Granted Patent B2

US 12,393,746 · App. 17/371,900 · Granted Aug 19, 2025

Inventor: Haining Du (Shenzhen, CN)

Assignee: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED

G06F30/15G06F30/10G06F30/20

Abstract

Claims (102)

1. A driving simulation method for improving an autonomous driving system configured for controlling an autonomous vehicle, performed by an electronic device, the method comprising:

receiving parameters through a first configuration interface;

determining, based on the parameters, a reference vehicle-to-vehicle distance, a reference vehicle speed, and a background vehicle quantity of a target lane, the target lane comprising a lane in which a test vehicle and background vehicles are located;

determining an initial location of each of the background vehicles in the target lane based on a location of the test vehicle, the reference vehicle-to-vehicle distance, and a random distance deviation;

determining an initial speed of each of the background vehicles in the target lane based on the reference vehicle speed and a random speed deviation; and

simulating traveling of each of the background vehicles in a simulated traffic environment based on the initial location and the initial speed of each of the background vehicles and the background vehicle quantity, further comprising:

fixing a speed of a leading background vehicle;

updating speeds of other background vehicles following the leading background vehicle according to predefined speed constraints in a following model, such that the background vehicles all stay within an effective test range of the test vehicle;

controlling the test vehicle in the simulated traffic environment to change a lane in accordance with a decision algorithm of the autonomous driving system:

in response to the test vehicle changing the lane, updating transverse speeds and locations of the background vehicles in the simulated traffic environment;

recording driving data of the test vehicle until a predefined simulation time is achieved; and

generating a test report based on the driving data of the test vehicle, wherein the test report represents performance of the autonomous driving system.

2. The method according to claim 1 , wherein

the first configuration interface comprises a configuration control configured to configure a traffic state indication parameter, and a configuration control configured to configure a background vehicle quantity indication parameter;

the receiving parameters inputted through the first configuration interface comprises:

receiving a target traffic state indication parameter and a target background vehicle quantity indication parameter inputted through the configuration controls on the first configuration interface; and

the determining, based on the parameters, a reference vehicle-to-vehicle distance, a reference vehicle speed, and a background vehicle quantity of a target lane comprises:

determining the reference vehicle-to-vehicle distance and the reference vehicle speed based on a target traffic flow macroscopic fundamental diagram and the target traffic state indication parameter; and

determining the background vehicle quantity of the target lane based on the target background vehicle quantity indication parameter of the target lane.

3. The method according to claim 2 , wherein the target traffic state indication parameter comprises a target traffic density; and

the determining the reference vehicle-to-vehicle distance and the reference vehicle speed based on a target traffic flow macroscopic fundamental diagram and the target traffic state indication parameter comprises:

using an inverse of the target traffic density as the reference vehicle-to-vehicle distance; and

determining a target traffic state corresponding to the target traffic density from the target traffic flow macroscopic fundamental diagram, determining a target traffic capacity that the target traffic state has in the target traffic flow macroscopic fundamental diagram, and using a ratio of the target traffic capacity to the target traffic density as the reference vehicle speed.

4. The method according to claim 3 , wherein

the target traffic flow macroscopic fundamental diagram comprises a first linear segment and a second linear segment located in a coordinate system, a horizontal axis of the coordinate system representing a traffic density, a vertical axis of the coordinate system representing a traffic capacity, and each point in the first linear segment and the second linear segment representing a traffic state;

the determining a target traffic state corresponding to the target traffic density from the target traffic flow macroscopic fundamental diagram comprises:

establishing the following constraint relationships between the first linear segment and the second linear segment in the target traffic flow macroscopic fundamental diagram:

an ordinate of an intersection point of the first linear segment and the second linear segment is used as a maximum traffic capacity, an abscissa of the intersection point of the first linear segment and the second linear segment is used as a critical traffic density, and an abscissa of an intersection point of the second linear segment and the horizontal axis is used as a traffic jam density; and

using, as the target traffic state, a point, in the first linear segment and the second linear segment that conform to the constraint relationships, whose abscissa is the target traffic density; and

the determining a target traffic capacity that the target traffic state has in the target traffic flow macroscopic fundamental diagram comprises:

using an ordinate of the target traffic state as the target traffic capacity that the target traffic state has in the target traffic flow macroscopic fundamental diagram.

5. The method according to claim 2 , wherein the target traffic state indication parameter comprises a target traffic capacity and a target vehicle speed; and

the determining the reference vehicle-to-vehicle distance and the reference vehicle speed based on a target traffic flow macroscopic fundamental diagram and the target traffic state indication parameter comprises:

determining a target traffic state corresponding to the target traffic capacity and a target vehicle speed from the target traffic flow macroscopic fundamental diagram, determining a target traffic density that the target traffic state has in the target traffic flow macroscopic fundamental diagram, and using an inverse of the target traffic density as the reference vehicle-to-vehicle distance; and

determining the target vehicle speed as the reference vehicle speed.

