipulation parameter, a second motion state parameter, and second environmental information, wherein the second manipulation parameter is a manipulation parameter for controlling the travel of the vehicle at a second moment, the second motion state parameter is used to indicate the motion state of the vehicle at the second moment, and the second environmental information is used to indicate the surrounding environment of the vehicle at the second moment; The processing module is further configured to: determine a second expected manipulation parameter based on the driver model according to the second motion state parameter and the second environmental information; The processing module is specifically used to: Determine whether abnormal driving behavior exists based on a first manipulation parameter sequence and a first expected manipulation parameter sequence, wherein the first manipulation parameter sequence includes the first manipulation parameter and the second manipulation parameter, and the first expected manipulation parameter sequence includes the first expected manipulation parameter and the second expected manipulation parameter.

15. The device according to claim 14, wherein The processing module is further configured to: determining a first confidence level according to the first manipulation parameter sequence and the first expected manipulation parameter sequence, wherein the first confidence level is used to represent a degree of similarity between the first manipulation parameter sequence and the first expected manipulation parameter sequence; The processing module is specifically used to: When the first confidence level is greater than or equal to a first threshold, it is determined whether abnormal driving behavior exists according to the first operating parameter sequence and the first expected manipulation parameter sequence.

16. The device according to claim 15, characterized in that The processing module is further configured to: The driver model is optimized according to the first manipulation parameter sequence when the first confidence level is greater than or equal to a second threshold value, and the second threshold value is less than the first threshold value.

17. The device according to any one of claims 12 to 16, characterized in that The acquisition module is further configured to: Acquiring first information, where the first information is used to indicate a user manipulation preference; The processing module is further configured to determine the driver model based on the first information.

18. The device according to claim 17, wherein The processing module is further configured to: determining an evaluation function according to the first information; The processing module is specifically used to: determining a value of the evaluation function according to the first manipulation parameter sequence and the first expected manipulation parameter sequence; When the value of the evaluation function is greater than or equal to a third threshold, it is determined that abnormal driving behavior exists.

19. The device according to any one of claims 12 to 18, characterized in that The processing module is further configured to: determining the operating condition type of the vehicle according to the first motion state parameter and / or the first motion state parameter; The evaluation function is determined according to the operating condition type of the vehicle.

20. The device according to any one of claims 12 to 19, characterized in that The processing module is further configured to: When it is determined that abnormal driving behavior exists, the user is prompted that the abnormal driving behavior exists.

21. The device according to any one of claims 12 to 20, characterized in that The processing module is further configured to: When it is determined that abnormal driving behavior exists, the vehicle is controlled to be in an automatic driving mode.

22. The device according to any one of claims 12 to 21, characterized in that The transportation tool is a vehicle, and the manipulation parameter includes at least one of a steering wheel angle, an accelerator pedal travel, and a brake pedal travel.

23. A device, characterized in that The method comprises a processor and a memory, wherein the memory is used to store program instructions, and the processor is used to call the program instructions to execute the method according to any one of claims 1 to 11.

24. A means of transport, characterized in that: A device comprising the apparatus of any one of claims 12 to 23.

25. A computer-readable storage medium, characterized in that The computer-readable medium stores a program code, and when the program code is run on a computer, the computer is caused to perform the method according to any one of claims 1 to 11.

26. A computer program product, characterized in that The method comprises a computer program code, which, when running on a computer, causes the computer to perform the method according to any one of claims 1 to 11.

27. A chip, characterized in that: The chip includes a processor and a data interface, and the processor reads instructions stored in a memory through the data interface to execute the method according to any one of claims 1 to 11.

Images