System and method for reducing driving risk with hindsight

Patent metadata

Field	Value
Publication number	US-9836716-B2
Application number	US-38232506-A
Country	US
Kind code	B2
Filing date	May 9, 2006
Priority date	May 9, 2006
Publication date	Dec 5, 2017
Grant date	Dec 5, 2017

Abstract

A system and method for computer assisted event based reconstruction and forensic analysis of vehicle accidents is provided. The system comprises an event capture device that records audio, video, and other information that collectively comprise one or more events related to a vehicle accident. The event data, including the audio, video, and other related information, is provided to an evaluation server where it is stored in a database for forensic analysis of the events comprising the vehicle accident. Event data for a specific automobile accident event is analyzed and compared to similar types of automobile accident event data in order to forensically analyze the specific event and correlate causal relationships with key elements of the event data to determine the likely cause of the accident or the factors contributing to the accident. Correlation information is also stored at the evaluation server to provide historical data points about causal relationships and key elements.

First claim

The invention claimed is:

1. A computer-implemented method to generate a report of a vehicle accident, comprising:
capturing driving event data using an event detector in communication with an event capture device coupled to a vehicle, wherein the driving event data is associated with a vehicle accident, wherein the driving event data is selectively captured in response to a trigger by the event detector and comprises at least video data, audio data, metadata fields, and vehicle status information from before, during, and after the vehicle accident;
uploading the driving event data to an evaluation server;
receiving driving event data at the evaluation server from the event detector;
identifying one or more elements from the driving event data, wherein the one or more elements comprise individual data points from the driving event data, wherein identifying the one or more elements comprises:
performing a computer analysis of one of the video data for indicators that point to vehicle accident elements,
performing a computer analysis of one of the audio data for indicators that point to vehicle accident elements, comprising:
determining whether decibel level on the audio data indicates an accident has occurred, the audio data being recorded inside the vehicle,

analyzing metadata fields of the driving event data for temperature conditions as one of the one or more elements,
analyzing the driving data for vehicle status information as one of the one or more elements, wherein vehicle status information comprises speed and acceleration data before and during the accident, and
flagging elements determined by the computer analysis to be vehicle accident elements;

receiving historical information from a database, wherein the historical information comprises prior captured driving event data elements, and a plurality of correlations that describe how often a particular element of the prior captured driving event data elements is an accident causal factor;
correlating the one or more elements to the prior captured driving event data elements and the plurality of correlations;
determining one or more factors that contributed to a cause of the vehicle accident, comprising:
determining a first correlation relating to how often a first element of the prior captured driving event data elements is a first accident causal factor;
determining a second correlation relating to how often a second element of the prior captured driving event data elements is a second accident causal factor;
determining a third correlation relating to how often a combination of the first element and the second element is a third accident causal factor;
ranking the plurality of elements and combinations of elements, wherein in the event that the third correlation is greater than the first correlation and the third correlation is greater than the second correlation, the combination of the first element and the second element is ranked higher than the first element alone and the combination of the first element and the second element is ranked higher than the second element alone; and
determining the one or more factors that contributed to the cause of the vehicle accident based on the ranked plurality of elements and the ranked combinations of elements;

evaluating a group of events relating to a driver or a group of drivers based on an analysis of the driving event data;
creating a coaching session for the driver or the group of drivers based on the group of events;
providing the coaching session to the driver or the group of drivers to avoid the one or more factors that contribute to causes of vehicle accidents;
generating an interactive multimedia report of the vehicle accident, wherein the report includes (i) the determined one or more factors that contributed to the cause of the vehicle accident and (ii) a user interface to step through a reconstruction of at least one stage of the vehicle accident, wherein the step through includes at least the video data corresponding to a respective stage of the vehicle accident;
outputting the report; and
updating the historical information in the database, wherein the historical information is updated with the one or more elements, correlations, and the one or more factors, wherein updating the historical information in the database allows later forensic analyses to benefit from aggregated historical information.

2. A system to generate a report of a vehicle accident, comprising:
an event detector in communication with an event capture device coupled to a vehicle, wherein the event detector comprises a processor configured to:
capture driving event data, wherein the driving event data is associated with a vehicle accident, wherein the driving event data is selectively captured in response to a trigger by the event detector and comprises at least video data, audio data, metadata fields, and vehicle status information from before, during, and after the vehicle accident;
upload the driving event data to an evaluation server;
the evaluation server comprising one or more processors configured to:
receive driving event data at the evaluation server from the event detector;
identify one or more elements from the driving event data, wherein the one or more elements comprise individual data points from the driving event data, wherein the identifying of the one or more elements comprises:
perform a computer analysis of one of the video data for indicators that point to vehicle accident elements,
perform a computer analysis of one of the audio data for indicators that point to vehicle accident elements, comprising to:
determine whether decibel level on the audio data indicates an accident has occurred, the audio data being recorded inside the vehicle,

analyze metadata fields of the driving event data for temperature conditions as one of the one or more elements,
analyze the driving data for vehicle status information as one of the one or more elements, wherein vehicle status information comprises speed and acceleration data before and during the accident, and
flag elements determined by the computer analysis to be vehicle accident elements;

evaluate a group of events relating to a driver or a group of drivers based on an analysis of the driving event data;
create a coaching session for the driver or the group of drivers based on the group of events;
provide the coaching session to the driver or the group of drivers to avoid the one or more factors that contribute to causes of vehicle accidents;
receive historical information from a database, wherein the historical information comprises prior captured driving event data elements, and a plurality of correlations that describe how often a particular element of the prior captured driving event data elements is an accident causal factor;
correlate the one or more elements to the prior captured driving event data elements and the plurality of correlations;
determine one or more factors that contributed to a cause of the vehicle accident, comprising:
determine a first correlation relating to how often a first element of the prior captured driving event data elements is an accident causal factor;
determine a second correlation relating to how often a second element of the prior captured driving event data elements is an accident causal factor;
determine a third correlation relating to how often a combination of the first element and the second element is an accident causal factor;
rank the plurality of elements and combinations of elements, wherein in the event that the third correlation is greater than the first correlation and the third correlation is greater than the second correlation, the combination of the first element and the second element is ranked higher than the first element alone and the combination of the first element and the second element is ranked higher than the second el

Assignees

• Gunderson Charlie
• Lafleur Tom
• Lurie Doron
• Moeller Bruce
• Lytx Inc

Inventors

• Gunderson Charlie
• Lafleur Tom
• Lurie Doron
• Moeller Bruce

Classifications

• G07C5/0858

  wherein the data carrier is removable · CPC title

• G06Q10/087 Primary

  Inventory or stock management, e.g. order filling, procurement or balancing against orders · CPC title

• G07C5/0891

  Video recorder in combination with video camera · CPC title