Method and apparatus for detecting complexity of traveling scenario of vehicle
View Patent ↗A method for detecting a complexity of a traveling scenario of a vehicle includes obtaining a travelling speed of the vehicle and a travelling speed of a target vehicle, determining, based on the traveling speed of the vehicle and the traveling speed of the target vehicle, a dynamic complexity of a traveling scenario in which the vehicle is located, determining static information of each static factor in the traveling scenario in which the vehicle is currently located, obtaining, based on the static information of each static factor, a static complexity of the traveling scenario in which the vehicle is located, and obtaining, based on the dynamic complexity and the static complexity, a comprehensive complexity of the traveling scenario in which the vehicle is located.
1 . A method comprising:
obtaining, using a sensor, a first traveling speed of a vehicle and a second traveling speed of a target vehicle, wherein the target vehicle meets a preset distance condition with the vehicle;
obtaining an included angle (θ ij ) between a first traveling direction and a relative traveling speed between the vehicle and the target vehicle based on the first traveling speed and the second traveling speed;
obtaining an initial complexity (f(θ ij )) corresponding to the target vehicle by substituting the included angle in the following equation:
f
(
θ
i
j
)
=
1
2
{
1
-
cos
[
(
1
8
0
6
7
.
5
)
·
2
θ
ij
3
·
π
1
8
0
+
π
1
0
]
}
;
obtaining, based on the initial complexity, a dynamic complexity of a traveling scenario of the vehicle;
obtaining, from a positioning system and a high-definition map, static information of each static factor in the traveling scenario;
obtaining, based on the static information, a static complexity of the traveling scenario;
obtaining, based on the dynamic complexity and the static complexity, a comprehensive complexity of the traveling scenario; and
controlling, by a controller of the vehicle, a traveling decision of the vehicle based on the comprehensive complexity.
2 . The method of claim 1 , wherein obtaining the dynamic complexity comprises:
obtaining, for each of a plurality of target vehicles and based on the first traveling speed and third traveling speeds of the target vehicles, a complexity corresponding to the target vehicle, wherein the target vehicles comprise the target vehicle, and wherein the third traveling speeds comprise the second traveling speed; and
adding complexities corresponding to the target vehicles to obtain the dynamic complexity.
3 . The method of claim 2 , wherein obtaining the complexity corresponding to the target vehicle comprises correcting the initial complexity based on a distance between the target vehicle and the vehicle and a relative speed between the target vehicle and the vehicle to obtain the complexity corresponding to the target vehicle.
4 . The method of claim 1 , wherein the static factor comprises a plurality of static factors, and wherein obtaining the static complexity comprises:
obtaining values of the static factors based on static information of the static factors;
obtaining complexities respectively corresponding to the static factors based on a correspondence between the values of the static factors and the complexities; and
obtaining, based on the complexities, the static complexity.
5 . The method of claim 4 , wherein obtaining the static complexity comprises:
obtaining weights respectively corresponding to the static factors;
respectively multiplying the complexities corresponding to the values of the static factors by the weights to obtain weighted complexities respectively corresponding to the static factors; and
adding the weighted complexities to obtain the static complexity.
6 . The method of claim 5 , further comprising:
obtaining, for a first value of a first static factor, N sample images corresponding to the first value, wherein N is an integer greater than 1;
separately performing image recognition on the N sample images to obtain a predicted value corresponding to each sample image;
counting a quantity M of sample images whose corresponding predicted values are different from corresponding ground truth values, wherein a ground truth value uniquely identifies the first value; and
obtaining a ratio of M to N as a complexity corresponding to the first value.
7 . The method of claim 1 , wherein the static factor comprises at least one of a road type, a quantity of co-directional lanes, a lane width, a central isolation form, an isolation form between a motor vehicle and a non-motor vehicle, a traffic sign, or a traffic light.
