Autonomous vehicle operation feature monitoring and evaluation of effectiveness

Patent metadata

Field	Value
Publication number	US-2022005291-A1
Application number	US-201715421508-A
Country	US
Kind code	A1
Filing date	Feb 1, 2017
Priority date	May 20, 2014
Publication date	Jan 6, 2022
Grant date	—

Abstract

Methods and systems for monitoring use and determining risks associated with operation of a vehicle having one or more autonomous operation features are provided. According to certain aspects, operating data may be recorded during operation of the vehicle. This may include information regarding the vehicle, the vehicle environment, use of the autonomous operation features, and/or control decisions made by the features. The control decisions may include actions the feature would have taken to control the vehicle, but which were not taken because a vehicle operator was controlling the relevant aspect of vehicle operation at the time. The operating data may be recorded in a log, which may then be used to determine risk levels associated with vehicle operation based upon risk levels associated with the autonomous operation features. The risk levels may further be used to adjust an insurance policy associated with the vehicle.

First claim

1 . A computer system for monitoring vehicles having one or more autonomous systems or features, comprising:
one or more processors; and a non-transitory program memory coupled to the one or more processors and storing executable instructions that, when executed by the one or more processors, cause the computer system to:
receive, from an autonomous vehicle-mounted transceiver, information regarding capabilities and features of an autonomous system of an autonomous vehicle;
receive, from the autonomous vehicle-mounted transceiver, (i) autonomous system sensor data or vehicle-mounted sensor data collected or generated at a time immediately prior to a vehicle collision, and (ii) control signal data indicating both (a) one or more control decisions generated and implemented to control the autonomous vehicle by the autonomous system immediately prior to the vehicle collision instead of one or more unimplemented control decisions and (b) the one or more unimplemented control decisions generated by the autonomous system for controlling an aspect of operation of the autonomous vehicle immediately prior to the vehicle collision but not implemented to control the autonomous vehicle;
determine, by the one or more processors applying a machine learning model trained using (i) previous autonomous system sensor data, (ii) driver data, or (iii) vehicle environmental data, a risk score for each of the one or more unimplemented control decisions based upon the autonomous system sensor data or the vehicle-mounted sensor data;
determine, by the one or more processors, whether the one or more control decisions made and implemented by the autonomous system prior to the vehicle collision were preferred control decisions by applying the machine learning model to (1) the information regarding the capabilities and features of the autonomous system of the autonomous vehicle, (2) the autonomous system sensor data or the vehicle-mounted sensor data collected or generated at the time immediately prior to the vehicle collision, and (3) the risk score for each of the one or more unimplemented control decisions; and
assign, by the one or more processors, a percentage of fault of the vehicle collision to the autonomous system based upon whether or not the one or more control decisions made by the autonomous system immediately prior to the vehicle collision were preferred control decisions.

2 . The computer system of claim 1 , wherein the executable instructions further cause the computer system to adjust a risk level or the machine learning model for the autonomous vehicle or autonomous system based upon the one or more control decisions made by the autonomous system immediately prior to the vehicle collision.

3 . (canceled)

4 . The computer system of claim 1 , wherein the one or more unimplemented control decisions include an alternative control decision not selected by the autonomous system to control the autonomous vehicle.

5 . The computer system of claim 1 , wherein the one or more unimplemented control decisions include a control decision not implemented because an autonomous operation feature associated with the control decisions was disabled.

6 . The computer system of claim 1 , wherein the control signal data is entered into a log of operating data that includes an indication of a reason why the one or more control decisions or the one or more unimplemented control decisions were executed or not executed by the autonomous system.

7 . The computer system of claim 6 , wherein the reason why the one or more unimplemented control decisions were not executed by the autonomous system was (i) that software of the autonomous system was corrupted, or (ii) a sensor of the autonomous system was not working properly.

8 . The computer system of claim 6 , wherein the reason why the one or more unimplemented control decisions were not executed by the autonomous system was that the autonomous system was overridden by a human driver.

9 . The computer system of claim 6 , wherein external data is also entered into the log of operating data, the external data including information regarding a plurality of the following: road conditions, weather conditions, nearby traffic conditions, road type, construction conditions, presence of pedestrians, or presence of other obstacles.

10 . The computer system of claim 1 , wherein the one or more control decisions made by the autonomous system immediately prior to the vehicle collision includes a control decision implemented by the autonomous system to change lanes or to turn the autonomous vehicle.

11 . The computer system of claim 1 , wherein the one or more control decisions made by the autonomous system immediately prior to the vehicle collision includes a control decision implemented by the autonomous system to cause the autonomous vehicle to accelerate or decelerate, and wherein the control decision includes a rate of acceleration or deceleration.

12 . The computer system of claim 1 , wherein the one or more control decisions made by the autonomous system immediately prior to the vehicle collision includes a control decision implemented by the autonomous system to cause the autonomous vehicle to brake, and wherein the control decision includes an amount of force or pressure applied to brakes of the autonomous vehicle.

13 . A tangible, non-transitory computer-readable medium storing executable instructions for monitoring an autonomous vehicle having one or more autonomous operation features for controlling the autonomous vehicle that, when executed by at least one processor of a computer system, cause the computer system to:
receive, from an autonomous vehicle-mounted transceiver, information regarding capabilities and features of an autonomous system of the autonomous vehicle; receive, from the autonomous vehicle-mounted transceiver, (i) autonomous system sensor data or vehicle-mounted sensor data collected or generated at a time immediately prior to a vehicle collision, and (ii) control signal data indicating both (a) one or more control decisions generated and implemented to control the autonomous vehicle by the autonomous system immediately prior to the vehicle collision instead of one or more unimplemented control decisions and (b) the one or more unimplemented control decisions generated by the autonomous system for controlling an aspect of operation of the autonomous vehicle immediately prior to the vehicle collision but not implemented to control the autonomous vehicle; determine, by the at least one processor applying a machine learning model trained using (i) previous autonomous system sensor data, (ii) driver data, or (iii) vehicle environmental data, a risk score for each of the one or more unimplemented control decisions based upon the autonomous system data or the vehicle-mounted sensor data; determine, by the at least one processor, whether the one or more control decisions made and implemented by the autonomous system prior to the vehicle collision were preferred control decisions by applying the machine learning model to (1) the information regarding the capabilities and features of the autonomous system of the autonomous vehicle, (2) the autonomous system sensor data or vehicle-mounted sensor data collected or generated at the time immediately prior to the vehicle collision, and (3) the risk score for each of the one or more unimplemented control decisions; and assign, by the at least one processor, a percentage of fault of the vehicle collision to the autonomous system based upon whether or not the one or more control decisions made by the autonomous system immediately prior to the vehicle collision were preferred control decisions.

Assignees

• State Farm Mutual Automobile Insurance Co

Inventors

• Konrardy Blake
• Christensen Scott T
• Hayward Gregory
• Farris Scott

Classifications

• B60W60/005

  Handover processes (Handing over between remote control and on-board control or handing over between remote control arrangements G05D1/227) · CPC title

• G07C5/008 Primary

  communicating information to a remotely located station (transmission systems for measured values G08C) · CPC title

• G07C5/02

  Registering or indicating driving, working, idle, or waiting time only (apparatus forming part of taximeters G07B13/00) · CPC title

• G06Q40/08

  Insurance · CPC title

• G05D1/0088

  characterized by the autonomous decision making process, e.g. artificial intelligence, predefined behaviours (using knowledge based models G06N5/00) · CPC title