System and method for improving interaction of a plurality of autonomous vehicles with a driving environment including said vehicles

Patent metadata

Field	Value
Publication number	US-11964671-B2
Application number	US-202117338359-A
Country	US
Kind code	B2
Filing date	Jun 3, 2021
Priority date	Jul 8, 2019
Publication date	Apr 23, 2024
Grant date	Apr 23, 2024

Abstract

The disclosure relates to technology for determining vehicle environment interaction metrics and/or passenger behavioral metrics for autonomous vehicles in a driving environment. A group of one or more autonomous vehicles independently determine a VEI score for a target vehicle within a predefined vicinity of the group of one or more vehicles. A target vehicle may determine a VPB score based on passengers within the target vehicle. VEI scores and/or VPB scores determined for a target vehicle may be used by the target vehicle to reinforce and/or correct driving actions of the target vehicle.

First claim

What is claimed is:

1. A system for improving interaction of a plurality of autonomous vehicles within a driving environment including the autonomous vehicles, the system comprising:
a first autonomous vehicle of the plurality of autonomous vehicles; and
one or more sensors on the first autonomous vehicle, the one or more sensors configured to record environment interaction metrics of a second autonomous vehicle relating to how the second autonomous vehicle interacts with the driving environment, the first autonomous vehicle configured to transmit an indicator of the environment interaction metrics to the second autonomous vehicle for use by the second autonomous vehicle in one or more of reinforcing and correcting how the second autonomous vehicle drives.

2. The system of claim 1 , wherein the environment interaction metrics are based on one or more of how courteously and safely the second autonomous vehicle is interacting with the driving environment.

3. The system of claim 1 , wherein the environment interaction metrics comprise one or more of:
identification of the second autonomous vehicle, including for example a year, make and model of the second autonomous vehicle and registration information for the second autonomous vehicle;
current driving speed of the second autonomous vehicle;
location of the second autonomous vehicle;
actions of the second autonomous vehicle, including how often the second autonomous vehicle changes lanes;
number of passengers of the second autonomous vehicle; and
a type of sensors used in the second autonomous vehicle.

4. The system of claim 1 , wherein the environment interaction metrics are stored in a data tube and processed by an algorithm to obtain the indicator.

5. The system of claim 1 , wherein the indicator comprises a first indicator, the first autonomous vehicle further generating a second indicator based on behavioral characteristics of one or more passengers in the first autonomous vehicle, the first autonomous vehicle using the second indicator to one or more of reinforce and correct how the first autonomous vehicle drives.

6. The system of claim 5 , wherein the behavioral characteristics are captured by a second set of sensors in an interior of the first autonomous vehicle.

7. The system of claim 6 , wherein the second set of sensors capture at least one of facial expression, body pose and utterances of the one or more passengers in the first autonomous vehicle.

8. The system of claim 1 , wherein the first and second autonomous vehicles are within a predefined distance of each other.

9. The system of claim 8 , wherein the predefined distance is varied by the first autonomous vehicle based on the driving environment.

10. A system for improving interaction of a plurality of autonomous vehicles within a driving environment including the autonomous vehicles, the system comprising:
a first autonomous vehicle of the plurality of autonomous vehicles, the first autonomous vehicle comprising:
a first group of one or more sensors configured to capture a first set of data relating to characteristics of a second autonomous vehicle driving within the driving environment;
a second group of one or more sensors configured to capture a second set of data relating to behaviors of one or more passengers within the first autonomous vehicle; and
a computer executing at least one of a first algorithm and a second algorithm, the first algorithm instructing the computer to:
determine a vehicle environment interaction (VEI) radius around the first autonomous vehicle, the second autonomous vehicle being within the VEI radius of the first autonomous vehicle;
receive the first set of data captured from the first group of one or more sensors,
calculate a VEI score based on the first set of data captured, the VEI score relating to a manner in which the second autonomous vehicle is driving, and
communicate the VEI score from the first autonomous vehicle to the second autonomous vehicle for the second autonomous vehicle to one or more of reinforce and correct the manner in which the second autonomous vehicle drives; and

the second algorithm instructing the computer to:
receive the second set of data captured from the second group of one or more sensors,
calculate a vehicle passenger behavior (VPB) score based on the second set of data captured, the VPB score relating to perceived emotions of the one or more passengers in the first autonomous vehicle, and
use the VPB score calculated to one or more of reinforce and correct the manner in which the first autonomous vehicle drives.

11. The system of claim 10 , wherein a size of the VEI radius varies based on the driving environment.

12. The system of claim 10 , wherein the VEI score relates to one or more of how courteously and safely the second autonomous vehicle is interacting with the driving environment, as measured by the first autonomous vehicle.

13. The system of claim 10 , wherein the VPB score relates to one or more of how courteously and safely the first autonomous vehicle is interacting with the driving environment, as measured by the perceived emotions of the one or more passengers.

14. The system of claim 10 , further comprising a cloud service for storing information relating to the second autonomous vehicle, the computer further receiving the information from the cloud service for use in determining the VEI score.

15. The system of claim 10 , further comprising a third autonomous vehicle receiving information from the second autonomous vehicle, the information based on at least one of the VEI score and the VPB score, and the information used by the third autonomous vehicle to one or more of reinforce and correct the manner in which the third autonomous vehicle drives.

16. A method of improving interaction of a plurality of autonomous vehicles within a driving environment including the autonomous vehicles, the method comprising:
measuring, by each vehicle of a group of one or more subject vehicles of the plurality of autonomous vehicles, a set of attributes relating to a manner in which a target vehicle of the plurality of autonomous vehicles is driving;
storing the set of attributes measured in a vehicle environment interaction (VEI) tube;
calculating a VEI score for each set of attributes measured by each vehicle of the group of one or more subject vehicles using a neural network which takes data from the VEI tube as inputs, the one or more VEI scores representing how well the target vehicle is interacting with the group of one or more subject vehicles and the driving environment; and
communicating the one or more VEI scores to the target vehicle for the target vehicle to one or more of affirm and adjust driving actions of the target vehicle based on the one or more VEI scores.

17. A method of improving interaction of a plurality of autonomous vehicles within a driving environment including the autonomous vehicles, the method comprising:
measuring a set of attributes indicative of an emotional state of one or more passengers in an autonomous vehicle of the plurality of autonomous vehicles;
storing the set of attributes measured in a vehicle passenger behavior (VPB) tube;
calculating a VPB score for the set of attributes measured using a neural network which takes data from the VPB tube as inputs, the VPB scores representing a measure of an emotional state of at least one of the one or more passengers; and
adjusting driving actions of the autonomous vehicle based on the VPB scores.

Assignees

• Huawei Tech Co Ltd

Inventors

• Bill Luis
• Jia Chengcheng
• Yang Lei
• Yu Hai
• Huang Wei

Classifications

• B60W60/0015 Primary

  specially adapted for safety · CPC title

• B60W40/09

  Driving style or behaviour · CPC title

• B60W50/06

  Improving the dynamic response of the control system, e.g. improving the speed of regulation or avoiding hunting or overshoot · CPC title

• B60W2540/221

  Physiology, e.g. weight, heartbeat, health or special needs · CPC title

• B60W2540/223

  Posture, e.g. hand, foot, or seat position, turned or inclined · CPC title