Back to results
US 12,612,037 B2Granted
Collision prediction based on different types of impact
AssigneeAllstate Insurance Company
InventorLucas Antony
Life of this patent
Estimating the term…
Abstract
Implementations claimed and described herein provide systems and methods for generating a prediction for a likelihood of a collision based on a respective collision prediction algorithm for a particular type of impact. In one implementation, determining, by a collision prediction algorithm, a prediction score based on a subset of variables associated with the movement of a mobile device. The prediction score associated with a likelihood that the movement is associated with a particular type of impact associated with the first collision prediction algorithm. A prediction that the movement is not association with the first type of impact when the prediction score is below a threshold score is outputted.
Claims16
A system comprising:
- one or more processors;
- a display with a user interface;
- and
- a memory unit storing computer-executable instructions, which when executed by the one or more processors, cause the system to:
- store kinematic variables associated with movement of a mobile device that occurred around a time of an event;
- extract a plurality of subsets of the kinematic variables associated respectively with a plurality of time windows that span one or more amounts of time around the time of the event, wherein each time window is associated with a different type of event of a plurality of types of events and a different collision prediction algorithm of a plurality of prediction algorithms, wherein a first subset of the plurality of types of events is associated with a collision, and wherein a second subset of the plurality of types of events is not associated with a collision;
- process, by each of the plurality of collision prediction algorithms a corresponding subset of the kinematic variables, wherein each of the plurality of collision prediction algorithms outputs a prediction score indicative of a likelihood that the movement is associated with a corresponding type of event;
- and
- determine, based on an aggregate of respective prediction scores, whether the movement is associated with a collision.
6 more elements
A computer-implemented method comprising:
- training a first collision prediction algorithm of a plurality of prediction algorithms with a training dataset comprising a dataset from counterfactual collisions where movement was not associated with a first type of event and actual historical instances associated with the first type of event that occurred around a time of an event;
- storing kinematic variables associated with the movement of a mobile device;
- extracting a plurality of subset of the kinematic variables associated respectively with a plurality of time windows that span one or more amounts of time around the time of the event, wherein each time window is associated with a different type of event of a plurality of types of events and a different collision prediction algorithm of a plurality of prediction algorithms, wherein a first subset of the plurality of types of events is associated with a collision, and wherein a second subset of the plurality of types of events is not associated with a collision;
- processing, by each of the plurality of collision prediction algorithms a corresponding subset of the kinematic variables, wherein each of the plurality of collision prediction algorithms outputs a prediction score indicative of a likelihood that the movement is associated with a corresponding type of event;
- and
- determining, based on an aggregate of respective prediction scores, whether the movement is associated with a collision.
3 more elements
A non-transitory computer-readable medium comprising instructions, the instructions, when executed by a computing system, cause the computing system to:
- store global positioning system (GPS) speed variables, GPS altitude variables, and accelerometer magnitude variables associated with movement of a mobile device that occurred around a time of an event;
- extract a plurality of subsets of the global positioning system (GPS) speed variables, the GPS altitude variables, and the accelerometer magnitude variables associated respectively with a plurality of time windows that span one or more amounts of time around the time of the event, wherein each time window is associated with a different type of event of a plurality of types of events and a different collision prediction algorithm of a plurality of prediction algorithms, wherein a first subset of the plurality of types of events is associated with a collision, and wherein a second subset of the plurality of types of events is not associated with a collision;
- process, by each of the plurality of collision prediction algorithms a corresponding subset of the GPS speed variables, GPS altitude variables, and accelerometer magnitude variables, wherein each of the plurality of collision prediction algorithms outputs a prediction score indicative of a likelihood that the movement is associated with a corresponding type of event, wherein the first type of event is associated with the first collision prediction algorithm;
- and
- determine, based on an aggregate of respective prediction scores, whether the movement is associated with a collision.
2 more elements
Dependencies are inferred from the claim text.