Steering mode selection using machine learning

Patent metadata

Field	Value
Publication number	US-11254351-B2
Application number	US-201715457411-A
Country	US
Kind code	B2
Filing date	Mar 13, 2017
Priority date	Mar 13, 2017
Publication date	Feb 22, 2022
Grant date	Feb 22, 2022

Abstract

Technical features of a steering system include a control module that dynamically determines an operating mode based on a set of input signal values such as lateral acceleration signal values and corresponding handwheel position values. The control module dynamically determines and learns classification boundaries between multiple operating modes based the input signal values. The control module further calibrates the steering system according to operating mode that is determined using the classification boundaries.

First claim

What is claimed is:

1. A steering system comprising:
a control module configured to dynamically determine an operating mode based on a set of input signal values comprising a plurality of lateral acceleration signal values and respectively corresponding handwheel position values; and
the control module configured to calibrate the steering system according to operating mode that is determined.

2. The steering system of the claim 1 , wherein the set of input signal values is a first set of input signal values, the operating mode is a first operating mode, and wherein the control module is further configured to:
receive a second set of input signal values comprising a second plurality of lateral acceleration values and respectively corresponding handwheel position values;
dynamically determine a second operating mode based on the second set of input signal values; and
switch the steering system from the first operating mode to the second operating mode by calibrating the steering system according to the second operating mode.

3. The steering system of claim 2 , wherein the control module switches the steering system from the first operating mode to the second operating mode without manual request for the switch.

4. The steering system of claim 2 , wherein the first operating mode has a first torque assist feedback value and the second operating mode has a second torque assist feedback value.

5. The steering system of claim 1 , wherein the set of input signal values further comprises a plurality of yaw rate signal values.

6. The steering system of claim 2 , wherein the control module is further configured to indicate to an electronic control unit the switch from the first operating mode to the second operating mode.

7. The steering system of claim 1 , wherein the control module is further configured to indicate to an electronic control unit the dynamically detected operating mode.

8. The steering system of claim 1 , wherein the control module is configured to dynamically determine the operating mode based on the input signal values by comparing the input signal values with a decision boundary, the decision boundary being determined by machine learning.

9. The steering system of claim 8 , wherein the control module is configured to determine the decision boundary based on:
a min-table that comprises an entry for a handwheel position indicative of a minimum lateral acceleration signal value at said handwheel position; and
a max-table that comprises an entry for said handwheel position indicative of a maximum lateral acceleration signal value at said handwheel position.

10. The steering system of claim 9 , wherein the control module determines the decision boundary based on the min-table and the max-table in response to the lateral acceleration signal values in the min-table and the max-table being symmetric across handwheel position values.

11. A method implemented by a steering system, the method comprising:
operating according to a first operating mode;
receiving a set of input signal values comprising a plurality of lateral acceleration signal values and respectively corresponding handwheel position values;
determining a second operating mode based on the set of input signal values and a decision boundary; and
switching the steering system from the first operating mode to the second operating mode by calibrating the steering system according to the second operating mode.

12. The method of claim 11 , further comprising:
determining, prior to switching, a handwheel velocity of the steering system; and
switching the steering system from the first operating mode to the second operating mode in response to the handwheel velocity being below a predetermined threshold.

13. The method of claim 11 , wherein the first operating mode has a first torque assist feedback value and the second operating mode has a second torque assist feedback value.

14. The method of claim 11 , wherein the decision boundary is a first decision boundary based on a first set of sample input values, and the method further comprising, determining a second decision boundary based on a second set of sample input values in response to a change in a predetermined set of components.

15. The method of claim 11 , further comprising indicating to an electronic control unit of a component of a vehicle, the second operating mode.

16. A power steering system comprising:
a control module configured to:
dynamically learn a classification boundary between a first operating mode and a second operating mode, the classification boundary based on a first set of input signal values comprising a plurality of lateral acceleration signal values and respectively corresponding handwheel position values;
collect a second set of input signal values during operation of the power steering system in the first operating mode; and
dynamically switch from the first operating mode to the second operating mode based on the second set of input signal values corresponding to the second operating mode.

17. The power steering system of claim 16 , wherein the control module determines the classification boundary by populating a min-table and a max-table using the first set of input signal values, wherein:
the min-table comprises an entry for each respective handwheel position, the entry indicative of a minimum lateral acceleration signal value at said handwheel position; and
the max-table comprises an entry for the each respective handwheel position, the entry indicative of a maximum lateral acceleration signal value at said handwheel position.

18. The power steering system of claim 17 , wherein the control module determines the classification boundary based on a difference between the maximum lateral acceleration signal values and the minimum lateral acceleration signal values.

19. The power steering system of claim 17 , wherein the control module determines the classification boundary in response to determining that the lateral acceleration signal values in the min-table and the max-table are symmetric across handwheel position values.

20. The power steering system of claim 16 , wherein the control module is further configured to determine, prior to switching the operating modes, a handwheel velocity, and to switch the first operating mode to the second operating mode in response to the handwheel velocity being below a predetermined threshold.

Assignees

• Steering Solutions Ip Holding

Inventors

• Ryne Zaki
• Mitra Rakesh
• Yang Tao
• Sanford Scott T
• Story Michael R

Classifications

• G05B13/0265 Primary

  the criterion being a learning criterion · CPC title

• B62D6/08

  responsive only to {driver} input torque · CPC title

• B62D5/0463 Primary

  calculating assisting torque from the motor based on driver input · CPC title

• B62D5/0457

  characterised by control features of the drive means as such · CPC title

• B62D5/0421

  Electric motor acting on or near steering gear · CPC title