Centralized scheduling system using event loop for operating autonomous driving vehicles

Patent metadata

Field	Value
Publication number	US-10732634-B2
Application number	US-201715640842-A
Country	US
Kind code	B2
Filing date	Jul 3, 2017
Priority date	Jul 3, 2017
Publication date	Aug 4, 2020
Grant date	Aug 4, 2020

Abstract

An event queue is maintained to store IO events generated from a number of sensors and timer events generated for a number of autonomous driving modules. For each of the events pending in the event queue, in response to determining that the event is an IO event, the data associated with the IO event is stored in a data structure associated with the sensor in a global store. In response to determining that the event is a timer event, a worker thread associated with the timer event is launched. The worker thread executes one of the autonomous driving modules triggered or initiated the timer event. Input data is retrieved from the global store and provided to the worker thread to allow the worker thread to process the input data.

First claim

What is claimed is:

1. A computer-implemented method for operating an autonomous driving vehicle (ADV), the method comprising:
maintaining an event queue to store input and output (TO) events generated from a plurality of sensors and timer events generated for a plurality of autonomous driving modules, the autonomous driving modules including a perception module to perceive a driving environment surrounding the ADV based on sensor data obtained from the sensors and a planning module to plan a trajectory to autonomously drive the ADV through the driving environment; and
for each of the events stored in the event queue, iteratively performing
in response to determining that the event is an IO event, storing data associated with the IO event in a global store, and
in response to determining that the event is a timer event,
launching a worker thread associated with the timer event, including identifying a first autonomous driving module based on the timer event, and launching the worker thread to execute a first autonomous driving module that triggered the timer event, and
identifying, by hashing an identifier of the timer event, one or more data structures of the global store to retrieve input data stored in the global store that is associated with the timer event; and
providing the input data retrieved from the global store to the worker thread to allow the worker thread to process the input data;

receiving a notification from the worker thread indicating that the input data has been processed;
in response to the notification, extracting first output data from the notification; and
storing the first output data in the global store.

2. The method of claim 1 , wherein the input data is retrieved from a first data structure of the one or more data structures stored in the global store, wherein the first data structure is specified by hashing the identifier of the timer event.

3. The method of claim 1 , wherein providing the input data to the first autonomous driving module comprises passing a memory pointer of a memory location of the global store to the worker thread without copying the input data.

4. The method of claim 1 , wherein the first output data is stored in a first data structure of the one or more data structures of the global store, and wherein the first data structure is associated with the first autonomous driving module.

5. The method of claim 1 , wherein each data structure of the global store is associated with one of the plurality of sensors or one of the plurality of autonomous driving modules.

6. The method of claim 1 , wherein in response to the notification, one or more second data structures of the global store that are associated with the first autonomous driving module are identified, the one or more second data structures being used to store the first output data.

7. 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:
maintaining an event queue to store input and output (TO) events generated from a plurality of sensors and timer events generated for a plurality of autonomous driving modules, the autonomous driving modules including a perception module to perceive a driving environment surrounding the ADV based on sensor data obtained from the sensors and a planning module to plan a trajectory to autonomously drive the ADV through the driving environment; and
for each of the events stored in the event queue, iteratively performing
in response to determining that the event is an IO event, storing data associated with the IO event in a global store, and
in response to determining that the event is a timer event,
launching a worker thread associated with the timer event, including identifying a first autonomous driving module based on the timer event, and launching the worker thread to execute a first autonomous driving module that triggered the timer event;
identifying, by hashing an identifier of the timer event, one or more data structures of the global store to retrieve input data stored in the global store that is associated with the timer event; and
providing the input data retrieved from the global store to the worker thread to allow the worker thread to process the input data

receiving a notification from the worker thread indicating that input data has been processed;
in response to the notification, extracting first output data from the notification; and
storing the first output data in the global store.

8. The machine-readable medium of claim 7 , wherein the input data is retrieved from a first data structure of the one or more data structures stored in the global store, wherein the first data structure is specified by hashing the identifier of the timer event.

9. The machine-readable medium of claim 7 , wherein providing the input data to the first autonomous driving module comprises passing a memory pointer of a memory location of the global store to the worker thread without copying the input data.

10. The machine-readable medium of claim 7 , wherein the first output data is stored in a first data structure of the one or more data structures of the global store, and wherein the first data structure is associated with the first autonomous driving module.

11. The machine-readable medium of claim 7 , wherein each data structure of the global store is associated with one of the plurality of sensors or one of the plurality of autonomous driving modules.

12. The machine-readable medium of claim 7 , wherein in response to the notification, one or more second data structures of the global store that are associated with the first autonomous driving module are identified, the one or more second data structures being used to store the first output data.

13. 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
maintaining an event queue to store input and output (TO) events generated from a plurality of sensors and timer events generated for a plurality of autonomous driving modules, the autonomous driving modules including a perception module to perceive a driving environment surrounding the ADV based on sensor data obtained from the sensors and a planning module to plan a trajectory to autonomously drive the ADV through the driving environment, and
for each of the events stored in the event queue, iteratively performing
in response to determining that the event is an IO event, storing data associated with the IO event in a global store, and
in response to determining that the event is a timer event,
launching a worker thread associated with the timer event, including identifying a first autonomous driving module based on the timer event, and launching the worker thread to execute a first autonomous driving module that triggered the timer event; and
providing the input data retrieved from the global store to the worker thread to allow the worker thread to process the input data

receiving a notification from the worker thread indicating that the input data has been processed;
in response to the notification, extracting first output data from the notification; and
storing the first output data in the global store.

14. The system of claim 13 , wherein the first input data is retrieved from a first data structure stored in the global store, wherein the first data structure is specified by the timer event.

15. The system of claim 13 , wherein providing the input data to t

Assignees

• Baidu Usa Llc
• Baidu Us Llc

Inventors

• Yang Yiqing
• Yu Siyang
• Liu Xuan
• Cao Yu
• Li Zhang
• Zhan Jun
• Yang Guang

Classifications

• G06F9/542

  Event management; Broadcasting; Multicasting; Notifications · CPC title

• G06F13/26

  with priority control · CPC title

• G06F9/4831 Primary

  with variable priority · CPC title

• G06Q10/00

  Administration; Management · CPC title

• G05D1/0212 Primary

  with means for defining a desired trajectory (involving a plurality of land vehicles G05D1/0287) · CPC title