6. The method according to claim 5 , wherein

the target traffic flow macroscopic fundamental diagram comprises a coordinate system and a first linear segment and a second linear segment located in the coordinate system, a horizontal axis of the coordinate system representing a traffic density, a vertical axis of the coordinate system representing a traffic capacity, and each point in the first linear segment and the second linear segment representing a traffic state;

the determining a target traffic state corresponding to the target traffic capacity and a target vehicle speed from the target traffic flow macroscopic fundamental diagram comprises:

establishing the following constraint relationships between the first linear segment and the second linear segment in the target traffic flow macroscopic fundamental diagram:

an ordinate of an intersection point of the first linear segment and the second linear segment is used as a maximum traffic capacity, an abscissa of the intersection point of the first linear segment and the second linear segment is used as a critical traffic density, and an abscissa of an intersection point of the second linear segment and the horizontal axis is used as a traffic jam density; and

determining, as the target traffic state, a point, in the first linear segment and the second linear segment that conform to the constraint relationships, whose ordinate is the target traffic capacity and which is connected to an origin of the coordinate system to form a line with a slope that is the target vehicle speed; and

the determining a target traffic density that the target traffic state has in the target traffic flow macroscopic fundamental diagram comprises:

using an abscissa of the target traffic state as the target traffic density that the target traffic state has in the target traffic flow macroscopic fundamental diagram.

7. The method according to claim 3 , further comprising:

displaying a second configuration interface, and receiving traffic flow quantification parameters inputted through the second configuration interface, the traffic flow quantification parameters comprising a free-flow vehicle speed, a jam density, a critical density, and a maximum traffic capacity;

generating, according to the free-flow vehicle speed, the critical density, and the maximum traffic capacity, the first linear segment configured for representing a free traveling state of the vehicles;

generating, according to the critical density, the maximum traffic capacity, and the jam density, the second linear segment configured for representing a jammed traveling state of the vehicles; and

constructing the target traffic flow macroscopic fundamental diagram according to the coordinate system and the first linear segment and the second linear segment located in the coordinate system.

8. The method according to claim 2 , wherein the target background vehicle quantity indication parameter comprises a background vehicle quantity parameter used for indicating the background vehicle quantity of the target lane.

9. The method according to claim 2 , wherein the target background vehicle quantity indication parameter comprises a location range indication parameter; and

the determining the background vehicle quantity of the target lane based on the target background vehicle quantity indication parameter of the target lane comprises:

determining the background vehicle quantity of the target lane according to the location range indication parameter of target background vehicles and the reference vehicle-to-vehicle distance.

10. The method according to claim 9 , wherein the location range indication parameter of the target background vehicles comprises a first distance parameter and a second distance parameter, the first distance parameter being used for indicating a maximum distance between the test vehicle and a background vehicle in a forward direction of the test vehicle, and the second distance parameter being used for indicating a maximum distance between the test vehicle and a background vehicle in a reverse direction of the test vehicle; and

the determining the background vehicle quantity of the target lane according to the location range indication parameter of target background vehicles and the reference vehicle-to-vehicle distance comprises:

determining a ratio of the first distance parameter to the reference vehicle-to-vehicle distance, and using the ratio as a quantity of the background vehicles in the target lane in front of the test vehicle; and

determining a ratio of the second distance parameter to the reference vehicle-to-vehicle distance, and using the ratio as a quantity of the background vehicles in the target lane in rear of the test vehicle.

11. The method according to claim 1 , wherein the first configuration interface comprises a configuration control configured to configure the reference vehicle-to-vehicle distance, a configuration control configured to configure the reference vehicle speed, and a configuration control configured to configure the background vehicle quantity of the target lane; and

the receiving parameters inputted through the first configuration interface comprises:

receiving parameters inputted through the configuration controls on the first configuration interface, the parameters comprising the reference vehicle-to-vehicle distance, the reference vehicle speed, and the background vehicle quantity of the target lane.

12. The method according to claim 1 , wherein when determining the initial location of an i th background vehicle in the target lane, the random distance deviation is a normal random variable conforming to a first normal distribution, and when determining the initial speed of the i th background vehicle in the target lane, the random speed deviation is a normal random variable conforming to a second normal distribution, a value for i being in a range of 1 to N, N being a total quantity of the background vehicles.

13. An electronic device, comprising a processor and a memory,

the memory being configured to store a plurality of computer programs that, when executed by the processor, cause the electronic device to perform a driving simulation method for improving an autonomous driving system configured for controlling an autonomous vehicle, the driving simulation method including a plurality of operations including:

receiving parameters through a first configuration interface;

determining, based on the parameters, a reference vehicle-to-vehicle distance, a reference vehicle speed, and a background vehicle quantity of a target lane, the target lane comprising a lane in which a test vehicle and background vehicles are located;

determining an initial location of each of the background vehicles in the target lane based on a location of the test vehicle, the reference vehicle-to-vehicle distance, and a random distance deviation;

determining an initial speed of each of the background vehicles in the target lane based on the reference vehicle speed and a random speed deviation; and

simulating traveling of each of the background vehicles in a simulated traffic environment based on the initial location and the initial speed of each of the background vehicles and the background vehicle quantity, further comprising:

fixing a speed of a leading background vehicle;

updating speeds of other background vehicles following the leading background vehicle according to predefined speed constraints in a following model, such that the background vehicles all stay within an effective test range of the test vehicle;

controlling the test vehicle in the simulated traffic environment to change a lane in accordance with a decision algorithm of the autonomous driving system;

in response to the test vehicle changing the lane, updating transverse speeds and locations of the background vehicles in the simulated traffic environment:

recording driving data of the test vehicle until a predefined simulation time is achieved; and

generating a test report based on the driving data of the test vehicle, wherein the test report represents performance of the autonomous driving system.

