Vehicle control method, apparatus, and device

1. A method, comprising:

obtaining driving status information of a self-driving vehicle and first driving environment information of the self-driving vehicle when the self-driving vehicle is in a self-driving status, wherein the driving status information comprises a current speed of the self-driving vehicle;

obtaining a control instruction decision model through training based on training driving status information of the self-driving vehicle in a training status and training driving environment information of the self-driving vehicle in the training status;

calculating a coarse-grained control instruction of the self-driving vehicle based on the driving status information, the first driving environment information of a target vehicle and the self-driving vehicle, and the control instruction decision model, wherein the coarse-grained control instruction controls a driving manner of the self-driving vehicle, and wherein the coarse-grained control instruction comprises straight driving in a lane that is currently being used by the self-driving vehicle;

executing, based on second driving environment information, the coarse-grained control instruction when a preset distance of the self-driving vehicle to the target vehicle is greater than a secure distance, wherein the secure distance is between the target vehicle and the self-driving vehicle that does not cause the self-driving vehicle that is driving at a first current speed to come into contact with the target vehicle that is driving at a second current speed, and wherein the second driving environment information comprises driving environment information corresponding to the coarse-grained control instruction and when the self-driving vehicle is in the self-driving status;

obtaining, based on lane information of the self-driving vehicle and the driving status information of the self-driving vehicle, a fine-grained control instruction when executing the coarse-grained control instruction, wherein the fine-grained control instruction controls the current speed of the self-driving vehicle to be less than a preset expected speed that is defined for the lane that is currently being used by the self-driving vehicle, wherein the preset expected speed is less than a maximum speed that is defined for the lane, wherein the fine-grained control instruction comprises a value of a throttle of the self-driving vehicle, and wherein the lane information comprises first information about the lane of the self-driving vehicle that corresponds to the coarse-grained control instruction and that is on a road that the self-driving vehicle drives;

calculating the value of the throttle based on a first preset formula when the first current speed is not greater than the preset expected speed, and wherein the first preset formula comprises value of a throttle=preset throttle control coefficient×(preset expected speed−current speed+preset value);

executing the fine-grained control instruction; and

automatically controlling the self-driving vehicle in response to executing the fine-grained control instruction.

2. The method of claim 1 , further comprising obtaining the fine-grained control instruction based on the preset expected speed and the current speed.

3. The method of claim 1 , wherein the preset throttle control coefficient is based on a brand and a model of the self-driving vehicle.

4. The method of claim 1 , wherein the preset throttle control coefficient is 0.2.

5. The method of claim 1 , wherein the preset throttle control coefficient is based on a performance of the self-driving vehicle.

6. The method of claim 1 , wherein the preset value is 1.

7. An apparatus, comprising:

a memory storing instructions; and

a processor coupled to the memory and configured to execute the instructions that cause the apparatus to be configured to:

obtain driving status information of a self-driving vehicle and first driving environment information of the self-driving vehicle when the self-driving vehicle is in a self-driving status, wherein the driving status comprises a current speed of the self-driving vehicle;

obtain a control instruction decision model through training based on training driving status information of the self-driving vehicle in a training status and training driving environment information of the self-driving vehicle in the training status;

calculate a coarse-grained control instruction of the self-driving vehicle based on the driving status information, the first driving environment information of a target vehicle in relation to and the self-driving vehicle, and the control instruction decision model, wherein the coarse-grained control instruction controls a driving manner of the self-driving vehicle, and wherein the coarse-grained control instruction comprises straight driving and in a lane that is currently being used by the self-driving vehicle;

executed, based on second driving environment information, the coarse-grained control instruction when a preset distance of the self-driving vehicle to the target vehicle is greater than a secure distance, wherein the secure distance is between the target vehicle and the self-driving vehicle that does not cause the self-driving vehicle that is driving at a first current speed to come into contact with the target vehicle that is driving at a second current speed, and wherein the second driving environment information comprises driving environment information corresponding to the coarse-grained control instruction and when the self-driving vehicle is in the self-driving status;

obtain, based on lane information of the self-driving vehicle and the driving status information of the self-driving vehicle, the coarse-grained control instruction to control a parameter of the self-driving vehicle associated with the coarse-grained control instruction when executing the coarse-grained control instruction, wherein a fine-grained control instruction controls the current speed of the self-driving vehicle to be less than a preset expected speed that is defined for the lane that is currently being used by the self-driving vehicle, wherein the preset expected speed is less than a maximum speed that is defined for the lane, wherein the fine-grained control instruction comprises a value of a throttle of the self-driving vehicle, and wherein the lane information comprises first information about the lane of the self-driving vehicle that corresponds to the coarse-grained control instruction that is on a road the self-driving vehicle drives;

calculate the value of the throttle based on a first preset formula when the first current speed is not greater than the preset expected speed, and wherein the first preset formula comprises value of a throttle=preset throttle control coefficient×(preset expected speed−current speed+preset value);

execute the fine-grained control instruction; and

automatically control the self-driving vehicle in response to executing the fine-grained control instruction.

8. The apparatus of claim 7 , wherein the instructions further cause the apparatus to be configured to obtain the fine-grained control instruction based on the preset expected speed and the current speed.

9. The apparatus of claim 7 , wherein the preset throttle control coefficient is based on a brand and a model of the self-driving vehicle.

10. The apparatus of claim 7 , wherein the preset throttle control coefficient is 0.2.

11. The apparatus of claim 7 , wherein the preset throttle control coefficient is based on a performance of the self-driving vehicle.

