Exhaustive driving analytical systems and modelers
AssigneeAllstate Insurance Company
InventorsAnuradha Kodali, Sunil Chintakindi, Regina Madigan
Life of this patent
Estimating the term…
Abstract
Exhaustive driving analytical methods, systems, are apparatuses are described. The methods, systems, are apparatuses relate to utilizing partially available data associated with driver and/or driving behaviors to determine safety factors, identify times to react to events, and contextual information regarding the events. The methods, systems, and apparatuses described herein may determine, based on a systematic model, reactions and reaction times, compare the vehicle behavior (or lack thereof) to the modeled reactions and reaction times, and determine safety factors and instructions based on the comparison.
Claims20
A method comprising:
- receiving data from one or more sensors associated with a vehicle;
- identifying a vehicle event based on the data;
- identifying driver conditions based on the data;
- determining, based on the vehicle event, an expected performance of the vehicle and an estimated time to perform the expected performance of the vehicle;
- determining a vehicle reaction to the vehicle event using the identified driver conditions;
- determining a reaction time between identifying the vehicle event and the vehicle reaction;
- comparing the reaction time to the estimated time to perform the expected performance;
- determining a safety factor associated with a driver of the vehicle, the safety factor based on a driving error model, the data, the driver conditions, the vehicle event, and the comparison of the reaction time to the estimated time;
- and
- outputting a vehicle instruction based on the safety factor,
- wherein the vehicle instruction controls the vehicle to switch to an autonomous driving mode.
A system comprising:
- at least one processor;
- and
- memory storing instructions that, when executed by the at least one processor, cause at least one processor to:
- receive data from at least one sensor;
- identify a vehicle event based on the data;
- determine a driver condition based on the data;
- determine, based on the vehicle event, an expected performance of a vehicle and an estimated time to perform the expected performance of the vehicle;
- determine a vehicle reaction to the vehicle event using the driver condition;
- determine a reaction time between identifying the vehicle event and vehicle reaction;
- compare the reaction time to the estimated time;
- determine a safety factor associated with a driver of the vehicle, the safety factor determined using a driving error model, the data, the driver condition, the vehicle event, and the comparison of the reaction time to the estimated time;
- and
- outputting a vehicle instruction based on the safety factor,
- wherein the vehicle instruction controls the vehicle to change to an autonomous driving mode.
A non-transitory computer readable medium storing computer-executable instructions that, when executed, cause at least one processor to:
- receive data from one or more sensors associated with a vehicle;
- identify a vehicle event associated with the vehicle using the data;
- identify one or more driver conditions using the data;
- determine, based on the vehicle event, an expected performance of the vehicle and an estimated time to perform the expected performance of the vehicle;
- determine a vehicle reaction to the vehicle event using the driver conditions;
- determine a reaction time between identifying the vehicle event and the vehicle reaction;
- compare the reaction time to the estimated time to perform the expected performance;
- determine a safety factor associated with a driver of the vehicle, the safety factor based on a driving error model, the data, the one or more driver conditions, the event, and on the comparison of the reaction time to the estimated time;
- and
- output a vehicle instruction based on the safety factor,
- wherein the vehicle instruction controls the vehicle to switch to an autonomous driving mode.
Dependencies are inferred from the claim text.