Characterizing a vehicle collision
Described herein are examples of a system that processes information describing movement of a vehicle at a time related to a potential collision to reliably determine whether a collision occurred and/or one or more characteristics of the collision. In response to obtaining information regarding a potential collision, data describing movement of the vehicle before and/or after a time associated with the potential collision is analyzed to determine whether the collision occurred and/or to determine one or more collision characteristic(s). The analysis may be carried out at least in part using a trained classifier that classifies the vehicle movement data into one or more classes, where at least some the classes are associated with whether a collision occurred and/or one or more characteristics of a collision. If a collision is determined to be likely, one or more actions may be triggered based on the characteristic(s) of the collision.
1. At least one non-transitory computer-readable storage medium having encoded thereon executable instructions that, when executed by at least one processor, cause the at least one processor to carry out a method comprising:
receiving vehicle movement data from a device installed on a vehicle, the vehicle movement data describing operation of the vehicle during a time period surrounding a potential collision of the vehicle;
determining whether the vehicle experienced a collision and, in a case that the vehicle experienced a collision, one or more attributes of the collision, wherein the determining comprises analyzing at least the vehicle movement data using at least one classifier trained to identify types of collisions, and wherein the one or more attributes of the collision comprise an angle of impact;
in response to determining that the vehicle experienced the collision, sending notification of the one or more attributes of the collision to a remote device.
2. The at least one non-transitory computer-readable storage medium of claim 1 , wherein the one or more attributes of the collision comprise a forward-backward collision and/or a broadside collision.
3. The at least one non-transitory computer-readable storage medium of claim 2 , wherein the types of collisions that can be identified by the at least one classifier comprise a no-collision type, and wherein determining whether the vehicle experienced a collision comprises:
determining there is a match between the vehicle movement data and the no-collision type; and
in response to determining the match, determining the vehicle did not experience a collision.
4. The at least one non-transitory computer-readable storage medium of claim 3 , wherein determining there is a match between the vehicle movement data and the no-collision type comprises:
determining a probability indicating a level of match between the vehicle movement data and the non-collision type; and
in response to determining that the probability exceeds a threshold, determining that there is a match between the vehicle movement data and the no-collision type.
5. The at least one non-transitory computer-readable storage medium of claim 1 , wherein the one or more attributes of the collision comprise a severity of collision.
6. The at least one non-transitory computer-readable storage medium of claim 5 , wherein the method further comprises:
in response to determining that the one or more attributes of the collision comprise at least a severe collision, triggering a dispatch of a type of service, wherein the type of service is based on the angle of impact.
7. The at least one non-transitory computer-readable storage medium of claim 2 , wherein the one or more attributes of the collision comprise a front-end collision and/or a rear-end collision, and wherein the method further comprises:
in response to determining that the one or more attributes of the collision comprise at least one of a front-end collision and a rear-end collision, determining a liability of the potential collision.
8. The at least one non-transitory computer-readable storage medium of claim 2 , wherein:
the vehicle movement data comprises forward and backward movement data and right-left movement data; and
determining the one or more attributes of the collision comprises:
for a respective type of the types of collisions that can be identified by the at least one classifier, calculating a probability indicating a level of match between the respective type and the vehicle movement data; and
determining the one or more attributes of the collision based on the probabilities for the types of collision that can be identified by the at least one classifier.