Scenario based control of autonomous driving vehicle

Patent metadata

Field	Value
Publication number	US-11740628-B2
Application number	US-202016823066-A
Country	US
Kind code	B2
Filing date	Mar 18, 2020
Priority date	Mar 18, 2020
Publication date	Aug 29, 2023
Grant date	Aug 29, 2023

Abstract

In one embodiment, control of an autonomous driving vehicle (ADV) includes determining a current scenario of the ADV. Based on the scenario, a control algorithm is selected among a plurality of distinct control algorithms as the active control algorithm. One or more control commands are generated using the active control algorithm, based one or more target inputs. The control commands are applied to effect movement of the ADV.

First claim

What is claimed is:

1. A method for operating an autonomous driving vehicle (ADV), the method comprising:
perceiving a driving environment surrounding the ADV;
determining a current scenario of the ADV from a plurality of driving scenarios that includes a parking scenario and a normal driving scenario, based on the driving environment;
based on the current scenario, selecting a control algorithm among a plurality of distinct control algorithms, each of the control algorithms being associated with at least one of the driving scenarios;
generating a control command with the selected control algorithm by invoking a respective controller corresponding to the selected control algorithm from a plurality of controllers associated with the driving scenarios, based on one or more control target inputs, wherein each of the plurality of controllers share a common application programming interface (API) that is used to invoke the respective controller, and wherein a first of the plurality of distinct control algorithms that corresponds to the parking scenario generates the control command to reduce differences between a target heading of the ADV and a current heading of the ADV, more aggressively than a second of the plurality of distinct control algorithms that corresponds to the normal driving scenario; and
applying the control command to effect movement of the ADV, wherein each of the plurality of distinct control algorithms share in common, a common pre-processing protocol and a common post-processing protocol that formats and outputs the control command onto a vehicle communication bus, and wherein the common post processing protocol includes emergency logic that determines that an emergency condition is present, and in response to the emergency condition being present, the emergency logic generates an emergency command that overrides the control command.

2. The method of claim 1 , wherein the current scenario of the ADV is determined based on at least two of: map information, current vehicle location, or current vehicle speed.

3. The method of claim 1 , wherein under the parking scenario, the ADV is controlled to drive in any direction, and under the normal driving scenario the ADV is controlled to drive based on a driving lane.

4. The method of claim 1 , wherein the first of the plurality of distinct control algorithms that corresponds to the parking scenario allows throttle in forward and reverse, and the second of the plurality of distinct control algorithms that corresponds to the normal driving scenario does not allow throttle in reverse.

5. The method of claim 1 , wherein the parking scenario generates the control command to reduce differences between the target heading of the ADV and the current heading of the ADV on a frame-by-frame basis.

6. The method of claim 3 , wherein a first of the plurality of distinct control algorithms that corresponds to the parking scenario generates the control command to reduce differences between a target speed of the ADV and a current speed of the ADV less aggressively than a second of the plurality of distinct control algorithms that corresponds to the normal driving scenario.

7. The method of claim 1 , wherein the one or more control target inputs include at least one of: a target location of the ADV, a current location of the ADV, a target speed of the ADV, a current speed of the ADV, the target heading of the ADV, or a current heading of the ADV.

8. The method of claim 1 , wherein the control command includes at least one of: a steering command, a throttle command, or a braking command.

9. The method of claim 1 , wherein the common pre-processing protocol reads the current scenario to select the control algorithm based on the current scenario, and retrieves the one or more control target inputs from one or more of: a vehicle localization system, a vehicle planning system, or a chassis channel.

10. The method of claim 1 , wherein the control command is received by a throttle actuator, a steering actuator, or a braking actuator.

11. The method of claim 10 , wherein the control command causes the ADV to slow down.

12. The method of claim 1 , wherein each of the plurality of distinct control algorithms are configurable through a configuration file stored in electronic memory.

13. A non-transitory machine-readable medium having instructions stored therein, which when executed by a processor, cause the processor to perform operations of operating an autonomous driving vehicle (ADV), the operations comprising:
perceiving a driving environment surrounding the ADV;
determining a current scenario of the ADV from a plurality of driving scenarios that includes a parking scenario and a normal driving scenario, based on the driving environment;
based on the current scenario, selecting a control algorithm among a plurality of distinct control algorithms, each of the control algorithms being associated with at least one of the driving scenarios;
generating a control command with the selected control algorithm by invoking a respective controller corresponding to the selected control algorithm from a plurality of controllers associated with the driving scenarios, based on one or more control target inputs wherein each of the plurality of controllers share a common application programming interface (API) that is used to invoke the respective controller, and wherein a first of the plurality of distinct control algorithms that corresponds to the parking scenario generates the control command to reduce differences between a target heading of the ADV and a current heading of the ADV, more aggressively than a second of the plurality of distinct control algorithms that corresponds to the normal driving scenario; and
applying the control command to effect movement of the ADV, wherein each of the plurality of distinct control algorithms share in common, a common pre-processing protocol and a common post-processing protocol that formats and outputs the control command onto a vehicle communication bus, and wherein the common post processing protocol includes emergency logic that determines that an emergency condition is present, and in response to the emergency condition being present, the emergency logic generates an emergency command that overrides the control command.

14. The non-transitory machine-readable medium of claim 13 , wherein the current scenario of the ADV is determined based on at least two of: map information, current vehicle location, current vehicle speed.

15. The non-transitory machine-readable medium of claim 13 , wherein under the parking scenario, the ADV is controlled to drive in any direction, and under the normal driving scenario the ADV is controlled to drive based on a driving lane.

16. The non-transitory machine-readable medium of claim 1 , wherein the first of the plurality of distinct control algorithms that corresponds to the parking scenario allows throttle in forward and reverse, and the second of the plurality of distinct control algorithms that corresponds to the normal driving scenario does not allow throttle in reverse.

17. A data processing system comprising:
a processor; and
memory, coupled to the processor, to store instructions which, when executed by the processor, cause the processor to perform operations of operating an autonomous driving vehicle (ADV), the operations including:
perceiving a driving environment surrounding the ADV;
determining a current scenario of the ADV from a plurality of driving scenarios that includes a parking scenario and a normal driving scenario, based on the driving environment;
based on the curr

Assignees

• Baidu Usa Llc

Inventors

• Jiang Shu
• Luo Qi
• Miao Jinghao
• Hu Jiangtao
• Wang Yu
• Zhou Jinyun
• Tao Jiaming
• Xiao Xiangquan

Classifications

• B60W2556/50

  of positioning data, e.g. GPS [Global Positioning System] data · CPC title

• B60W2552/53

  Road markings, e.g. lane marker or crosswalk · CPC title

• B60W2552/05

  Type of road, e.g. motorways, local streets, paved or unpaved roads · CPC title

• B60W2050/0095

  Automatic control mode change · CPC title

• B60W30/18036

  Reversing · CPC title