Method and apparatus for exporting driving property index of autonomous vehicle

Abstract

Embodiments of the present disclosure disclose a method and apparatus for exporting a driving property index of an autonomous vehicle. A specific implementation of the method comprises: constructing a driving scenario of an autonomous vehicle by using a preset driving parameter; acquiring a driving state information set of the autonomous vehicle under the driving scenario; fitting a driving curve by using the driving state information set; and acquiring and exporting a driving property index of the autonomous vehicle by analyzing the driving curve. The implementation improves the accuracy of the acquired driving property index of the autonomous vehicle, controls the autonomous vehicle on the basis of the driving property index of high accuracy, and is helpful to realize safe driving of the autonomous vehicle.

What is claimed is: 
     
         1 . A computer-implemented method for exporting a driving property index of an autonomous vehicle, the method comprising:
 constructing a driving scenario of the autonomous vehicle by using a preset driving parameter;   acquiring a driving state information set of the autonomous vehicle under the constructed driving scenario, wherein the driving state information set comprises driving state information of the autonomous vehicle under the constructed driving scenario collected by sensors;   fitting, through a curve fitting method, a mathematical curve used as a driving curve by using the acquired driving state information set of the autonomous vehicle under the constructed driving scenario; and   analyzing the fitted mathematical curve, to acquire the driving property index of the autonomous vehicle from the fitted mathematical curve, and exporting the acquired driving property index of the autonomous vehicle.   
     
     
         2 . The method according to  claim 1 , wherein the fitting a mathematical curve used as a driving curve by using the acquired driving state information set of the autonomous vehicle under the constructed driving scenario, comprising:
 mapping each piece of the driving state information in the acquired driving state information set as a coordinate in a coordinate system, to obtain coordinates in the coordinate system; and   fitting, by using a continuous curve, points indicated by the obtained coordinates in the coordinate system, to obtain the fitted driving curve, wherein the weighted sum of squares of distances from the points indicated by the obtained coordinates to the fitted driving curve is minimized.   
     
     
         3 . The method according to  claim 1 , wherein the driving property index comprises at least one of a steering driving property index, an acceleration driving property index, or a deceleration driving property index. 
     
     
         4 . The method according to  claim 3 , wherein the driving property index comprises the steering driving property index, the driving scenario comprises a scenario with a steering angle of the autonomous vehicle changing from a first steering angle value to a second steering angle value, and the driving state information set comprises driving state information of the autonomous vehicle collected at intervals of every preset steering angle value under the driving scenario. 
     
     
         5 . The method according to  claim 4 , wherein the fitting a driving curve by using the driving state information set comprises:
 for each piece of driving state information in the driving state information set, generating a coordinate corresponding to the driving state information by taking a steering angle value corresponding to the driving state information as a first coordinate value and taking the driving state information as a second coordinate value; and   fitting a driving curve by using points indicated by coordinates corresponding to each piece of driving state information in the driving state information set.   
     
     
         6 . The method according to  claim 5 , wherein the driving curve comprises at least one of a steering time curve, a steering angular velocity curve, or a torque curve; and the steering driving property index comprises at least one of a steering response time delay, a steering duration, a steering overshoot, a steering overshoot time, an average angular velocity, a maximum angular velocity, or a maximum torque. 
     
     
         7 . The method according to  claim 3 , wherein the driving property index comprises the acceleration driving property index, the driving scenario comprises a scenario with an accelerator pedal opening of the autonomous vehicle changing from a first accelerator pedal opening value to a second accelerator pedal opening value, and the driving state information set comprises driving state information of the autonomous vehicle collected at intervals of every preset accelerator pedal opening value under the driving scenario. 
     
     
         8 . The method according to  claim 7 , wherein the fitting a driving curve by using the driving state information set comprises:
 for each piece of driving state information in the driving state information set, generating a coordinate corresponding to the driving state information by taking an accelerator pedal opening value corresponding to the driving state information as a first coordinate value and taking the driving state information as a second coordinate value; and   fitting a driving curve by using points indicated by coordinates corresponding to each piece of driving state information in the driving state information set.   
     
     
         9 . The method according to  claim 8 , wherein the driving curve comprises at least one of an acceleration time curve, an acceleration distance curve, a velocity curve, or an acceleration curve; and the acceleration driving property index comprises at least one of an accelerator pedal response time delay, an acceleration response time delay, an acceleration duration, an average velocity, an average acceleration, a maximum velocity, or a maximum acceleration. 
     
     
         10 . The method according to  claim 3 , wherein the driving property index comprises the deceleration driving property index, the driving scenario comprises a scenario with a brake pedal opening of the autonomous vehicle changing from a first brake pedal opening value to a second brake pedal opening value, and the driving state information set comprises driving state information of the autonomous vehicle collected at intervals of every preset brake pedal opening value under the driving scenario. 
     
     
         11 . The method according to  claim 10 , wherein the fitting a driving curve by using the driving state information set comprises:
 for each piece of driving state information in the driving state information set, generating a coordinate corresponding to the driving state information by taking a brake pedal opening value corresponding to the driving state information as a first coordinate value and taking the driving state information as a second coordinate value; and   fitting a driving curve by using points indicated by coordinates corresponding to each piece of driving state information in the driving state information set.   
     
     
         12 . The method according to  claim 11 , wherein the driving curve comprises at least one of a deceleration time curve, a deceleration distance curve, a velocity curve, or a deceleration curve; and the deceleration driving property index comprises at least one of a brake pedal response time delay, a deceleration response time delay, a deceleration duration, an average velocity, an average deceleration, a maximum velocity, or a maximum deceleration. 
     
     
         13 . The method according to  claim 1 , wherein the driving state information set comprises driving state information of the autonomous vehicle under the constructed driving scenario collected by sensors installed on the autonomous vehicle. 
     
     
         14 . An apparatus for exporting a driving property index of an autonomous vehicle, the apparatus comprising:
 at least one processor; and   a memory storing instructions, the instructions when executed by the at least one processor, cause the at least one processor to perform operations, the operations comprising:   constructing a driving scenario of the autonomous vehicle by using a preset driving parameter;   acquiring a driving state information set of the autonomous vehicle under the constructed driving scenario, wherein the driving state information set comprises driving state information of the autonomous vehicle under the constructed driving scenario collected by sensors;   fitting, through a curve fitting method, a mathematical curve used as a driving curve by using the acquired driving state information set of the autonomous vehicle under the constructed driving scenario; and   analyzing the fitted mathematical curve, to acquire the driving property index of the autonomous vehicle from the fitted mathematical curve, and exporting the acquired driving property index of the autonomous vehicle.   
     
     
         15 . A non-transitory computer readable storage medium storing a computer program, wherein the computer program, when executed by a processor, cause the processor to perform operations, the operation comprising:
 constructing a driving scenario of the autonomous vehicle by using a preset driving parameter;   acquiring a driving state information set of the autonomous vehicle under the constructed driving scenario, wherein the driving state information set comprises driving state information of the autonomous vehicle under the constructed driving scenario collected by sensors;   fitting, through a curve fitting method, a mathematical curve used as a driving curve by using the acquired driving state information set of the autonomous vehicle under the constructed driving scenario; and   analyzing the fitted mathematical curve, to acquire the driving property index of the autonomous vehicle from the fitted driving curve, and exporting the acquired driving property index of the autonomous vehicle.