US-20260125062-A1 - AUTONOMOUS VEHICLE OPERATION BASED ON REAL-TIME ANALYTICS

US20260125062A1US 20260125062 A1US20260125062 A1US 20260125062A1US-20260125062-A1

Abstract

Systems and methods are disclosed for operating an autonomous vehicle based on real-time operating data. The operating data may be data about vehicles, drivers, passengers, as well as relevant environmental conditions and contextual data. In some cases, historical data for the preceding data types may be used. The systems and methods may obtain a set of real-time operating data indicative of one or more behaviors of an autonomous vehicle. One or more operations may be performed on the set of real-time operating data. An instruction to modify a particular vehicle operation may be generated based on output from the operations and the one or more behaviors of the autonomous vehicle, and the instruction to modify the particular vehicle operation may be provided to a particular processor that is on-board the vehicle and that controls the particular vehicle operation, to thereby automatically modify the particular vehicle operation.

Inventors

Brian Mark Fields

Assignees

STATE FARM MUTUAL AUTOMOBILE INSURANCE COMPANY

Dates

Publication Date: 20260507
Application Date: 20251230

Claims (20)

1 . A method, comprising: receiving, by a processor, first data indicating a weather condition occurring during a portion of a trip taken by a vehicle; selecting, by the processor and based on occurrence of the weather condition during the portion of the trip, a model of a plurality of models, wherein the model is trained to generate a threshold corresponding to the weather condition; generating, by the processor and using the model, the threshold corresponding to the weather condition; determining, by the processor and based on second data indicating an operating parameter exhibited by the vehicle during the portion of the trip, that the operating parameter fails to satisfy the threshold; and providing, by the processor and based on determining that the operating parameter fails to satisfy the threshold, an instruction to an additional processor, the instruction, when executed by the additional processor, causing modification of a vehicle operation corresponding to the operating parameter.
2 . The method of claim 1 , wherein the first data comprises one or more of: data from one or more sensors carried by the vehicle, third-party weather data based on a geographic location of the vehicle, or information from an environmental monitoring application operating on a remote computing system communicatively connected to the processor.
3 . The method of claim 1 , wherein the second data comprises one or more of: sensor data from one or more sensors carried by the vehicle, indicator data comprising a current physical state of a component of the vehicle, or status data indicative of a current status of oil levels, fuel levels, or cabin conditions of the vehicle.
4 . The method of claim 1 , wherein the model is trained on training data comprising historical operating data indicating past behaviors of vehicles operating in conditions substantially similar to the weather condition.
5 . The method of claim 1 , wherein the weather condition is a first weather condition and the portion of the trip is a first portion of the trip, the method further comprising: receiving, by the processor, third data indicating a second weather condition occurring during a second portion of the trip; selecting, by the processor and based on the second weather condition, an additional model of the plurality of models; and generating, by the processor and using the additional model, an additional threshold, different from the threshold, corresponding to the second weather condition, wherein the instruction is provided based on determining that the operating parameter fails to satisfy the additional threshold.
6 . The method of claim 5 , further comprising: determining, by the processor, a vehicle response to a change in weather condition from the first weather condition to the second weather condition; determining, by the processor and based at least in part on the vehicle response, a vehicle performance score; and determining, by the processor and based at least in part on the vehicle performance score, an aspect of an insurance policy associated with the vehicle.
7 . The method of claim 6 , wherein the vehicle response comprises one or more of: a length of time between the change in the weather condition and a reaction by the vehicle, a content of the reaction, or a magnitude of the reaction.
8 . The method of claim 1 , further comprising: determining, by the processor, that the second data indicates an alert condition of one or more pre-determined alert conditions, wherein the instruction causes the modification of the vehicle operation to alleviate the alert condition.
9 . The method of claim 8 , wherein the one or more pre-determined alert conditions are specified by a driver of the vehicle or by an insurance provider of the vehicle.
10 . The method of claim 1 , further comprising: obtaining, by the processor, an indication that a driver of the vehicle has opted-in to automatic modification of vehicle operations, wherein the instruction causes an automatic modification of the vehicle operation based on obtaining the indication.
11 . The method of claim 1 , further comprising: obtaining, by the processor, a set of historical operating data associated with the vehicle; identifying, by the processor and based on the set of historical operating data, one or more alert conditions; and determining, by the processor, that the second data indicates an alert condition of the one or more alert conditions, wherein the instruction causes the modification of the vehicle operation to alleviate the alert condition.