8 . The method of claim 1 , further comprising:
obtaining environment information of an environment factor in the traveling scenario;
obtaining a value of the environment factor based on the environment information; and
obtaining, based on a stored correspondence between the value of the environment factor and a complexity correction coefficient, a target complexity correction coefficient corresponding to the environment factor, wherein obtaining the comprehensive complexity comprises:
obtaining a value by adding the dynamic complexity and the static complexity; and
multiplying the value by the target complexity correction coefficient to obtain the comprehensive complexity.
9 . The method of claim 8 , wherein the environment factor comprises at least one of a lighting condition, a weather condition, or a road surface condition.
10 . The method of claim 1 , further comprising:
when there is a first forward vehicle in a same lane as the vehicle:
determining, as a reference vehicle, a second forward vehicle that is in the first forward vehicle in the same lane and that has a minimum distance from the vehicle;
obtaining a third traveling speed of the reference vehicle and a first component of a first distance between the reference vehicle and the vehicle;
obtaining a first safe distance between the vehicle and the reference vehicle based on the third traveling speed, a preset maximum acceleration, a preset minimum deceleration of the vehicle, a preset maximum deceleration of the reference vehicle, and a preset driver reaction time;
obtaining a smaller value in the first safe distance and the first component; and
using, as the target vehicle, the second forward vehicle in the same lane and an adjacent lane, wherein the first component is not greater than the smaller value; and
when there is no first forward vehicle in the same lane:
obtaining a second safe distance between the vehicle and a third forward vehicle based on the third traveling speed, a second preset maximum acceleration, a second preset minimum deceleration of the vehicle, a second preset forward-vehicle traveling speed, the preset maximum deceleration, and the preset driver reaction time, wherein the second preset forward-vehicle traveling speed is 0; and
using, as the target vehicle, the third forward vehicle in the same lane in and the adjacent lane, wherein a third component of a distance between the third forward vehicle and the vehicle is not greater than the second safe distance.
11 . A controller comprising:
a memory configured to store instructions; and
a processor coupled to the memory and configured to:
obtain, using a sensor, a first traveling speed of a vehicle and a second traveling speed of a target vehicle, wherein the target vehicle meets a preset distance condition with the vehicle;
obtain an included angle (θ ij ) between a first traveling direction and a relative traveling speed between the vehicle and the target vehicle based on the first traveling speed and the second traveling speed;
obtain an initial complexity (f(θ ij )) corresponding to the target vehicle by substituting the included angle in the following equation:
f
(
θ
i
j
)
=
1
2
{
1
-
cos
[
(
1
8
0
6
7
.
5
)
·
2
θ
ij
3
·
π
1
8
0
+
π
1
0
]
}
;
obtain, based on the initial complexity, a dynamic complexity of a traveling scenario of the vehicle;
obtain, from a positioning system and a high-definition map, static information of each static factor in the traveling scenario;
obtain, based on the static information, a static complexity of the traveling scenario;
obtain, based on the dynamic complexity and the static complexity, a comprehensive complexity of the traveling scenario; and
control a traveling decision of the vehicle based on the comprehensive complexity.
12 . The controller of claim 11 , wherein the processor is further configured to:
obtain, for each of a plurality of target vehicles and based on the first traveling speed and third traveling speeds of the target vehicles, a complexity corresponding to the target vehicle, wherein the target vehicles comprise the target vehicle, and wherein the third traveling speeds comprise the second traveling speed; and
add complexities corresponding to the target vehicles to obtain the dynamic complexity of the traveling scenario.
13 . The controller of claim 11 , wherein the static factor comprises a plurality of static factors, and wherein the processor is further configured to:
obtain values of the static factors based on static information of the static factors;
obtain complexities respectively corresponding to the static factors based on a correspondence between the values of the static factors and the complexities; and
obtain, based on the complexities, the static complexity.
14 . The controller of claim 13 , wherein the processor is further configured to:
obtain weights respectively corresponding to the static factors;
respectively multiply the static complexities corresponding to the values of the static factors by the weights to obtain weighted complexities respectively corresponding to the static factors; and
add the weighted complexities to obtain the static complexity.
