Real-time vehicle accident prediction, warning, and prevention

Patent metadata

Field	Value
Publication number	US-2020148200-A1
Application number	US-201816186098-A
Country	US
Kind code	A1
Filing date	Nov 9, 2018
Priority date	Nov 9, 2018
Publication date	May 14, 2020
Grant date	—

Abstract

A vehicle and a method for the vehicle to detect a vehicle accident are provided. The method includes accessing at least one accident profile, wherein each accident profile contains information describing, for vehicles involved in a vehicle accident, data indicating vehicle operating parameters during the accident; collecting, in real time, data from the vehicle and at least one nearby vehicle, the data describing current vehicle operating parameters for the vehicles; comparing the data collected in real time with the at least one accident profile to determine a probability of a vehicle crash; and implementing preventive action responsive to the probability of a vehicle crash exceeds a probability threshold.

First claim

What is claimed is:

1 . A method for a vehicle to detect a vehicle accident, the method comprising:
accessing at least one accident profile, wherein each accident profile contains information describing, for vehicles involved in a vehicle accident, data indicating vehicle operating parameters during the accident; collecting, in real time, data from the vehicle and at least one nearby vehicle, the data describing current vehicle operating parameters for the vehicles; comparing the data collected in real time with the at least one accident profile to determine a probability of a vehicle crash; and implementing preventive action responsive to the probability of a vehicle crash exceeds a probability threshold.

2 . The method of claim 1 , wherein implementing preventive action responsive to the probability of a vehicle crash exceeds a probability threshold comprises at least one of:
generating an alert for the occupants of the vehicle; and controlling one or more of the operational systems of the vehicle to prevent the vehicle crash.

3 . The method of claim 1 , wherein controlling one or more of the operational systems of the vehicle to prevent the vehicle crash comprises at least one of:
slowing the vehicle; altering a path of the vehicle; and stopping the vehicle.

4 . The method of claim 1 , further comprising:
identifying the at least one nearby vehicle; accessing a vehicle profile for the at least one nearby vehicle, wherein each vehicle profile represents a history of the respective vehicle; and comparing the vehicle profile and the data collected in real time with the at least one accident profile to determine the probability of the vehicle crash.

5 . The method of claim 1 , further comprising:
identifying a driver of the at least one nearby vehicle; accessing at least one driver profile for each driver, wherein each driver profile represents a history of the respective driver; and comparing the at least one driver profile and the data collected in real time with the at least one accident profile to determine the probability of the vehicle crash.

6 . The method of claim 1 , further comprising:
collecting, in real time, from the vehicle, data describing environment parameters of the vehicle; comparing the data collected in real time with the at least one accident profile to determine the probability of the vehicle crash.

7 . The method of claim 1 , further comprising generating the one or more accident profiles, the generating comprising:
for each of a plurality of accidents, receiving and storing sensor data from a plurality of vehicle sensors indicating values of vehicle operating parameters during each accident; applying a machine learning algorithm to the data to identify the vehicle operating parameters and values that are indicative of the accident; and generating a profile representing the identified vehicle operating parameters and values.

8 . A vehicle comprising:
a processor; and a non-transitory machine-readable storage medium encoded with instructions executable by the processor, the machine-readable storage medium comprising instructions to cause the processor to perform a method comprising:
accessing at least one accident profile, wherein each accident profile contains information describing, for vehicles involved in a vehicle accident, data indicating vehicle operating parameters during the accident;
collecting, in real time, data from the vehicle and at least one nearby vehicle, the data describing current vehicle operating parameters for the vehicles;
comparing the data collected in real time with the at least one accident profile to determine a probability of a vehicle crash; and
implementing preventive action responsive to the probability of a vehicle crash exceeds a probability threshold.

9 . The vehicle of claim 8 , wherein implementing preventive action responsive to the probability of a vehicle crash exceeds a probability threshold comprises at least one of:
generating an alert for the occupants of the vehicle; and controlling one or more of the operational systems of the vehicle to prevent the vehicle crash.

10 . The vehicle of claim 8 , wherein controlling one or more of the operational systems of the vehicle to prevent the vehicle crash comprises at least one of:
slowing the vehicle; altering a path of the vehicle; and stopping the vehicle.

11 . The vehicle of claim 8 , wherein the method further comprises:
identifying the at least one nearby vehicle; accessing a vehicle profile for the at least one nearby vehicle, wherein each vehicle profile represents a history of the respective vehicle; and comparing the vehicle profile and the data collected in real time with the at least one accident profile to determine the probability of the vehicle crash.

12 . The vehicle of claim 8 , wherein the method further comprises:
identifying a driver of the at least one nearby vehicle; accessing at least one driver profile for each driver, wherein each driver profile represents a history of the respective driver; and comparing the at least one driver profile and the data collected in real time with the at least one accident profile to determine the probability of the vehicle crash.

13 . The vehicle of claim 8 , wherein the method further comprises:
collecting, in real time, from the vehicle, data describing environment parameters of the vehicle; and comparing the data collected in real time with the at least one accident profile to determine the probability of the vehicle crash.

14 . A non-transitory machine-readable storage medium encoded with instructions executable by a hardware processor of a computing component of a vehicle, the machine-readable storage medium comprising instructions to cause the hardware processor to perform a method comprising:
accessing at least one accident profile, wherein each accident profile contains information describing, for vehicles involved in a vehicle accident, data indicating vehicle operating parameters during the accident; collecting, in real time, data from the vehicle and at least one nearby vehicle, the data describing current vehicle operating parameters for the vehicles; comparing the data collected in real time with the at least one accident profile to determine a probability of a vehicle crash; and implementing preventive action responsive to the probability of a vehicle crash exceeds a probability threshold.

15 . The medium of claim 14 , wherein implementing preventive action responsive to the probability of a vehicle crash exceeds a probability threshold comprises at least one of:
generating an alert for the occupants of the vehicle; and controlling one or more of the operational systems of the vehicle to prevent the vehicle crash.

16 . The medium of claim 14 , wherein controlling one or more of the operational systems of the vehicle to prevent the vehicle crash comprises at least one of:
slowing the vehicle; altering a path of the vehicle; and stopping the vehicle.

17 . The medium of claim 14 , wherein the method further comprises:
identifying the at least one nearby vehicle; accessing a vehicle profile for the at least one nearby vehicle, wherein each vehicle profile represents a history of the respective vehicle; and comparing the vehicle profile and the data collected in real time with the at least one accident profile to determine the probability of the vehicle crash.

18 . The medium of claim 14 , wherein the method further comprises:
identifying a driver of the at leas

Assignees

• Toyota Motor North America Inc

Inventors

• Lerner Emily

Classifications

• G08G1/166

  for active traffic, e.g. moving vehicles, pedestrians, bikes · CPC title

• G06N20/00 Primary

  Machine learning · CPC title

• B60W30/09 Primary

  Taking automatic action to avoid collision, e.g. braking and steering · CPC title

• G08G1/162

  event-triggered · CPC title

• B60W30/0956

  the prediction being responsive to traffic or environmental parameters · CPC title