Extended model reference adaptive control algorithm for the vehicle actuation time-latency

Patent metadata

Field	Value
Publication number	US-11453409-B2
Application number	US-202016854718-A
Country	US
Kind code	B2
Filing date	Apr 21, 2020
Priority date	Apr 21, 2020
Publication date	Sep 27, 2022
Grant date	Sep 27, 2022

Abstract

Systems and methods are disclosed for reducing second order dynamics delays in a control subsystem (e.g. throttle, braking, or steering) in an autonomous driving vehicle (ADV) and increasing control system bandwidth by accounting for time-latency in a control subsystem actuation system. A control input is received from an ADV's autonomous driving system. The control input is translated into a control command of the control subsystem of the ADV. A reference actuation output and a predicted actuation output are generated corresponding to a by-wire (“real”) actuation action for the control subsystem. A control error is determined between the reference actuation action and the by-wire actuation action. A predicted control error is determined between the predicted actuation action and the between the by-wire actuation action. Adaptive gains are determined and applied to the by-wire actuation action to generate a second by-wire actuation action.

First claim

What is claimed is:

1. A computer-implemented method of improving performance of a control subsystem of an autonomous driving vehicle (ADV), the method comprising:
receiving, by a feedback controller and a reference system, a control input of an actuation system of the control subsystem of the ADV;
generating a predicted actuation output corresponding to the control input and a reference actuation output corresponding to the control input;
determining a control error between a first actuation command output of an actuator of the control subsystem and the reference actuation output, and determining a predicted control error between the first actuation command output and the predicted actuation output;
generating an adaptive gain that is based upon the control error and the predicted control error;
outputting a second actuation command to actuate the control subsystem, wherein the second actuation command is based upon the first actuation command output and the adaptive gain; and
generating a second predicted actuation output based at least in part upon a temporary actuation output and time-latency information for the temporary actuation output.

2. The method of claim 1 , wherein the reference actuation output is based at least in part on second order attributes of a dynamic model of the control subsystem.

3. The method of claim 2 , wherein the second order attributes of the control subsystem include one or more of: time-latency in initiating a control action of the control subsystem, a rise time, an overshoot amount, or a settling time of an actuator of the control subsystem.

4. The method of claim 1 , wherein the control subsystem of the ADV is one or more of a throttle control subsystem, a braking control subsystem, or a steering control subsystem.

5. The method of claim 1 , wherein generating the predicted actuation output further includes:
generating time-latency information and the predicted actuation output is based at least in part on the generated time-latency information.

6. The method of claim 1 ,
wherein the temporary actuation output is generated by applying the adaptive gain to the first actuation command output.

7. The method of claim 6 , further comprising generating a second reference actuation output, comprising:
applying a time-delay compensation function to the first actuation command output and the adaptive gain to generate a second temporary actuation output, and the second reference actuation output is based at least in part upon the second temporary actuation output.

8. The method of claim 7 , further comprising:
determining a second predicted control error between a second actuation command output and the second predicted actuation output;
determining a second control error between the second actuation command output and the second predicted actuation output; and
generating second adaptive gains based upon the second control error and the second predicted control error.

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:
receiving, by a feedback controller and a reference system, an actuation command of an actuation system of a control subsystem of an autonomous driving vehicle (ADV);
generating a predicted actuation output corresponding to the actuation command and a reference actuation output corresponding to the actuation command;
determining a control error between a first actuation command output of an actuator of the control subsystem and the reference actuation output, and determining a predicted control error between the first actuation command output and the predicted actuation output;
generating an adaptive gain that is based upon the control error and the predicted control error;
outputting a second actuation command to actuate the control subsystem, wherein the second actuation command is based upon the first actuation command output and the adaptive gain; and
generating a second predicted actuation output based at least in part upon a temporary actuation output and time-latency information for the temporary actuation output.

10. The medium of claim 9 , wherein the reference actuation output is based at least in part on second order attributes of a dynamic model of the control subsystem.

11. The medium of claim 10 , wherein the second order attributes of the control subsystem include one or more of: time-latency in initiating a control action of the control subsystem, a rise time, an overshoot amount, or a settling time of an actuator of the control subsystem.

12. The medium of claim 9 , wherein the control subsystem of the ADV is one or more of a throttle control subsystem, a braking control subsystem, or a steering control subsystem.

13. The medium of claim 9 , wherein generating the predicted actuation output further includes:
generating time-latency information and the predicted actuation output is based at least in part on the generated time-latency information.

14. The medium of claim 9 ,
wherein the temporary actuation output is generated by applying the adaptive gain to the first actuation command output.

15. The medium of claim 14 , the operations further comprising generating a second reference actuation output, comprising:
applying a time-delay compensation function to the first actuation command output and the adaptive gain to generate a second temporary actuation output, and the second reference actuation output is based at least in part upon the second temporary actuation output.

16. The medium of claim 15 , the operations further comprising:
determining a second predicted control error between a second actuation command output and the second predicted actuation output;
determining a second control error between the second actuation command output and the second predicted actuation output; and
generating second adaptive gains based upon the second control error and the second predicted control error.

17. 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:
receiving, by a feedback controller and a reference system, an actuation command of an actuation system of a control subsystem of an autonomous driving vehicle (ADV);
generating a predicted actuation output corresponding to the actuation command and a reference actuation output corresponding to the actuation command;
determining a control error between a first actuation command output of an actuator of the control subsystem and the reference actuation output, and determining a predicted control error between the first actuation command output and the predicted actuation output;
generating an adaptive gain that is based upon the control error and the predicted control error;
outputting a second actuation command to actuate the control subsystem, wherein the second actuation command is based upon the first actuation command output and the adaptive gain; and
generating a second predicted actuation output based at least in part upon a temporary actuation output and time-latency information for the temporary actuation output.

18. The system of claim 17 , wherein the control subsystem of the ADV is one or more of a throttle control subsystem, a braking control subsystem, or a steering control subsystem.

19. The system of claim 17 , wherein generating the predicted actuation output further includes:
gen

Assignees

• Baidu Usa Llc

Inventors

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

Classifications

• B62D15/025

  Active steering aids, e.g. helping the driver by actively influencing the steering system after environment evaluation (B62D1/28 takes precedence) · CPC title

• B60W2050/0037

  Mathematical models of vehicle sub-units · CPC title

• B60W2510/20

  Steering systems · CPC title

• B60W10/18

  including control of braking systems · CPC title

• B60W2050/0043

  Signal treatments, identification of variables or parameters, parameter estimation or state estimation · CPC title