Patent Issued for Systems and methods for modeling telematics, positioning, and environmental data (USPTO 11790458): State Farm Mutual Automobile Insurance Company

2023 NOV 03 (NewsRx) -- By a News Reporter-Staff News Editor at Insurance Daily News -- According to news reporting originating from Alexandria, Virginia, by NewsRx journalists, a patent by the inventors Bracero, Bernardo (Bloomington, IL, US), Fogg, Jeremy Shawn (Bloomington, IL, US), Gozur, Stan E. (Bloomington, IL, US), Menon, Sunish (Normal, IL, US), Moon, Phillip Sangpil (Bloomington, IL, US), Stoiber, Jeffrey Wilson (Bloomington, IL, US), Wu, Weixin (Normal, IL, US), filed on April 22, 2021, was published online on October 17, 2023.

The assignee for this patent, patent number 11790458, is State Farm Mutual Automobile Insurance Company (Bloomington, Illinois, United States).

Reporters obtained the following quote from the background information supplied by the inventors: “Vehicle insurance provides financial protection against physical damage and/or bodily injury caused by a vehicular accident. Other financial protections may be provided, such as vehicle theft protection or weather-related damage protection. Conventionally, vehicle insurance rates or premiums may be typically determined based upon a driver’s age and driving history, a vehicle make, model, and year, among a myriad of other factors.

“Some insurance policies (e.g., vehicle insurance, rental insurance, homeowners insurance, and/or property insurance) provide coverage for loss or damage to personal possessions of a policyholder during a policy claim (e.g., a formal request by the policyholder to an insurance provider for reimbursement for one or more personal possessions covered under an insurance policy). Loss events may include vehicle damage, residential fires, theft, vandalism and/or other events that cause partial or complete loss of the personal possessions of the policyholder.

“Policy coverage may typically be associated with the amount of risk or liability that is covered by the insurance provider for the policyholder’s possessions during these loss events. Insurance providers may typically set policy premiums based upon a number of factors including an amount of coverage that the policy provides (e.g., policy coverage or insurance coverage). An insurance policy may have different limits, such as coverage limits (e.g., limits of liability) and aggregate limits. Different types of insurance policies limits may typically include payout limits to a policy holder with respect to payouts over time, the maximum amount the insurer will pay, or a combination thereof.

“Insurance premiums and coverage rates may depend on, at least in part, coverage limits or limits of liability, also referred to as liability limits. At least some applications may benefit from accurately predicting the likelihood of insurance claims being made by policyholders. In such applications, insurance claim costs may be anticipated. Further, based upon the likelihood of insurance claims being made and their respective costs, insurance policy premium prices may be determined appropriately.

“However, current solutions may lack the ability to provide accurate predictions of liability limits for users. Current solutions may also be inefficient, cumbersome, untimely, burdensome, and/or have other drawbacks.”

In addition to obtaining background information on this patent, NewsRx editors also obtained the inventors’ summary information for this patent: “The present embodiments may relate to, inter alia, systems and methods for building a model to analyze collected data. The model may be built using historical data (e.g., historical user data and/or historical liability limit data) to analyze collected data including telematics, positioning, and/or environmental data. In some embodiments, the model may use the historical data to relate historical liability limit data to historical user data (e.g., personal data including filed claim data and natural loss data and telematics, positioning, and/or environmental data). Accordingly, the collected data may be input into the model to determine a liability limit for a user associated with the collected data. In some embodiments, the liability limit may be used to generate an insurance policy for the user, and the liability limit may be associated with a maximum amount for which an insurance company associated with the insurance policy is accountable.

“In an exemplary embodiment, the model may be created through the gathering of established user data records and historical data associated with a plurality of users. The user data records may include user driving history and insurance data (e.g., claims data, premium cost data, etc.). User data and historical data may also include positional data and/or telematics data reported from one or more measurement sensory devices, such as a GPS device, an accelerometer, a gyroscope, or other sensors mounted within user computing devices (e.g., mobile devices or tablets) or integrated into vehicles operated by the users. Historical user data may also include environmental data associated with the users or a surrounding area of the users (e.g., traffic data, pedestrian data, etc.). The model may be built by relating one or more sets of the historical data. In some embodiments, the model may be built by relating one or more of the historical vehicle positional, telematics data, and/or environmental data with the historical insurance data. For example, the model may be used to determine and/or predict an insurance liability amount based upon historical position, telematics, and/or environmental data.

