Segmenting a parking trajectory to control an autonomous driving vehicle to park

Patent metadata

Field	Value
Publication number	US-11485353-B2
Application number	US-201916399579-A
Country	US
Kind code	B2
Filing date	Apr 30, 2019
Priority date	Apr 30, 2019
Publication date	Nov 1, 2022
Grant date	Nov 1, 2022

Abstract

In one embodiment, a computer-implemented method of autonomously parking an autonomous driving vehicle, includes generating environment descriptor data describing a driving environment surrounding the autonomous driving vehicle (ADV), including identifying a parking space and one or more obstacles within a predetermined proximity of the ADV, generating a parking trajectory of the ADV based on the environment descriptor data to autonomously park the ADV into the parking space, including optimizing the parking trajectory in view of the one or more obstacles, segmenting the parking trajectory into one or more trajectory segments based on a vehicle state of the ADV, and controlling the ADV according to the one or more trajectory segments of the parking trajectory to autonomously park the ADV into the parking space without collision with the one or more obstacles.

First claim

What is claimed is:

1. A computer-implemented method of autonomously parking an autonomous driving vehicle (ADV), comprising:
generating, by a first processing thread, environment descriptor data describing a driving environment surrounding the ADV, including identifying a parking space and one or more obstacles within a predetermined proximity of the ADV;
generating, by a second processing thread, a first parking trajectory of the ADV based on the environment descriptor data to autonomously park the ADV into the parking space, optimizing the first parking trajectory in view of the one or more obstacles, wherein in response to moving of the one or more obstacles, the second processing thread is triggered to generate a second parking trajectory of the ADV, wherein the generating the first parking trajectory of the ADV is performed via the second processing thread executed independently by a different computing node having a separate set of processing resources decoupled from the first processing thread, wherein the first thread is executed based on a first processing cycle while the second thread is executed based on a second processing cycle, and wherein the second processing cycle has a 1 second or more longer period than a time period of the first processing cycle;
segmenting the second parking trajectory into one or more trajectory segments based on a vehicle state of the ADV, wherein the vehicle state comprises one or more parameters of the ADV including a position, a heading, and a speed of the ADV; and
controlling the ADV according to the one or more trajectory segments of the second parking trajectory to autonomously park the ADV into the parking space without collision with the one or more obstacles.

2. The method of claim 1 , further comprising storing trajectory data describing the second parking trajectory in a memory, wherein the stored trajectory data of the second parking trajectory can be utilized again in response to determining the second thread fails to produce a new parking trajectory within a predetermined period of time.

3. The method of claim 2 , wherein the predetermined period of time is 5 seconds.

4. The method of claim 1 , wherein the second parking trajectory is updated to avoid a collision with the one or more obstacles proximate to the ADV moving at different points in time in response to moving of the one or more obstacles.

5. The method of claim 1 , wherein the one or more trajectory segments include at least one of a forward driving segment or a reverse driving segment.

6. The method of claim 1 , further comprising performing a safety check of the ADV's surrounding area based on the environment descriptor data, while the ADV is moving along the second parking trajectory to autonomously park in the parking space.

7. The method of claim 1 , wherein the time period of the first processing cycle is 100 milliseconds.

8. The method of claim 1 , wherein the first processing thread runs at a higher frequency than the second processing thread.

9. A non-transitory machine-readable medium having instructions stored therein, which when executed by a processor, cause the processor to perform operations, the operations comprising:
generating, by a first processing thread, environment descriptor data describing a driving environment surrounding an autonomous driving vehicle (ADV), including identifying a parking space and one or more obstacles within a predetermined proximity of the ADV;
generating, by a second processing thread, a first parking trajectory of the ADV based on the environment descriptor data to autonomously park the ADV into the parking space, optimizing the first parking trajectory in view of the one or more obstacles, wherein in response to moving of the one or more obstacles, the second processing thread is triggered to generate a second parking trajectory of the ADV, wherein the generating the first parking trajectory of the ADV is performed via the second processing thread executed independently by a different computing node having a separate set of processing resources decoupled from the first processing thread, wherein the first thread is executed based on a first processing cycle while the second thread is executed based on a second processing cycle, and wherein the second processing cycle has a 1 second or more longer period than a time period of the first processing cycle;
segmenting the second parking trajectory into one or more trajectory segments based on a vehicle state of the ADV, wherein the vehicle state comprises one or more parameters of the ADV including a position, a heading, and a speed of the ADV; and
controlling the ADV according to the one or more trajectory segments of the second parking trajectory to autonomously park the ADV into the parking space without collision with the one or more obstacles.

10. The machine-readable medium of claim 9 , wherein the operations further comprise storing trajectory data describing the second parking trajectory in a memory, wherein the stored trajectory data of the second parking trajectory can be utilized again in response to determining the second thread fails to produce a new parking trajectory within a predetermined period of time.

11. The machine-readable medium of claim 9 , wherein the second parking trajectory is updated to avoid a collision with the one or more obstacles proximate to the ADV moving at different points in time in response to moving of the one or more obstacles.

12. The machine-readable medium of claim 9 , wherein the one or more trajectory segments include at least one of a forward driving segment or a reverse driving segment.

13. The machine-readable medium of claim 9 , wherein the operations further comprise performing a safety check of the ADV's surrounding area based on the environment descriptor data, while the ADV is moving along the second parking trajectory to autonomously park in the parking space.

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, the operations including:
generating, by a first processing thread, environment descriptor data describing a driving environment surrounding an autonomous driving vehicle (ADV), including identifying a parking space and one or more obstacles within a predetermined proximity of the ADV,
generating, by a second processing thread, a first parking trajectory of the ADV based on the environment descriptor data to autonomously park the ADV into the parking space, optimizing the first parking trajectory in view of the one or more obstacles, wherein in response to moving of the one or more obstacles, the second processing thread is triggered to generate a second parking trajectory of the ADV, wherein the generating the first parking trajectory of the ADV is performed via the second processing thread executed independently by a different computing node having a separate set of processing resources decoupled from the first processing thread, wherein the first thread is executed based on a first processing cycle while the second thread is executed based on a second processing cycle, and wherein the second processing cycle has a 1 second or more longer period than a time period of the first processing cycle,
segmenting the second parking trajectory into one or more trajectory segments based on a vehicle state of the ADV, wherein the vehicle state comprises one or more parameters of the ADV including a position, a heading, and a speed of the ADV, and
controlling the ADV according to the one or more trajectory segments of the second parking tr

Assignees

• Baidu Usa Llc

Inventors

• Zhou Jinyun
• He Runxin
• Luo Qi
• Miao Jinghao
• Hu Jiangtao
• Wang Yu
• Xu Jiaxuan
• Jiang Shu

Classifications

• B60W30/06 Primary

  Automatic manoeuvring for parking · CPC title

• B60W60/001

  Planning or execution of driving tasks · CPC title

• G06V20/586

  of parking space · CPC title

• G06V20/58

  Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads · CPC title

• G05D1/0088

  characterized by the autonomous decision making process, e.g. artificial intelligence, predefined behaviours (using knowledge based models G06N5/00) · CPC title