Method for identifying abnormal driving behavior

1 . A method for identifying abnormal driving behavior of a vehicle as the vehicle is driving, comprising:

automatically obtaining current vehicle driving behavior data including vehicle control parameters obtained from an electronic control unit or control system or a sensor of the vehicle;

determining, based on the current vehicle driving behavior data, whether the vehicle is in a state of suspicious abnormal driving behavior;

obtaining current vehicle driving scenario data based on the vehicle being in the state of suspicious abnormal driving behavior;

classifying, using a trained neural network, the current vehicle driving scenario data based on ambient environment information of the vehicle representing a driving scenario in which the vehicle is located, wherein the ambient environment information of the vehicle includes continuous parameters and discrete parameters and discretization processing is performed on the continuous parameters, to determine a current vehicle driving scenario of the vehicle;

determining from a plurality of predetermined algorithm logic types a specific algorithm logic that matches the current vehicle driving scenario of the vehicle; and

determining, based on at least the current vehicle driving behavior data and the specific algorithm logic, whether the suspicious abnormal driving behavior is abnormal driving behavior or normal driving behavior.

2 . The method according to claim 1 , wherein the determining, based on the current vehicle driving behavior data and the specific algorithm logic current vehicle driving scenario, whether the vehicle is in the state of suspicious abnormal driving behavior comprises:

extracting statistical feature values of the current vehicle driving behavior data, and performing cluster analysis on the statistical feature values to obtain the suspicious abnormal driving behavior, wherein the vehicle control parameters include at least one of the following: a vehicle speed, a vehicle acceleration, a vehicle orientation angle, or a lane line deviation value of the vehicle.

3 . The method according to claim 2 , wherein the cluster analysis comprises at least one of the following methods: principal components analysis (PCA), kernel principal components analysis (KPCA), locally linear embedding (LLE), or Laplacian eigenmap (LE).

4 . The method according to claim 2 ,

wherein the classifying the current vehicle driving scenario data to determine the current vehicle driving scenario comprises:

classifying the current vehicle driving scenario data using the trained neural network, to determine the current vehicle driving scenario, wherein the current vehicle driving scenario data comprises at least one of the following: a vehicle information parameter, an other-vehicle information parameter, a traffic signal parameter, a lane line parameter, or a road information parameter; and

wherein the determining, based on at least the vehicle driving behavior data and the current vehicle driving scenario, whether the suspicious abnormal driving behavior is abnormal driving behavior comprises:

determining, based on at least the suspicious abnormal driving behavior and the current driving scenario, whether the suspicious abnormal driving behavior is abnormal driving behavior or normal driving behavior.

5 . The method according to claim 1 , wherein the trained neural network comprises at least one of the following: a convolutional neural network (CNN) or an extreme learning machine.

6 . The method according to claim 1 , wherein the current vehicle driving scenario comprises at least one of the following: intersection deceleration, road section deceleration, or lane line pressing.

7 . The method according to claim 1 , wherein the ambient environment information includes a vehicle information parameter, an other-vehicle information parameter, a traffic signal parameter, a lane line parameter, or a road information parameter.

8 . The method according to claim 1 , wherein the ambient environment information is obtained using a laser radar, a millimeter wave radar, a camera, or an ultrasound radar.

9 . An automated driving assistance system, comprising:

at least one processor; and

a memory coupled to the at least one processor and storing programming instructions for execution by the at least one processor, wherein the programming instructions instruct the at least one processor to perform the following operations:

automatically obtaining current vehicle driving behavior data of a vehicle as the vehicle is driving, the current vehicle driving behavior data including vehicle control parameters obtained from an electronic control unit or control system or a sensor of the vehicle;

obtaining current vehicle driving scenario data; and

determining, based on the current vehicle driving behavior data, whether the vehicle is in a state of suspicious abnormal driving behavior, and based on the vehicle being in the state of suspicious abnormal driving behavior, classifying the current vehicle driving scenario data, using a trained neural network, based on ambient environment information of the vehicle representing a driving scenario in which the vehicle is located to determine a current vehicle driving scenario, wherein the ambient environment information of the vehicle includes continuous parameters and discrete parameters and discretization processing is performed on the continuous parameters, determining from a plurality of predetermined algorithm logic types a specific algorithm logic that matches the current vehicle driving scenario of the vehicle and determining, based on at least the current vehicle driving behavior data and the specific algorithm logic, whether the suspicious abnormal driving behavior is abnormal driving behavior or normal driving behavior.