“In another exemplary embodiment, systems and methods may provide feedback to users with respect to driving conditions, intersections, or the like. For example, a user may be provided with feedback with respect to their traveling speed when operating a vehicle in view of a posted speed limit. In another example, a user may be provided with data pertaining to a planned driving route. Data with respect to a certain route may indicate the number of traffic incidents that have occurred along the route over a certain time period (e.g., the past six months). Additionally or alternatively, users may be notified of dangerous areas (e.g., intersections) along a certain route. An optimal route may be suggested that is considered to be the lowest risk, or safest route. The optimal route may be determined using a combination of location data, historical telematics data, among other factors, such as weather data.

“In one aspect, a modeling computing device including at least one processor in communication with a memory device may be provided. The at least one processor may be configured to: (i) retrieve, from the at least one memory device, historical data associated with a plurality of users, wherein the historical data includes historical liability limit data and historical user data, and wherein the historical user data includes at least one of historical personal information, historical vehicle telematics data, and historical environmental data, (ii) generate a model that relates the historical liability limits data and the historical user data, (iii) store the model in the at least one memory device, (iv) collect current user data associated with a candidate user, wherein the current user data includes current personal information, current vehicle telematics data, and current environmental data, and/or (v) analyze the collected current user data using the generated model. The modeling computing device may include additional, less, or alternate actions, including those discussed elsewhere herein.

“In another aspect, a computer-implemented method implemented by a modeling computing device including at least one processor in communication with at least one memory device may be provided. The computer-implemented method may include (i) retrieving, from the at least one memory device, historical data associated with a plurality of users, wherein the historical data includes historical liability limit data and historical user data, and wherein the historical user data includes at least one of historical personal information, historical vehicle telematics data, and historical environmental data, (ii) generating a model that relates the historical liability limits data and the historical user data, (iii) storing the model in the at least one memory device, (iv) collecting current user data associated with a candidate user, wherein the current user data includes current personal information, current vehicle telematics data, and current environmental data, and/or (v) analyzing the collected current user data using the generated model. The computer-implemented method may include additional, less, or alternate actions, including those discussed elsewhere herein.”

The claims supplied by the inventors are:

“1. A modeling computing device comprising at least one processor in communication with at least one memory device, the at least one processor configured to: retrieve, from the at least one memory device, historical data associated with a plurality of users, wherein the historical data includes historical liability amount data and historical user data, and wherein the historical user data includes at least one of historical personal information, historical vehicle telematics data, and historical environmental data; create a first plurality of training datasets including the historical liability amount data and the historical user data; build, using one or more machine learning programs, a model based upon the first plurality of training datasets; store the model in the at least one memory device; collect current user data of a candidate user associated with a vehicle, wherein the current user data includes current personal information, current vehicle telematics data, and current environmental data, wherein the current vehicle telematics data and the current environmental data are gathered by one or more sensors during operation of the vehicle, and wherein the one or more sensors include a GPS device, an accelerometer, a gyroscope, a camera, and a sensor installed within the vehicle or located remotely from the vehicle; create a second plurality of training datasets by updating the first plurality of training datasets to include the collected current user data; update the model by applying the second plurality of training datasets to the model; and execute the updated model to determine a current liability amount for the candidate user.

“2. The modeling computing device of claim 1, wherein the model is configured to predict a liability amount based upon the historical user data, wherein the liability amount includes a liability limit, a liability loss amount, and a claim amount, and wherein the at least one processor is further configured to: apply the updated model to further collected current user data to determine a further current liability amount for the candidate user; and further update the updated model based upon the determined further current liability amount.

“3. The modeling computing device of claim 1, wherein the at least one processor is further configured to: transmit the determined current liability amount to at least one third party, wherein the at least one third party includes an insurance company; and generate an insurance policy for the insurance company based upon the determined current liability amount.

“4. The modeling computing device of claim 1, wherein the one or more machine learning programs include machine learning, artificial intelligence, or a combination thereof, and wherein the at least one processor is further configured to build the first plurality of training datasets using the historical data associated with the plurality of users, the historical data including historical insurance data.

“5. The modeling computing device of claim 1, wherein: the historical vehicle telematics data comprises one or more historical measurements collected during historical operation of the vehicle including historical velocity, acceleration, direction, and driver behavior characteristics; and the current vehicle telematics data comprises one or more current measurements collected during current operation of the vehicle including current velocity, acceleration, direction, and driver behavior characteristics.

“6. The modeling computing device of claim 1, wherein the historical environmental data includes at least one of past traffic data and past pedestrian data; and the current environmental data includes at least one of current traffic data and current pedestrian data.