14. The electronic device according to claim 13 , wherein

the first configuration interface comprises a configuration control configured to configure a traffic state indication parameter, and a configuration control configured to configure a background vehicle quantity indication parameter;

the receiving parameters inputted through the first configuration interface comprises:

receiving a target traffic state indication parameter and a target background vehicle quantity indication parameter inputted through the configuration controls on the first configuration interface; and

the determining, based on the parameters, a reference vehicle-to-vehicle distance, a reference vehicle speed, and a background vehicle quantity of a target lane comprises:

determining the reference vehicle-to-vehicle distance and the reference vehicle speed based on a target traffic flow macroscopic fundamental diagram and the target traffic state indication parameter; and

determining the background vehicle quantity of the target lane based on the target background vehicle quantity indication parameter of the target lane.

15. The electronic device according to claim 14 , wherein the target traffic state indication parameter comprises a target traffic density; and

the determining the reference vehicle-to-vehicle distance and the reference vehicle speed based on a target traffic flow macroscopic fundamental diagram and the target traffic state indication parameter comprises:

using an inverse of the target traffic density as the reference vehicle-to-vehicle distance; and

determining a target traffic state corresponding to the target traffic density from the target traffic flow macroscopic fundamental diagram, determining a target traffic capacity that the target traffic state has in the target traffic flow macroscopic fundamental diagram, and using a ratio of the target traffic capacity to the target traffic density as the reference vehicle speed.

16. The electronic device according to claim 14 , wherein the target traffic state indication parameter comprises a target traffic capacity and a target vehicle speed; and

the determining the reference vehicle-to-vehicle distance and the reference vehicle speed based on a target traffic flow macroscopic fundamental diagram and the target traffic state indication parameter comprises:

determining a target traffic state corresponding to the target traffic capacity and a target vehicle speed from the target traffic flow macroscopic fundamental diagram, determining a target traffic density that the target traffic state has in the target traffic flow macroscopic fundamental diagram, and using an inverse of the target traffic density as the reference vehicle-to-vehicle distance; and

determining the target vehicle speed as the reference vehicle speed.

17. The electronic device according to claim 13 , wherein when determining the initial location of an i th background vehicle in the target lane, the random distance deviation is a normal random variable conforming to a first normal distribution, and when determining the initial speed of the i th background vehicle in the target lane, the random speed deviation is a normal random variable conforming to a second normal distribution, a value for i being in a range of 1 to N, N being a total quantity of the background vehicles.

18. A non-transitory computer-readable storage medium, storing a plurality of computer programs, the plurality of computer programs, when executed by a processor of an electronic device, cause the electronic device to perform a driving simulation method for improving an autonomous driving system configured for controlling an autonomous vehicle, the driving simulation method including a plurality of operations including:

receiving parameters through a first configuration interface;

determining, based on the parameters, a reference vehicle-to-vehicle distance, a reference vehicle speed, and a background vehicle quantity of a target lane, the target lane comprising a lane in which a test vehicle and background vehicles are located;

determining an initial location of each of the background vehicles in the target lane based on a location of the test vehicle, the reference vehicle-to-vehicle distance, and a random distance deviation;

determining an initial speed of each of the background vehicles in the target lane based on the reference vehicle speed and a random speed deviation; and

simulating traveling of each of the background vehicles in a simulated traffic environment based on the initial location and the initial speed of each of the background vehicles and the background vehicle quantity, further comprising:

fixing a speed of a leading background vehicle;

updating speeds of other background vehicles following the leading background vehicle according to predefined speed constraints in a following model, such that the background vehicles all stay within an effective test range of the test vehicle;

controlling the test vehicle in the simulated traffic environment to change a lane in accordance with a decision algorithm of the autonomous driving system;

in response to the test vehicle changing the lane, updating transverse speeds and locations of the background vehicles in the simulated traffic environment:

recording driving data of the test vehicle until a predefined simulation time is achieved; and

generating a test report based on the driving data of the test vehicle, wherein the test report represents performance of the autonomous driving system.

19. The non-transitory computer-readable storage medium according to claim 18 , wherein when determining the initial location of an i th background vehicle in the target lane, the random distance deviation is a normal random variable conforming to a first normal distribution, and when determining the initial speed of the i th background vehicle in the target lane, the random speed deviation is a normal random variable conforming to a second normal distribution, a value for i being in a range of 1 to N, N being a total quantity of the background vehicles.

Assignments (1)

Priority Claims (1)

Continuity (2)

References Cited (26)