Open space planner profiling tool for autonomous vehicle

Patent metadata

Field	Value
Publication number	US-12097887-B2
Application number	US-202117398359-A
Country	US
Kind code	B2
Filing date	Aug 10, 2021
Priority date	Aug 10, 2021
Publication date	Sep 24, 2024
Grant date	Sep 24, 2024

Abstract

According to various embodiments, systems, methods, and media for evaluating an open space planner in an autonomous vehicle are disclosed. In one embodiment, an exemplary method includes receiving, at a profiling application, a record file recorded by the ADV while driving in an open space using the open space planner, and a configuration file specifying parameters of the ADV; extracting planning messages and prediction messages from the record file, each extracted message being associated with the open space planner. The method further includes generating features from the planning message and the prediction messages in view of the specified parameters of the ADV; and calculating statistical metrics from the features. The statistical metrics are then provided to an automatic tuning framework for tuning the open space planner.

First claim

What is claimed is:

1. A computer-implemented method of evaluating an open space planner of an autonomous driving vehicle (ADV), the method comprising:
receiving, at a profiling application executed by a processor, a record file and a configuration file specifying parameters of the ADV, the record file having prior driving records recorded by the ADV while driving in an open space using the open space planner;
extracting, by the profiling application, a plurality of planning messages and a plurality of prediction messages from the record file, wherein each extracted message is associated with the open space planner;
generating, by the profiling application, a plurality of features from the planning messages and the prediction messages in view of the specified parameters of the ADV, wherein the plurality of features measure latency, controllability, safety, and comfort of trajectories generated by the ADV;
calculating, by the profiling application, a plurality of statistical metrics from the plurality of features; and
providing, by the profiling application, the plurality of statistical metrics to an automatic tuning framework for tuning the open space planner.

2. The method of claim 1 , wherein extracting the plurality of planning messages and the plurality of prediction messages associated with the open space planner further comprises:
extracting, from the record file, each message in the record file, wherein the record file includes at least one message in addition to the plurality of planning messages associated with the open space planner and the plurality of prediction messages associated with the open space planner;
filtering out the at least one message based on a determination the at least one message is unrelated to the open space planner; and
aligning the plurality of planning messages and the plurality of prediction messages based on timestamps included in each of the plurality of planning messages and each of the plurality of prediction messages.

3. The method of claim 1 , wherein the profiling application is configured to run either on the ADV or on a cloud server.

4. The method of claim 1 , wherein generating the plurality of features includes extracting one or more of the plurality of features from the plurality of planning messages, and calculating one or more of the plurality of features based on the one or more extracted features and the parameters of the ADV.

5. The method of claim 1 , wherein the parameters of the ADV specified in the configuration file includes a steering ratio, a wheel base, and a maximum speed of the ADV.

6. The method of claim 1 , wherein the statistical metrics include a mean, and a 95 percentile calculated from the plurality of features.

7. The method of claim 1 , wherein the statistical metrics are displayed on a graphical user interface as visualization plots, and are sent to a user as an email report.

8. A non-transitory machine-readable medium having instructions stored therein, which when executed by a processor, cause the processor to perform operations of evaluating an open space planner of an autonomous driving vehicle (ADV), the operations comprising:
receiving a record file and a configuration file specifying parameters of the ADV, the record file having prior driving records recorded by the ADV while driving in an open space using the open space planner;
extracting a plurality of planning messages and a plurality of prediction messages from the record file, wherein each extracted message is associated with the open space planner;
generating a plurality of features from the planning messages and the prediction messages in view of the specified parameters of the ADV, wherein the plurality of features measure latency, controllability, safety, and comfort of trajectories generated by the ADV;
calculating a plurality of statistical metrics from the plurality of features; and
providing the plurality of statistical metrics to an automatic tuning framework for tuning the open space planner.

9. The non-transitory machine-readable medium of claim 8 , wherein extracting the plurality of planning messages and the plurality of prediction messages further comprises:
extracting, from the record file, each message in the record file, wherein the record file includes at least one message in addition to the plurality of planning messages associated with the open space planner and the plurality of prediction messages associated with the open space planner;
filtering out the at least one message based on a determination that the at least one message is unrelated to the open space planner; and
aligning the plurality of planning messages and the plurality of prediction messages based on timestamps included in each of the plurality of planning messages and each of the plurality of prediction messages.

10. The non-transitory machine-readable medium of claim 8 , wherein the statistical metrics are displayed on a graphical user interface as visualization plots, and are sent to a user as an email report.

11. The non-transitory machine-readable medium of claim 8 , wherein generating the plurality of features includes extracting one or more of the plurality of features from the plurality of planning messages, and calculating one or more of the plurality of features based on the one or more extracted features and the parameters of the ADV.

12. The non-transitory machine-readable medium of claim 8 , wherein the parameters of the ADV specified in the configuration file includes a steering ratio, a wheel base, and a maximum speed of the ADV.

13. The non-transitory machine-readable medium of claim 8 , wherein the statistical metrics include a mean, and a 95 percentile calculated from the plurality of features.

14. A data processing system, comprising:
a processor; and
a memory coupled to the processor to store instructions, which when executed by the processor, cause the processor to perform operations of evaluating an open space planner of an autonomous driving vehicle (ADV), the operations comprising:
receiving a record file and a configuration file specifying parameters of the ADV, the record file having prior driving records recorded by the ADV while driving in an open space using the open space planner;
extracting plurality of planning messages and a plurality of prediction messages from the record file, wherein each extracted message is associated with the open space planner;
generating a plurality of features from the planning messages and the prediction messages in view of the specified parameters of the ADV, wherein the plurality of features measure latency, controllability, safety, and comfort of trajectories generated by the ADV;
calculating a plurality of statistical metrics from the plurality of features; and
providing plurality of statistical metrics to an automatic tuning framework for tuning the open space planner.

15. The data processing system of claim 14 , wherein the statistical metrics are displayed on a graphical user interface as visualization plots, and are sent to a user as an email report.

16. The data processing system of claim 14 , wherein the profiling application is configured to run either on the ADV or on a cloud server.

17. The data processing system of claim 14 , wherein extracting the plurality of planning messages and the plurality of prediction messages further comprises:
extracting, from the record file, each message in the record file, wherein the record file includes at least one message in addition to the plurality of planning messages associated with the open space planner and the plurality of

Assignees

• Baidu Usa Llc

Inventors

• Jiang Shu
• Luo Qi
• Cao Yu
• Lin Weiman
• Wang Yu
• Sun Hongyi

Classifications

• B60W2720/10

  Longitudinal speed · CPC title

• B60W2510/20

  Steering systems · CPC title

• B60W2050/0083

  Setting, resetting, calibration · CPC title

• B60W50/045

  Monitoring control system parameters · CPC title

• B60W2556/45

  External transmission of data to or from the vehicle · CPC title