Risk evaluation based on vehicle operator behavior

Patent metadata

Field	Value
Publication number	US-10089692-B1
Application number	US-201514592277-A
Country	US
Kind code	B1
Filing date	Jan 8, 2015
Priority date	Mar 15, 2013
Publication date	Oct 2, 2018
Grant date	Oct 2, 2018

Abstract

A method comprises retrieving data about vehicle operator behavior via a computer network and clustering the data about vehicle operator behavior into a plurality of groups of data, each of the plurality of groups of data representing a type of movement of the vehicle operator. The method further includes determining a numerical level of risk corresponding to each of the plurality of groups of data by executing a learning routine and generating a communication to be transmitted to user of a remote computing device. The communication is based on the one or more of the numerical levels of risk or the types of movements corresponding to the plurality of groups of data.

First claim

We claim:

1. A method comprising:
generating, in a motion sensing device disposed in a vehicle and adapted to monitor movements of a vehicle operator, sensor data;
pre-processing, in an on-board computer of the vehicle, the sensor data to generate data about vehicle operator behavior, wherein pre-processing the sensor data includes identifying a plurality of types of movement of the vehicle operator;
retrieving, by a computing device disposed at a location different from a location of the vehicle, the data about vehicle operator behavior from the vehicle via a computer network;
clustering, by one or more processors of the computing device that are specifically configured by a first routine, the data about vehicle operator behavior into a plurality of groups of data, each of the plurality of groups of data representing a different one of the plurality of types of movement of the vehicle operator;
determining, by the one or more processors, a numerical level of risk corresponding to each of the plurality of groups of data by executing a learning routine specifically configured to determine the numerical level of risk based on correlations between types of movement of vehicle operators and levels of risk;
repeatedly sending, by the one or more processors, an alert to the vehicle operator, wherein the alert is generated based on the one or more of the numerical levels of risk or the types of movements corresponding to the plurality of groups of data, and wherein the alert includes one or more of an audible alert, a visual alert, or a tactile alert;
receiving an alert confirmation from the vehicle operator indicating that the vehicle operator has received the alert; and
in response to receiving the alert confirmation, ceasing sending the alert to the vehicle operator.

2. The method of claim 1 , wherein the learning routine is trained by executing the learning routine with a set of training data about past vehicle operator movements.

3. The method of claim 1 , further comprising generating, by the one or more processors, an insurance rate for an insurance product, wherein the insurance rate is consistent with the numerical levels of risk corresponding to each of the plurality of groups of data.

4. The method of claim 3 , wherein generating the insurance rate for an insurance product consistent with the numerical levels of risk corresponding to each of the plurality of groups of data includes generating a risk index corresponding to the vehicle operator, the risk index being a collective measure of risk based on the numerical levels of risk corresponding to the plurality of groups of data.

5. The method of claim 1 , wherein determining, with one or more processors, the numerical level of risk corresponding to each of the plurality of groups of data includes:
determining, by executing the learning routine with one or more processors, whether at least some of the plurality of groups of data represent a type of behavior associated with relatively high risk in comparison to other types of behavior.

6. The method of claim 5 , wherein determining whether the at least some of the plurality of groups of data represent the type of behavior associated with relatively high risk includes identifying the presence of a particular one of the plurality of groups of data, wherein the presence of the particular ones of the plurality of groups of data within the whole of the data about vehicle operator behavior is associated with relatively high risk.

7. The method of claim 1 , further comprising:
updating training of the learning routine based on the retrieved data about vehicle operator behavior corresponding to the vehicle operator.

8. The method of claim 1 , wherein the plurality of types of movement of the vehicle operator include two or more of:
mobile device use;
vehicle feature adjustments;
eating or drinking;
smoking;
grooming and personal hygiene;
reading a map or navigation device;
reaching for objects inside the vehicle; or
looking at objects or events outside the vehicle.

9. A system comprising:
a motion sensing module disposed in a vehicle and configured to:
monitor movements of a vehicle operator inside of the vehicle, and
generate sensor data indicative of the monitored movements of the vehicle operator;

an on-board computer of the vehicle configured to:
pre-process the sensor data to generate data about vehicle operator behavior, at least by identifying a plurality of types of movement of the vehicle operator;

an interface disposed in the vehicle, the interface connecting the on-board computer to a computing device disposed at a location different from a location of the vehicle; and
the computing device, wherein the computing device is specially configured to:
retrieve the data about vehicle operator behavior from the on-board computer via the interface,
cluster the data about vehicle operator behavior into a plurality of groups of data, each of the plurality of groups of data representing a different one of the plurality of types of movement of the vehicle operator,
determine a numerical level of risk corresponding to each of the plurality of groups of data by executing a learning routine specifically configured to determine the numerical level of risk based on correlations between types of movement of vehicle operators and levels of risk,
repeatedly send an alert to the vehicle operator, wherein the alert is generated based on the one or more of the numerical levels of risk or the types of movements corresponding to the plurality of groups of data, and wherein the alert includes one or more of an audible alert, a visual alert, or a tactile alert,
receive an alert confirmation from the vehicle operator indicating that the vehicle operator has received the alert, and
in response to receiving the alert confirmation, cease sending the alert to the vehicle operator.

10. The system of claim 9 , wherein the computer device is further configured to update training of the learning routine based on the retrieved data about vehicle operator behavior.

11. The system of claim 9 , wherein determining the numerical level of risk corresponding to each of the plurality of groups of data includes determining, by executing the learning routine with one or more processors, whether at least some of the plurality of groups of data represent a type of behavior associated with relatively high risk in comparison to other types of behavior.

Assignees

• State Farm Mutual Automobile Insurance Co
• State Farm Mututal Automobile Insurance Company

Inventors

• Sanchez Kenneth J
• Chan Aaron Scott
• Baker Matthew R
• Zettinger Matthew
• Nepomuceno John A
• Fields Brian Mark

Classifications

• G06Q40/08 Primary

  Insurance · CPC title

• G06Q40/00

  Finance; Insurance; Tax strategies; Processing of corporate or income taxes · CPC title