US-12617418-B2 - Systems and methods for detecting vehicle idling and determining classifications for the vehicle idling

US12617418B2US 12617418 B2US12617418 B2US 12617418B2US-12617418-B2

Abstract

A device may receive road facing camera (RFC) video data, driver facing camera (DFC) video data, and idling events data associated with a vehicle, and may receive traffic data associated with the vehicle. The device may determine that an idling event of the idling events data is an idling event trigger, and may process the idling event and the DFC video data, based on the idling event being an idling event trigger and with a first machine learning model, to determine a behavior of a driver of the vehicle. The device may process the behavior of the driver, the RFC video data, and the traffic data, with a second machine learning model, to determine a score for the idling event, and may determine a classification for the idling event based on the score and a score threshold. The device may perform one or more actions based on the classification.

Inventors

Francesco DE FELICE
Filippo VALENTE
Giovanni Pini
Tommaso MUGNAI

Assignees

VERIZON PATENT AND LICENSING INC.

Dates

Publication Date: 20260505
Application Date: 20230908

Claims (20)

1 . A method, comprising: receiving, by a device, idling events data associated with a vehicle; receiving, by the device, traffic data associated with the vehicle, wherein the traffic data is received from a source external to the vehicle, and wherein the traffic data and the idling events data are received from different sources; determining, by the device, that an idling event of the idling events data is an idling event trigger; processing, by the device, based on the idling event being an idling event trigger, and with a first machine learning model, the idling events data and the traffic data, to determine a first score for the idling event, based on determining that video data is not available, wherein the video data is associated with a vehicle device; processing, by the device, the idling events data, the video data, and the traffic data, with a machine learning model and based on the idling event being an idling event trigger, to determine a second score for the idling event, based on determining that the video data is available; determining, by the device, a classification for the idling event based on the first score or the second score, and a score threshold; and performing, by the device, one or more actions based on the classification.
2 . The method of claim 1 , wherein the classification provides an indication of whether the idling event is an acceptable idling event relative to the score threshold.
3 . The method of claim 1 , wherein determining the classification for the idling event based on the first score or the second score, and the score threshold comprises: determining whether the first score or the second score satisfies the score threshold; and determining the classification for the idling event based on whether the first score or the second score satisfies the score threshold.
4 . The method of claim 1 , wherein performing the one or more actions comprises: generating a notification based on the classification; and providing the notification to the vehicle.
5 . The method of claim 1 , wherein performing the one or more actions comprises one or more of: determining a rating for a driver of the vehicle based on the classification; or scheduling the driver of the vehicle for training based on the classification.
6 . The method of claim 1 , wherein performing the one or more actions comprises one or more of: causing the vehicle to be disabled based on the classification; or retraining the machine learning model based on the classification.
7 . The method of claim 1 , wherein the idling events data associated with the vehicle is generated by a vehicle tracking system of the vehicle.
8 . A device, comprising: one or more processors configured to: receive road facing camera (RFC) video data and idling events data associated with a vehicle; receive traffic data associated with the vehicle, wherein the traffic data is received from a source external to the vehicle, and wherein the traffic data and the idling events data are received from different sources; determine that an idling event of the idling events data is an idling event trigger; selectively perform: processing the idling events data and the traffic data, based on the idling event being an idling event trigger and with a first machine learning model, to determine a first score for the idling event, based on determining that the RFC video data is not available, or processing the idling events data, the RFC video data, and the traffic data, based on the idling event being an idling event trigger and with a machine learning model, to determine a second score for the idling event, based on determining that the RFC video data is available; determine a classification for the idling event based on the first score or the second score, and a score threshold; and perform one or more actions based on the classification.
9 . The device of claim 8 , wherein the classification provides an indication of whether the idling event is an acceptable idling event relative to the score threshold.
10 . The device of claim 8 , wherein the one or more processors, to determine the classification for the idling event based on the first score or the second score, and the score threshold, are configured to: determine whether the first score or the second score satisfies the score threshold; and determine the classification for the idling event based on whether the first score or the second score satisfies the score threshold.
11 . The device of claim 8 , wherein the one or more processors, to perform the one or more actions, are configured to: generate a notification based on the classification; and provide the notification to the vehicle.
12 . The device of claim 8 , wherein the one or more processors, to perform the one or more actions, are configured to one or more of: determine a rating for a driver of the vehicle based on the classification; or schedule the driver of the vehicle for training based on the classification.
13 . The device of claim 8 , wherein the one or more processors, to perform the one or more actions, are configured to one or more of: cause the vehicle to be disabled based on the classification; or retrain the machine learning model based on the classification.
14 . The device of claim 8 , wherein the idling events data associated with the vehicle is generated by a vehicle tracking system of the vehicle.
15 . A non-transitory computer-readable medium storing a set of instructions, the set of instructions comprising: one or more instructions that, when executed by one or more processors of a device, cause the device to: receive road facing camera (RFC) video data, driver facing camera (DFC) video data, and idling events data associated with a vehicle; receive traffic data associated with the vehicle, wherein the traffic data is received from a source external to the vehicle, and wherein the traffic data and the idling events data are received from different sources; determine that an idling event of the idling events data is an idling event trigger; selectively perform: processing the idling events data and the traffic data, based on the idling event being an idling event trigger and with a first machine learning model, to determine a first score for the idling event, based on determining that the RFC video data and the DFC video data are not available, performing, based on determining that the RFC video data and the DFC video data are available: processing the idling events data and the DFC video data, based on the idling event being an idling event trigger and with a first machine learning model, to determine a behavior of a driver of the vehicle; and processing the behavior of the driver, the RFC video data, and the traffic data, with a second machine learning model, to determine a second core for the idling event; determine a classification for the idling event based on the first score or the second score, and a score threshold; and perform one or more actions based on the classification.
16 . The non-transitory computer-readable medium of claim 15 , wherein the behavior of the driver includes one of the driver being located in the vehicle and the vehicle is moving, the driver being located in the vehicle and the vehicle is not moving, or the driver not being located in the vehicle.
17 . The non-transitory computer-readable medium of claim 15 , wherein the one or more instructions, that cause the device to determine the classification for the idling event based on the first score or the second score and the score threshold, cause the device to: determine whether the first score or the second score satisfies the score threshold; and determine the classification for the idling event based on whether the first score or the second score satisfies the score threshold.
18 . The non-transitory computer-readable medium of claim 15 , wherein the one or more instructions, that cause the device to perform the one or more actions, cause the device to: generate a notification based on the classification; and provide the notification to the vehicle.
19 . The non-transitory computer-readable medium of claim 15 , wherein the one or more instructions, that cause the device to perform the one or more actions, cause the device to one or more of: determine a rating for a driver of the vehicle based on the classification; or schedule the driver of the vehicle for training based on the classification.
20 . The non-transitory computer-readable medium of claim 15 , wherein the one or more instructions, that cause the device to perform the one or more actions, cause the device to one or more of: cause the vehicle to be disabled based on the classification; or retrain the first machine learning model and the second machine learning model based on the classification.

