Efficient automatic gear shift using computer vision

Patent metadata

Field	Value
Publication number	US-11655893-B1
Application number	US-202117509666-A
Country	US
Kind code	B1
Filing date	Oct 25, 2021
Priority date	Oct 25, 2021
Publication date	May 23, 2023
Grant date	May 23, 2023

Abstract

An apparatus comprising an interface and a processor. The interface may be configured to receive pixel data of an exterior environment of a vehicle. The processor may be configured to process the pixel data arranged as video frames, perform computer vision operations to detect objects in the video frames, extract characteristics about the objects detected, determine driving conditions in response to an analysis of the characteristics and generate a control signal. The control signal may be configured to perform a gear shift. The driving conditions may be used to predict a future drivetrain configuration of the vehicle. The gear shift may be performed if a comparison of the future drivetrain configuration with a current drivetrain configuration of the vehicle meets a threshold condition. The gear shift may not be performed if the comparison does not meet the threshold condition.

First claim

The invention claimed is:

1. An apparatus comprising:
an interface configured to receive pixel data of an exterior environment of a vehicle; and
a processor configured to (i) process said pixel data arranged as video frames, (ii) perform computer vision operations to detect objects in said video frames, (iii) extract characteristics about said objects detected, (iv) determine driving conditions in response to an analysis of said characteristics and (v) generate a control signal, wherein
(a) said control signal is configured to perform a gear shift,
(b) said driving conditions are used to predict a future drivetrain configuration of said vehicle,
(c) said gear shift is performed if a comparison of said future drivetrain configuration with a current drivetrain configuration of said vehicle meets a threshold condition, and
(d) said gear shift is not performed if said comparison does not meet said threshold condition.

2. The apparatus according to claim 1 , wherein said processor is further configured to (i) determine whether said gear shift is unnecessary and (ii) prevent performing said gear shift when said gear shift is determined to be unnecessary.

3. The apparatus according to claim 2 , wherein preventing said gear shift determined to be unnecessary increases an efficiency of said vehicle by avoiding a power interruption caused by a shift time of said gear shift.

4. The apparatus according to claim 1 , wherein said future drivetrain configuration comprises a future speed of said vehicle and said current drivetrain configuration comprises a current speed of said vehicle.

5. The apparatus according to claim 1 , wherein said future drivetrain configuration comprises a future RPM of a motor and said current drivetrain configuration comprises a current RPM of said motor.

6. The apparatus according to claim 1 , wherein said future drivetrain configuration comprises a future region of a torque curve and said current drivetrain configuration comprises a current region of said torque curve.

7. The apparatus according to claim 1 , wherein said current drivetrain configuration is determined in response to reading sensors of said vehicle.

8. The apparatus according to claim 1 , wherein said driving conditions comprise a plurality of factors that indicate that said vehicle will slow down from a current speed of said vehicle.

9. The apparatus according to claim 8 , wherein said processor is further configured to analyze and provide a weighting to each of said factors detected to determine a confidence level that said vehicle will slow down from said current speed of said vehicle.

10. The apparatus according to claim 8 , wherein said factors comprise a type of detected vehicle, whether said detected vehicle is in a same lane as said vehicle, whether said detected vehicle is likely to change move in front of said vehicle, a speed of said detected vehicle, whether an alternate lane is available and driving habits of a driver of said vehicle.

11. The apparatus according to claim 1 , wherein said driving conditions are determined in response to (i) detecting a road sign and (ii) determining a vehicle reaction in response to said road sign.

12. The apparatus according to claim 11 , wherein (i) said road sign is a stop sign, (ii) said vehicle reaction to said stop sign is to slow down and (iii) said comparison of said future drivetrain configuration to said current drivetrain configuration does not meet said threshold condition.

13. The apparatus according to claim 11 , wherein (i) said road sign is a speed limit that is less than a current speed of said vehicle, (ii) said vehicle reaction to said speed limit is to slow down and (iii) said comparison of said future drivetrain configuration to said current drivetrain configuration does not meet said threshold condition.

14. The apparatus according to claim 1 , wherein said driving conditions comprise at least one of a road incline, a road curve and difficult terrain.

15. The apparatus according to claim 14 , wherein said difficult terrain comprises at least one of mud, snow, sand and crushed stone.

16. The apparatus according to claim 1 , wherein (i) said processor is configured to implement a neural network model, (ii) said driving conditions are provided as input to said neural network model and (iii) said neural network model is configured to determine whether said gear shift is unnecessary in response to said input.

17. The apparatus according to claim 16 , wherein (i) said neural network model is configured to receive updates from a neural network model source, (ii) said neural network source model is configured to update said neural network model in response to fleet learning and (iii) said fleet learning comprises a plurality of vehicles implementing said processor uploading said driving conditions with a corresponding labeled drivetrain configuration performed in response to said driving conditions to said neural network source model as training data.

18. The apparatus according to claim 1 , wherein (i) said driving conditions are further determined in response to receiving at least one of map data and live traffic data from an external source and (ii) said processor is configured to perform sensor fusion to combine information from one or more of said map data, said live traffic data and said analysis of said characteristics about said objects in order to determine said future drivetrain configuration.

19. The apparatus according to claim 1 , wherein said apparatus is configured to implement efficient automatic gear shift using computer vision.

Assignees

• Ambarella Int Lp

Inventors

• Pertsel Shimon

Classifications

• B60L2250/24

  by lever actuation · CPC title

• B60L2250/16

  by display · CPC title

• B60L2240/68

  Traffic data · CPC title

• B60L2240/647

  Surface situation of road, e.g. type of paving · CPC title

• B60L2240/507

  Operating parameters · CPC title