15 . The controller of claim 13 , wherein the processor is further configured to:
obtain, for a first value of a first static factor, N sample images corresponding to the first value, wherein N is an integer greater than 1;
separately perform image recognition on the N sample images to obtain a predicted value corresponding to each sample image;
count a quantity M of sample images whose corresponding predicted values are different from corresponding ground truth values, wherein a ground truth value uniquely identifies the first value; and
obtain a ratio of M to N as a third complexity corresponding to the first value.
16 . The controller of claim 11 , wherein the static factor comprises at least one of a road type, a quantity of co-directional lanes, a lane width, a central isolation form, an isolation form between a motor vehicle and a non-motor vehicle, a traffic sign, or a traffic light.
17 . The controller of claim 11 , wherein the processor is further configured to:
obtain environment information of an environment factor in the traveling scenario;
obtain a value of the environment factor based on the environment information;
obtain, based on a stored correspondence between the value of the environment factor and a complexity correction coefficient, a target complexity correction coefficient corresponding to the value of the environment factor;
obtain a value by adding the dynamic complexity and the static complexity; and
multiply the value by the target complexity correction to obtain the comprehensive complexity.
18 . The controller of claim 17 , wherein the environment factor comprises at least one of a lighting condition, a weather condition, or a road surface condition.
19 . The controller of claim 11 , wherein the processor is further configured to:
when there is a first forward vehicle in a same lane as the vehicle:
determine, as a reference vehicle, a second forward vehicle that is in the first forward vehicle in the same lane and that has a minimum distance from the vehicle;
obtain a third traveling speed of the reference vehicle and a first component of a first distance between the reference vehicle and the vehicle;
obtain a first safe distance between the vehicle and the reference vehicle based on the third traveling speed, a preset maximum acceleration, a preset minimum deceleration of the vehicle, a preset maximum deceleration of the reference vehicle, and a preset driver reaction time;
obtain a smaller value in the first safe distance and the first component; and
use, as the target vehicle, the second forward vehicle in the same lane and an adjacent lane, wherein the first component is not greater than the smaller value; and
when there is no first forward vehicle in the same lane:
obtain a second safe distance between the vehicle and a third forward vehicle based on the third traveling speed, a second preset maximum acceleration, a second preset minimum deceleration of the vehicle, a second preset forward-vehicle traveling speed, the preset maximum deceleration, and the preset driver reaction time, wherein the second preset forward-vehicle traveling speed is 0; and
use, as the target vehicle, the third forward vehicle in the same lane and the adjacent lane, wherein a third component of a distance between the forward vehicle and the vehicle is not greater than the second safe distance.
20 . A computer program product comprising computer-executable instructions that are stored on a non-transitory computer-readable storage medium and that, when executed by a processor of a controller, cause the controller to:
obtain, using a sensor, a first traveling speed of a vehicle and a second traveling speed of a target vehicle, wherein the target vehicle meets a preset distance condition with the vehicle;
obtain an included angle (θ ij ) between a first traveling direction and a relative traveling speed between the vehicle and the target vehicle based on the first traveling speed and the second traveling speed;
obtain an initial complexity ((f(θ ij )) corresponding to the target vehicle by substituting the included angle in the following equation:
f
(
θ
i
j
)
=
1
2
{
1
-
cos
[
(
1
8
0
6
7
.
5
)
·
2
θ
ij
3
·
π
1
8
0
+
π
1
0
]
}
;
obtain, based on the initial complexity, a dynamic complexity of a traveling scenario of the vehicle;
obtain, from a positioning system and a high-definition map, static information of each static factor in the traveling scenario;
obtain, based on the static information, a static complexity of the traveling scenario;
obtain, based on the dynamic complexity and the static complexity, a comprehensive complexity of the traveling scenario; and
control a traveling decision of the vehicle based on the comprehensive complexity.