Description

BACKGROUND With today's modern automobile engines, no more than thirty seconds of idling is needed on winter days before driving the vehicle. Nevertheless, it is common practice to run a vehicle for several minutes in order to warm up the vehicle. However, idling only warms the vehicle's engine, but fails to warm wheel bearings, steering, suspension, transmission, and tires of the vehicle. These parts also need to be warmed up, and the only way to do that is to drive the vehicle. BRIEF DESCRIPTION OF THE DRAWINGS FIGS. 1A-1F are diagrams of an example associated with detecting vehicle idling and determining classifications for the vehicle idling. FIG. 2 is a diagram illustrating an example of training and using a machine learning model. FIG. 3 is a diagram of an example environment in which systems and/or methods described herein may be implemented. FIG. 4 is a diagram of example components of one or more devices of FIG. 3. FIGS. 5-7 are flowcharts of example processes for detecting vehicle idling and determining classifications for the vehicle idling. DETAILED DESCRIPTION OF EXAMPLE EMBODIMENTS The following detailed description of example implementations refers to the accompanying drawings. The same reference numbers in different drawings may identify the same or similar elements. Although some vehicles may need to idle, vehicle idling creates environmental and health issues by increasing the amount of vehicle exhaust in the air. Vehicle exhaust contains many pollutants that are linked to asthma and other lung diseases, allergies, heart disease, increased risk of infections and cancer, and other health problems. Air pollution is one of the top causes for climate change and all its catastrophic consequences. Higher levels of air pollution have been linked to increased school absences, hospital visits, and even premature deaths. Vehicle emissions are still present and harmful even when you cannot see the exhaust. Vehicle idling also creates cost issues and vehicle damage. For each hour spent idling, a typical light duty truck wastes approximately one gallon of diesel fuel and a typical car wastes 0.2 gallons of gasoline. Vehicle idling can damage the vehicle's transmission or overheat the vehicle's engine. The problems associated with vehicle idling are further exasperated with a fleet of vehicles since a fleet may include thousands, tens of thousands, hundreds of thousands, and/or the like of vehicles associated with vehicle idling. Thus, current techniques for operating a vehicle and/or managing a fleet of vehicles consume computing resources (e.g., processing resources, memory resources, communication resources, and/or the like), networking resources, and/or other resources associated with creating unnecessary air pollution with vehicle idling, creating health issues with the unnecessary air pollution, increasing operating costs associated with the vehicle or the fleet of vehicles, causing damage to the vehicle or the fleet of vehicles, and/or the like. Some implementations described herein relate to a video system that detects vehicle idling and determines classifications for the vehicle idling. For example, the video system may receive road facing camera (RFC) video data, driver facing camera (DFC) video data, and idling events data associated with a vehicle, and may receive traffic data associated with the vehicle. The video system may determine that an idling event of the idling events data is an idling event trigger, and may process the idling event and the DFC video data, based on the idling event being an idling event trigger and with a first model (e.g., a machine learning model), to classify a behavior of a driver of the vehicle. The video system may process the behavior of the driver, the RFC video data, and the traffic data, with a second machine learning model, to determine a score for the idling event, and may determine a classification for the idling event based on the score and a score threshold. The video system may perform one or more actions based on the classification. In this way, the video system detects vehicle idling and determines classifications for the vehicle idling. For example, the video system may identify and classify vehicle idling events utilizing, e.g., a set of machine learning models based on available data. The video system may utilize the identification and classification of the vehicle idling events to score fleets on idling behavior, alert drivers of unnecessary vehicle idling, to provide training to drivers associated with idling, and/or the like. The video system may utilize a first or basic machine learning model that identifies and classifies vehicle idling events based on vehicle tracking unit (VTU) data. The video system may utilize a second or enhanced machine learning model that identifies and classifies vehicle idling events based on VTU data and RFC video data. The video system may utilize a third or ultimate machine learning model that identifies and cla