Method for smartphone-based accident detection

Patent metadata

Field	Value
Publication number	US-10137889-B2
Application number	US-201715727972-A
Country	US
Kind code	B2
Filing date	Oct 9, 2017
Priority date	Aug 20, 2015
Publication date	Nov 27, 2018
Grant date	Nov 27, 2018

Abstract

A method and system for detecting an accident of a vehicle, the method including: receiving a movement dataset collected at least at one of a location sensor and a motion sensor arranged within the vehicle, during a time period of movement of the vehicle, extracting a set of movement features associated with at least one of a position, a velocity, and an acceleration characterizing the movement of the vehicle during the time period, detecting a vehicular accident event from processing the set of movement features with an accident detection model, and in response to detecting the vehicular accident event, automatically initiating an accident response action.

First claim

We claim:

1. A system for determining a vehicular accident associated with first user and a vehicle, the system comprising:
a first application executing on a first device associated with the first user, wherein the first application is operable to collect a first movement dataset sampled from at least one of a first motion sensor and a first location sensor of the first device;
an accident detection processing system operable to:
receive a traffic dataset describing traffic conditions proximal a vehicle location, wherein the vehicle location is extracted from the first movement dataset, wherein the traffic conditions comprise a traffic level, a traffic law, and accident data;
retrieve an accident detection model; and
determine a vehicular accident event in substantially real-time with the accident detection model, the traffic dataset, and the first movement dataset, and based on the traffic level exceeding a traffic level threshold.

2. The system of claim 1 , wherein the accident detection processing system is operable to:
extract a vehicle motion characteristic from the first movement dataset, wherein the vehicle motion characteristic describes movement of the vehicle; and
detect satisfaction of a threshold condition based on the vehicle motion characteristic and the traffic dataset, wherein retrieval of the accident detection model and determination of the vehicular accident event are in response to detection of the satisfaction of the threshold condition.

3. The system of claim 1 , further comprising:
a second application executing on a second device associated with a second user, wherein the second application is operable to collect a second movement dataset sampled from at least one of a second motion sensor and a second location sensor of the second device while the second device is proximal the first device,
wherein the accident detection processing system is operable to determine the vehicular accident event with the accident detection model, the traffic dataset, the first movement dataset, and the second movement dataset.

4. The system of claim 1 , further comprising:
a set of supplementary applications executing on a set of supplementary devices associated with a plurality of users and a plurality of vehicles, wherein the set of supplementary applications is operable to collect a set of supplementary movement datasets sampled at sensors of the set of supplementary devices during driving sessions associated with the plurality of vehicles,
wherein the accident detection model comprises an accident detection machine learning model, and
wherein the accident detection processing system is operable to:
train the accident detection machine learning model based on the set of supplementary movement datasets; and
determine the vehicular accident event with the accident detection machine learning model, the traffic dataset, and the first movement dataset.

5. The system of claim 4 , wherein the accident detection processing system is operable to:
collect a set of supplementary traffic datasets associated with locations of the plurality of vehicles during the driving sessions; and
train the accident detection machine learning model based on the set of supplementary movement datasets and the set of supplementary traffic datasets.

6. The system of claim 1 , wherein the accident detection processing system is operable to: in response to determination of the vehicular accident event, initiate an accident response action associated with the vehicular accident.

7. The system of claim 6 , further comprising: a second application executing on a second device associated with a second user, wherein the accident response action comprises transmission of an accident-related communication to the second device through the second application.

8. The system of claim 7 , wherein the second user is associated with an emergency service, and wherein the accident-related communication comprises location data indicating an accident location of the vehicular accident.

9. The system of claim 1 , wherein the first device comprises the vehicle, wherein the first application is executing on the vehicle, and wherein the accident detection processing system comprises a remote server system operable to receive the first movement dataset from the vehicle, and to determine the vehicular accident event.

10. The system of claim 1 , wherein the first device comprises a mobile computing device, wherein the first application is executing on the mobile computing device, wherein the accident detection processing system comprises the first application, and wherein the first application executing on the mobile computing device is operable to determine the vehicular accident event with the accident detection model, the traffic dataset, and the first movement dataset.

11. The system of claim 10 , wherein the mobile computing device comprises a microphone, wherein the first application is operable to collect an audio dataset sampled at the microphone and comprising audio emitted by the vehicle during a time period associated with the first movement dataset, and wherein the accident detection processing system is operable to determine the vehicular accident event with the accident detection model, the audio dataset, the traffic dataset, and the first movement dataset.

12. The system of claim 11 , wherein the accident detection processing system is operable to transmit at least one of the extracted vehicle location and audio from the audio dataset to an emergency service in response to determination of the vehicular accident event.

13. A method for determining a vehicular accident associated with a user and a vehicle, the method comprising:
collecting a movement dataset corresponding to at least one of a location sensor and a motion sensor of a mobile computing device proximal the vehicle;
extracting a vehicle motion characteristic from the movement dataset, wherein the vehicle motion characteristic describes movement of the vehicle;
receiving a traffic dataset describing traffic conditions proximal a vehicle location, wherein the vehicle location is extracted from the movement dataset, wherein the traffic conditions comprise a traffic level, a traffic law, and accident data;
determining a vehicular accident event in substantially real-time based on an accident detection model, the vehicle motion characteristic, the traffic dataset, and the traffic level exceeding a traffic level threshold; and
after determining the vehicular accident event, facilitating an accident response action associated with the vehicular accident.

14. The method of claim 13 , wherein facilitating the accident response action comprises facilitating the accident response action based on the traffic dataset.

15. The method of claim 13 , wherein the movement dataset is collected during a time period, wherein the method further comprises collecting a biosignal dataset sampled at a biosignal detector proximal the user during the time period, and wherein determining the vehicular accident event comprises determining the vehicular accident event based on the accident detection model, the vehicle motion characteristic, and the biosignal dataset.

16. The method of claim 15 , wherein the mobile computing device comprises the biosignal detector, and wherein facilitating the accident response action comprises facilitating the accident response action based on the biosignal dataset.

17. The method of claim 13 , wherein the vehicle motion characteristic comprises a vehicle speed characteristic, and wherein the method further comprises: in respon

Assignees

• Zendrive Inc

Inventors

• Pal Jayanta
• Islam Bipul
• Choudhury Romit Roy
• Risbood Pankaj
• Matus Jonathan
• Verma Vishal

Classifications

• G06N7/01

  Probabilistic graphical models, e.g. probabilistic networks · CPC title

• G06K2009/00939

  Physics · mapped topic

• B60W30/08 Primary

  Active safety systems} predicting or avoiding probable or impending collision {or attempting to minimise its consequences · CPC title

• G06K9/00664

  Physics · mapped topic

• H04M1/72538

  Electricity · mapped topic