“7. A computer-implemented method implemented by a modeling computing device including at least one processor in communication with at least one memory device, the computer-implemented method comprising: retrieving, from the at least one memory device, historical data associated with a plurality of users, wherein the historical data includes historical liability amount data and historical user data, and wherein the historical user data includes at least one of historical personal information, historical vehicle telematics data, and historical environmental data; creating, by the at least one processor, a first plurality of training datasets including the historical liability amount data and the historical user data; building, by the at least one processor using one or more machine learning programs, a model based upon the first plurality of training datasets; storing the model in the at least one memory device; collecting current user data of a candidate user associated with a vehicle, wherein the current user data includes current personal information, current vehicle telematics data, and current environmental data, wherein the current vehicle telematics data and the current environmental data are gathered by one or more sensors during operation of the vehicle, and wherein the one or more sensors include a GPS device, an accelerometer, a gyroscope, a camera, and a sensor installed within the vehicle or located remotely from the vehicle; creating, by the at least one processor, a second plurality of training datasets by updating the first plurality of training datasets to include the collected current user data; updating the model by applying the second plurality of training datasets to the model; and executing the updated model to determine a current liability amount for the candidate user.

“8. The computer-implemented method of claim 7, wherein the model is configured to predict a liability amount based upon the historical user data, wherein the liability amount includes a liability limit, a liability loss amount, and a claim amount, and wherein the computer-implemented method further comprises: applying the updated model to further collected current user data to determine a further current liability amount for the candidate user; and further updating the updated model based upon the determined further current liability amount.

“9. The computer-implemented method of claim 7 further comprising: transmitting the determined current liability amount to at least one third party, wherein the at least one third party includes an insurance company; and generating an insurance policy for the insurance company based upon the determined current liability amount.

“10. The computer-implemented method of claim 7 wherein the one or more machine learning programs include using machine learning, artificial intelligence, or a combination thereof, and wherein the method further comprises building the first plurality of training datasets using the historical data associated with the plurality of users, the historical data including historical insurance data.

“11. The computer-implemented method of claim 7, wherein: the historical vehicle telematics data comprises one or more historical measurements collected during historical operation of the vehicle including velocity, acceleration, direction, and driver behavior characteristics; and the current vehicle telematics data comprises one or more current measurements collected during current operation of the vehicle including current velocity, acceleration, direction, and driver behavior characteristics.

“12. The computer-implemented method of claim 7, wherein: the historical environmental data includes at least one of past traffic data and past pedestrian data; and the current environmental data includes at least one of current traffic data and current pedestrian data.

“13. At least one non-transitory computer-readable medium having computer-executable instructions embodied thereon, wherein when executed by a modeling computing device including at least one processor in communication with at least one memory device, the computer-executable instructions cause the at least one processor to: retrieve, from the at least one memory device, historical data associated with a plurality of users, wherein the historical data includes historical liability amount data and historical user data, and wherein the historical user data includes at least one of historical personal information, historical vehicle telematics data, and historical environmental data; create a first plurality of training datasets including the historical liability amount data and the historical user data; build, using one or more machine learning programs, a model based upon the first plurality of training datasets; store the model in the at least one memory device; collect current user data of a candidate user associated with a vehicle, wherein the current user data includes current personal information, current vehicle telematics data, and current environmental data, wherein the current vehicle telematics data and the current environmental data are gathered by one or more sensors during operation of the vehicle, and wherein the one or more sensors include a GPS device, an accelerometer, a gyroscope, a camera, and a sensor installed within the vehicle or located remotely from the vehicle; create a second plurality of training datasets by updating the first plurality of training datasets to include the collected current user data; update the model by applying the second plurality of training datasets to the model; and execute the updated model to determine a current liability amount for the candidate user.

“14. The at least one non-transitory computer-readable medium of claim 13, wherein the computer-executable instructions further cause the at least one processor to: apply the updated model to further collected current user data to determine a further current liability amount for the candidate user, wherein the further current liability amount includes a liability limit, a liability loss amount, and a claim amount; and update the model based upon the determined further current liability amount.”

There are additional claims. Please visit full patent to read further.

For more information, see this patent: Bracero, Bernardo. Systems and methods for modeling telematics, positioning, and environmental data. U.S. Patent Number 11790458, filed April 22, 2021, and published online on October 17, 2023. Patent URL (for desktop use only): https://ppubs.uspto.gov/pubwebapp/external.html?q=(11790458)&db=USPAT&type=ids

(Our reports deliver fact-based news of research and discoveries from around the world.)