Method, apparatus and non-transitory computer-readable storage medium for warning of dangerous driving behavior

1 . A method to warn dangerous driving behaviors, the method is applied on an electronic apparatus, the method comprising:

acquiring point cloud data collected by a lidar of a vehicle;

dividing the point cloud data into ground point cloud data and non-ground point cloud data;

identifying ground objects and non-ground objects based on the ground point cloud data and the non-ground point cloud data;

analyzing driving behaviors of a driver of the vehicle based on movement changes of the vehicle relative to the ground objects and the non-ground objects;

determining whether the driving behaviors of the driver comprises at least one dangerous driving behavior; and

outputting a safety warning when the driving behaviors of the driver is determined to comprise the dangerous driving behavior.

2 . The method of claim 1 , wherein the at least one dangerous driving behavior comprises driving in reverse and failing to maintain a safe distance.

3 . The method of claim 1 , wherein the dividing the point cloud data into ground point cloud data and non-ground point cloud data comprises:

converting the point cloud data from a point cloud coordinate system to a coordinate system of the vehicle;

classifying point data in the point cloud data with height values less than a preset height threshold as the ground point cloud data; and

classifying point data in the point cloud data with height values greater or equal to the preset height threshold as the non-ground point cloud data.

4 . The method of claim 1 , further comprising:

converting the non-ground point cloud data into a bird's-eye view representation;

dividing the bird's-eye view representation into equally sized unit grids, with each unit grid corresponding to an area above a ground surface;

clustering gridded non-ground point cloud data using a clustering algorithm to obtain shapes of the non-ground objects;

extracting point data closest to the vehicle from each non-ground object as two-dimensional feature of each non-ground object according to a distance between each non-ground object and the vehicle; and

tracking each non-ground object whose distance relative to the vehicle is less than or equal to a safe braking distance of the vehicle according to the two-dimensional feature of each non-ground object.

5 . The method of claim 2 , wherein the determining whether the driving behaviors of the driver comprises at least one dangerous driving behavior comprises:

analyzing whether directions of lanes of the ground objects are opposite to the driving direction of the vehicle;

determining that the driving behavior is the driving in reverse among the at least one dangerous driving behavior when the directions of lanes of the ground objects are opposite to the driving direction of the vehicle.

6 . The method of claim 2 , wherein the determining whether the determining whether the driving behaviors of the driver comprises at least one dangerous driving behavior comprises:

analyzing whether driving directions of a plurality of other vehicles around the vehicle in the adjacent lane are opposite to the driving direction of the vehicle when there is a non-double solid line in a lane of the ground objects; and

determining that the driving behavior is the driving in reverse among the dangerous driving behavior when the driving directions of the plurality of the other vehicles are opposite to the driving direction of the vehicle.

7 . The method of claim 2 , wherein the determining whether the driving behaviors of the driver comprises at least one dangerous driving behavior comprises:

analyzing whether distances between the vehicle and the other vehicles around the vehicle are greater than or equal to a safe braking distance when there are other vehicles in front of the non-ground objects;

determining that the driving behavior is the failing to maintain the safe distance among the dangerous driving behavior when the distances between the vehicle and the other vehicles are not greater than or equal to the safe braking distance.

8 . The method of claim 1 further comprising:

determining whether the driver has a record of the at least one dangerous driving behavior in a history of driving of the vehicle;

when the driver is determined to has the record of the at least one of the dangerous driving behavior, determining a road rage level corresponding to each of the at least one dangerous driving behavior, wherein the road rage level indicates a possibility that the driver has a road rage; and

predicting a road rage state of the driver based on the road rage levels corresponding to the at least one dangerous driving behaviors.

9 . An apparatus configured for warning of dangerous driving behavior, the apparatus comprising:

a non-transitory memory storage storing processor-executable instructions; and

at least one processor coupled to the memory to receive the processor-executable instructions, wherein, upon execution of the processor executable instructions, the at least one processor:

acquiring point cloud data collected by a lidar of a vehicle;

dividing the point cloud data into ground point cloud data and non-ground point cloud data;

identifying ground objects and non-ground objects based on the ground point cloud data and the non-ground point cloud data;

analyzing driving behaviors of a driver of the vehicle based on movement changes of the vehicle relative to the ground objects and the non-ground objects;

determining whether the driving behaviors of the driver comprises at least one dangerous driving behavior; and

outputting a safety warning when the driving behaviors of the driver is determined to comprise the dangerous driving behavior.

10 . A non-transitory computer readable storage medium storing processor-executable instructions which, when executed by at least one processor, cause the at least one processor to perform a method for warning of dangerous driving behavior, the method comprising:

acquiring point cloud data collected by a lidar of a vehicle;

dividing the point cloud data into ground point cloud data and non-ground point cloud data;

identifying ground objects and non-ground objects based on the ground point cloud data and the non-ground point cloud data;

analyzing driving behaviors of a driver of the vehicle based on movement changes of the vehicle relative to the ground objects and the non-ground objects;

determining whether the driving behaviors of the driver comprises at least one dangerous driving behavior; and

outputting a safety warning when the driving behaviors of the driver is determined to comprise the dangerous driving behavior.