12 . A system, comprising: a processor; a plurality of sensors configured to interface with the processor; and a non-transitory memory storing computer-executable instructions that, when executed, cause the processor to: receive first data indicating a weather condition occurring during a portion of a trip taken by a vehicle; select, based on the weather condition during the portion of the trip, a model of a plurality of models, wherein the model is trained to generate a threshold corresponding to the weather condition; generate, using the model, the threshold corresponding to the weather condition; determine, based on second data indicating an operating parameter exhibited by the vehicle during the portion of the trip, that the operating parameter fails to satisfy the threshold; and provide, based on determining that the operating parameter fails to satisfy the threshold, an instruction to an additional processor, the instruction, when executed by the additional processor, causing modification of a vehicle operation corresponding to the operating parameter.
13 . The system of claim 12 , wherein the instructions further cause the processor to: determine, based on the second data, that the second data indicates an alert condition of one or more pre-determined alert conditions, wherein the instruction causes the modification of the vehicle operation to alleviate the alert condition.
14 . The system of claim 13 , wherein the vehicle operation comprises at least one of: braking, accelerating, steering, lane change, or turning.
15 . The system of claim 12 , wherein the weather condition is a first weather condition and the portion of the trip is a first portion of the trip, the instructions further causing the processor to: receive third data indicating a second weather condition occurring during a second portion of the trip; select, based on the second weather condition, an additional model of the plurality of models; and generate, using the additional model, an additional threshold, different from the threshold, corresponding to the second weather condition, wherein the instruction is provided based on determining that the operating parameter fails to satisfy the additional threshold.
16 . The system of claim 12 , wherein the second data comprises time-series data representing: the operating parameter at each time instance, and the weather condition at the respective time instance.
17 . A tangible, non-transitory computer-readable medium storing executable instructions for operating an autonomous vehicle based on real-time operating data that, when executed by a processor of a system, cause the processor to: receive first data indicating a weather condition occurring during a portion of a trip taken by a vehicle; select, based on the weather condition during the portion of the trip, a model of a plurality of models, wherein the model is trained to generate a threshold corresponding to the weather condition; generate, using the model, a threshold corresponding to the weather condition; determine, based on second data indicating an operating parameter as exhibited by the vehicle during the portion of the trip, that the operating parameter fails to satisfy the threshold; and provide, based on determining that the operating parameter fails to satisfy the threshold, an instruction to an additional processor, the instruction, when executed by the additional processor, causing modification of a vehicle operation corresponding to the operating parameter.
18 . The tangible, non-transitory computer-readable medium of claim 17 , wherein the executable instructions further cause the processor to: obtain a set of historical operating data associated with the vehicle; identify, based on the set of historical operating data, one or more alert conditions; and determine, based on the second data, that the second data indicates an alert condition of the one or more alert conditions, wherein the instruction causes the modification of the vehicle operation to alleviate the alert condition.
19 . The tangible, non-transitory computer-readable medium of claim 17 , wherein the weather condition is a first weather condition and the portion of the trip is a first portion of the trip, and the executable instructions further cause the processor to: receive third data indicating a second weather condition occurring during a second portion of the trip; select, based on the second weather condition, an additional model of the plurality of models; and generate, using the additional model, an additional threshold, different from the threshold, corresponding to the second weather condition, wherein the instruction is provided based on determining that the operating parameter fails to satisfy the additional threshold.
20 . The tangible, non-transitory computer-readable medium of claim 19 , wherein: the model is a context-specific model characterized by a first plurality of weights trained on historical operating data indicating past behaviors of vehicles operating in conditions substantially similar to the first weather condition, and the additional model is characterized by a second plurality of weights, different from the first plurality of weights, trained on historical operating data indicating past behaviors of vehicles operating in conditions substantially similar to the second weather condition.