10 . The system according to claim 9 , wherein the system comprises the electronic control unit (ECU).

11 . The system according to claim 9 , wherein the system comprises at least one of the following: a laser radar, a millimeter wave radar, an ultrasonic radar, or a digital camera.

12 . The system according to claim 9 , wherein the determining, based on the current vehicle driving behavior data, whether the vehicle is in the state of suspicious abnormal driving behavior comprises extracting statistical feature values of the current vehicle driving behavior data, and wherein cluster analysis is performed on the statistical feature values to obtain the suspicious abnormal driving behavior, wherein the vehicle control parameters include at least one of the following: a vehicle speed, a vehicle acceleration, a vehicle orientation angle, or a lane line deviation value of the vehicle.

13 . The system according to claim 12 , wherein the cluster analysis comprises at least one of the following methods: principal components analysis (PCA), kernel principal components analysis (KPCA), locally linear embedding (LLE), or Laplacian eigenmap (LE).

14 . The system according to claim 9 , wherein:

the trained neural network is used to classify the current vehicle driving scenario data, to determine the current vehicle driving scenario, wherein the current vehicle driving scenario data comprises at least one of the following: a vehicle information parameter, an other-vehicle information parameter, a traffic signal parameter, a lane line parameter, or a road information parameter; and

the determining, based on at least the vehicle driving behavior data and the current vehicle driving scenario, whether the suspicious abnormal driving behavior is abnormal driving behavior comprises: determining, based on at least the suspicious abnormal driving behavior and the current driving scenario, whether the suspicious abnormal driving behavior is abnormal driving behavior or normal driving behavior.

15 . The system according to claim 9 , wherein the trained neural network comprises at least one of the following: a convolutional neural network (CNN) or an extreme learning machine (ELM).

16 . The system according to claim 9 , wherein the current vehicle driving scenario comprises at least one of the following: intersection deceleration, road section deceleration, or lane line pressing.

17 . A computer program product comprising computer-executable instructions stored on a non-transitory computer-readable storage medium that, when executed by a processor, cause an apparatus to:

automatically obtain current vehicle driving behavior data of a vehicle as the vehicle is driving, the current vehicle driving behavior data including vehicle control parameters obtained from an electronic control unit or control system or a sensor of the vehicle;

determine, based on the current vehicle driving behavior data, whether a vehicle is in a state of suspicious abnormal driving behavior;

obtain current vehicle driving scenario data based on the vehicle being in the state of suspicious abnormal driving behavior;

classifying, using a trained neural network, the current vehicle driving scenario data based on ambient environment information of the vehicle representing a driving scenario in which the vehicle is located, wherein the ambient environment information of the vehicle includes continuous parameters and discrete parameters and discretization processing is performed on the continuous parameters, to determine a current vehicle driving scenario of the vehicle;

determining from a plurality of predetermined algorithm logic types a specific algorithm logic that matches the current vehicle driving scenario of the vehicle; and

determine, based on at least the current vehicle driving behavior data and the specific algorithm logic, whether the suspicious abnormal driving behavior is abnormal driving behavior or normal driving behavior.

18 . The computer program product of claim 17 , wherein the instructions to determine, based on the current vehicle driving behavior data, whether the vehicle is in the state of suspicious abnormal driving behavior comprises instructions that cause the apparatus to extract statistical feature values of the current vehicle driving behavior data, and perform cluster analysis on the statistical feature values to obtain the suspicious abnormal driving behavior, wherein the vehicle control parameters include at least one of the following: a vehicle speed, a vehicle acceleration, a vehicle orientation angle, or a lane line deviation value of the vehicle.

19 . The computer program product of claim 17 ,

wherein the instructions to classify the current vehicle driving scenario data to determine the current driving scenario comprise instructions to: classify the current vehicle driving scenario data using the trained neural network, to determine the current vehicle driving scenario, wherein the current vehicle driving scenario data comprises at least one of the following: a vehicle information parameter, an other-vehicle information parameter, a traffic signal parameter, a lane line parameter, or a road information parameter; and

wherein the instructions to determine, based on at least the vehicle driving behavior data and the current vehicle driving scenario, whether the suspicious abnormal driving behavior is abnormal driving behavior comprise instructions to: determine, based on at least the suspicious abnormal driving behavior and the current driving scenario, whether the suspicious abnormal driving behavior is abnormal driving behavior or normal driving behavior.