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

obtain driving status information of a self-driving vehicle and first driving environment information of the self-driving vehicle when the self-driving vehicle is in a self-driving status, wherein the driving status information comprises a current speed of the self-driving vehicle;

obtain a control instruction decision model through training based on training driving status information of the self-driving vehicle in a training status and training driving environment information of the self-driving vehicle in the training status;

calculate a coarse-grained control instruction of the self-driving vehicle based on the driving status information, the first driving environment information of a target vehicle and the self-driving vehicle, and the control instruction decision model, wherein the coarse-grained control instruction controls a driving manner of the self-driving vehicle, and wherein the coarse-grained control instruction comprises straight driving and in a lane that is currently being used by the self-driving vehicle;

execute, based on second driving environment information, the coarse-grained control instruction when a preset distance of the self-driving vehicle to the target vehicle is greater than a secure distance, wherein the secure distance is between the target vehicle and the self-driving vehicle and that does not cause the self-driving vehicle that is driving at a first current speed to come into contact with the target vehicle that is driving at a second current speed, and wherein the second driving environment information comprises driving environment information corresponding to the coarse-grained control instruction and when the self-driving vehicle is in the self-driving status;

obtain, based on lane information of the self-driving vehicle and the driving status information of the self-driving vehicle, the coarse-grained control instruction to control a parameter of the self-driving vehicle associated with the coarse-grained control instruction when executing the coarse-grained control instruction, wherein the lane information comprises a preset expected speed of the lane in which the self-driving vehicle drives straight, wherein a fine-grained control instruction controls the current speed of the self-driving vehicle to be less than the preset expected speed that is defined for the lane that is currently being used by the self-driving vehicle, and wherein the preset expected speed is less than a maximum speed that is defined for the lane, wherein the fine-grained control instruction comprises a value of a throttle of the self-driving vehicle, and wherein the lane information comprises first information about the lane of the self-driving vehicle that corresponds to the coarse-grained control instruction that is on a road the self-driving vehicle drives;

calculate the value of the throttle based on a first preset formula when the first current speed is not greater than the preset expected speed, and wherein the first preset formula comprises value of a throttle=preset throttle control coefficient×(preset expected speed−current speed+preset value);

execute the fine-grained control instruction; and

automatically control the self-driving vehicle in response to executing the fine-grained control instruction.

13. The computer program product of claim 12 , wherein the preset throttle control coefficient is based on a brand and a model of the self-driving vehicle.

14. The computer program product of claim 12 , wherein the preset throttle control coefficient is 0.2.

15. The computer program product of claim 12 , wherein the preset throttle control coefficient is based on a performance of the self-driving vehicle.

16. A device, comprising:

a memory configured to store program code that corresponds to executable program code;

a bus coupled to the memory; and

a processor coupled to the memory and configured for mutual communication using the bus, wherein the processor is configured to execute the program code that causes the device to be configured to:

obtain driving status information of a self-driving vehicle and first driving environment information of the self-driving vehicle when the self-driving vehicle is in a self-driving status, wherein the driving status information comprises a current speed of the self-driving vehicle;

obtain a control instruction decision model through training based on training driving status information of the self-driving vehicle in a training status and training driving environment information of the self-driving vehicle in the training status;

calculate a coarse-grained control instruction of the self-driving vehicle based on the driving status information, the first driving environment information of a target vehicle and the self-driving vehicle, and the control instruction decision model, wherein the coarse-grained control instruction comprises straight driving and controls a driving manner of the self-driving vehicle;

execute, based on second driving environment information, to execute the coarse-grained control instruction when a preset distance of the self-driving vehicle to the target vehicle is greater than a secure distance, wherein the coarse-grained control instruction comprises straight driving in a lane that is currently being used by the self-driving vehicle, wherein the secure distance is between the target vehicle and the self-driving vehicle and that does not cause the self-driving vehicle that is driving at a first current speed to come into contact with the target vehicle that is driving at a second current speed, and wherein the second driving environment information comprises driving environment information corresponding to the coarse-grained control instruction and when the self-driving vehicle is in the self-driving status;

obtain, based on lane information of the self-driving vehicle and the driving status information of the self-driving vehicle, the coarse-grained control instruction to control a parameter of the self-driving vehicle associated with the coarse-grained control instruction when executing the coarse-grained control instruction, wherein a fine-grained control instruction comprises a value of a throttle of the self-driving vehicle, wherein the fine-grained control instruction controls the current speed of the self-driving vehicle to be less than a preset expected speed that is defined for the lane that is currently being used by the self-driving vehicle, wherein the preset expected speed is less than a maximum speed that is defined for the lane, wherein the lane information comprises the preset expected speed of the lane in which the self-driving vehicle drives straight, and wherein the lane information comprises first information about the lane of the self-driving vehicle that corresponds to the coarse-grained control instruction that is on a road the self-driving vehicle drives;

calculate the value of the throttle based on a first preset formula when the first current speed is not greater than the preset expected speed, and wherein the first preset formula comprises value of a throttle=preset throttle control coefficient×(preset expected speed−current speed+preset value);

execute the fine-grained control instruction; and

automatically control the self-driving vehicle in response to executing the fine-grained control instruction.

17. The device of claim 16 , wherein the preset throttle control coefficient is based on a brand and a model of the self-driving vehicle.

18. The device of claim 16 , wherein the preset throttle control coefficient is 0.2.

19. The device of claim 16 , wherein the preset throttle control coefficient is based on a performance of the self-driving vehicle.