Description

CROSS REFERENCE TO RELATED APPLICATION This application is a continuation of, and claims priority to U.S. Patent Application Serial No. 18/745,249, filed on June 17, 2024, is a continuation of, and claims priority to U.S. Patent Application Serial No. 15/596,495, filed on May 16, 2017, now U.S. Patent No. 12,013,695, issued June 18, 2024 and entitled “AUTONOMOUS VEHICLE OPERATION BASED ON REAL-TIME ANALYTICS,” the entire contents of which is incorporated herein by reference. TECHNICAL FIELD The present disclosure generally relates to systems and methods for operating an autonomous vehicle based on real-time operating data, in particular, generating and providing instructions to modify particular vehicle operations based on real-time operating data. BACKGROUND Autonomous, or self-driving, cars are becoming more and more common. An autonomous vehicle is a vehicle that exercises varying levels of control over the capabilities of the vehicle. In particular, an autonomous vehicle is capable of sensing its environment and navigating roadways without human interaction. Autonomous vehicles may exercise control of the vehicle with some driver assistance, partially automate the vehicle, conditionally automate the vehicle, engage in a high degree of automating the vehicle, or fully automate the vehicle. Autonomous vehicles still struggle with certain tasks and environments. For example, inclement weather may pose a problem for autonomous vehicles, reckless drivers on the road, changes in road conditions (e.g. detours and rerouted roads), problematic segments of roads (e.g. intersections that are known to be busy), as well as making sense of all the data collected on the road as the vehicle travels. SUMMARY The present disclosure generally relates to systems and methods for operating an autonomous vehicle based on real-time operating data. Embodiments of example systems and methods are summarized below. The methods and systems summarized below may include additional, less, or alternate actions, including those discussed elsewhere herein. In one embodiment, a computer-implemented method for operating an autonomous vehicle based on real-time operating data, the method includes obtaining, at one or more processors, a set of real-time operating data indicative of one or more behaviors of the autonomous vehicle. Performing, at the one or more processors, one or more operations on the set of real-time operating data, wherein the one or more operations are based on the set of real-time operating data. Generating, at the one or more processors, an instruction to modify a particular vehicle operation based on output from the operations and the one or more behaviors of the autonomous vehicle; and providing, by the one or more processors, the instruction to modify the particular vehicle operation to a particular processor that is on-board the vehicle and that controls the particular vehicle operation, to thereby automatically modify the particular vehicle operation. In one embodiment, a computer-implemented method for operating an autonomous vehicle based on real-time operating data, the method includes obtaining, at one or more processors, a set of real-time operating data indicative of one or more behaviors of the autonomous vehicle, wherein the set of real-time operating data obtained from a set of sensors. Performing, at the one or more processors, one or more operations on the set of real-time operating data, wherein the one or more operations are based on the set of real-time operating data. Comparing, at the one or more processors, the output of the one or more operations to a set of vehicle performance data, wherein comparing further comprises identifying alert conditions based on the comparison. Generating, at the one or more processors, an instruction to modify a particular vehicle operation based on the comparison and the one or more behaviors of the autonomous vehicle; and providing, by the one or more processors, the instruction to modify the particular vehicle operation to a particular processor that is at the vehicle and that controls the particular vehicle operation, to thereby automatically modify the particular vehicle operation. In another embodiment, a system for operating an autonomous vehicle based on real-time operating data, the system including a network interface configured to interface with a processor; a plurality of sensors affixed to the vehicle and configured to interface with the processor; a memory configured to store non-transitory computer executable instructions and configured to interface with the processor; and the processor configured to interface with the memory, wherein the processor is configured to execute the non-transitory computer executable instructions. The non-transitory computer executable instructions cause the processor to obtain a set of real-time operating data indicative of one or more behaviors of the autonomous vehicle; perform one or more operations on the set of real-time