Sensor-Based Mobile Device Usage Classification for Driver Risk
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems fail to accurately distinguish between different types of mobile device usage by a driver, leading to inaccurate risk assessments and increased false positives in determining the riskiness of driving behavior.
Innovation Solution
A system and method that utilizes sensors and machine learning algorithms to analyze user device information, including motion, orientation, and proximity data, to classify device usage accurately, distinguishing between holding and mounting of the device, and assess risk scores based on user interactions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional systems approximate mobile device presence in a vehicle, then device detection capability is provided, but measurement precision of device usage types deteriorates
Solution Approach 1:
The patent segments device usage into distinct states (mounted, held, portable) based on sensor data patterns. By dividing the detection space into discrete classification categories, the system achieves precise differentiation of device usage types without requiring overly complex processing, directly resolving the contradiction between measurement precision and device complexity.
Solution Approach 2:
The system dynamically transitions between device usage states based on real-time sensor data analysis. The classification is not static but adapts as the device moves between mounted, held, and portable states, enabling accurate measurement of device usage types while maintaining manageable system complexity through event-driven state transitions.
2Reliability
If conventional systems detect device presence, then basic monitoring capability is provided, but reliability of risk assessment deteriorates due to false positives
Solution Approach 1:
The system employs feedback loops where sensor data continuously informs state classification, which in turn triggers appropriate risk assessment actions. The classification system feeds back into the monitoring process, adjusting detection sensitivity based on the current device state, thereby improving reliability of risk assessment while managing the difficulty of detection through adaptive thresholds.
Solution Approach 2:
The patent introduces an intermediary classification layer between raw sensor detection and final risk assessment. This intermediary state classification system translates complex sensor patterns into discrete device usage states, serving as a mediator that simplifies the detection-measurement challenge while enhancing the reliability of subsequent risk evaluation.
3Measurement precision
If the system classifies device usage into detailed states, then measurement precision of usage types is improved, but loss of information increases due to complex data processing
Solution Approach 1:
The system extracts only the essential features from sensor data needed for device state classification, separating critical information (device position, motion patterns) from extraneous data. By taking out only the necessary information for classification, the system achieves high measurement precision while minimizing information loss and processing overhead.
Solution Approach 2:
The patent transforms continuous sensor parameters into discrete device usage states through parameter changes. By converting continuous motion and position data into categorical classifications (mounted, held, portable), the system maintains high classification accuracy while reducing data complexity and minimizing information loss during processing.
Data Source
AI summary
A method for monitoring device usage includes: receiving user device information; based on the user interaction data, determining a state associated with the user device; and based on the state of the user device, determining a set of tap parameters. Additionally or alternatively, the method can include any or all of: determining a risk score based on at least on the set of tap parameters; determining an output based on the risk score; and/or any other suitable processes performed in any suitable order.