Mobile Device Usage Classification for Driver Risk Scoring
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 inconsistent risk assessments and potential false positives in determining the riskiness of driving behavior.
Innovation Solution
A system and method that utilizes sensors and processing units to collect and analyze motion, orientation, and proximity data from a user's device to differentiate between various types of device usage, such as holding versus mounting, and assess risk scores based on these interactions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional systems detect mobile device presence in a vehicle, then device usage can be monitored, but the systems cannot accurately distinguish between different types of device usage (holding vs. mounting)
Solution Approach 1:
The patent segments device usage into distinct categories (holding vs. mounting) by analyzing multiple sensor parameters separately (acceleration, orientation, proximity) and combining their classifications to determine the overall device state, thereby achieving precise differentiation without requiring a single complex sensor system
Solution Approach 2:
The patent introduces an intermediary classification process that uses sensor data to infer device state indirectly. Instead of directly detecting whether a device is held or mounted, the system uses motion patterns, orientation changes, and proximity sensor readings as intermediate indicators to classify device usage behavior
2Productivity
If conventional systems classify all device usage as risky, then risk assessment is simplified, but false positives increase due to inability to distinguish low-risk usage patterns
Solution Approach 1:
The patent applies local quality by assigning different risk levels to different device usage states. Instead of uniformly classifying all device usage as high-risk, the system determines local risk characteristics based on specific device states (e.g., mounted devices during autonomous driving vs. manually held devices during active driving), allowing nuanced risk assessment that reduces false positives while maintaining efficiency
Solution Approach 2:
The patent changes risk assessment parameters dynamically based on device state classification. The system adjusts risk scores according to detected device usage patterns, vehicle motion states, and driving conditions, transforming a static binary risk assessment into a dynamic multi-parameter evaluation that improves reliability without sacrificing productivity
3Measurement precision
If the system uses multiple sensors and data processing to distinguish device usage types, then classification accuracy improves, but computational complexity and processing time increase
Solution Approach 1:
The patent applies preliminary action by pre-defining classification thresholds and decision rules for different device states. The system establishes predetermined criteria for distinguishing holding vs. mounting behaviors based on sensor parameter ranges, allowing rapid classification without requiring complex real-time computation, thus reducing processing time while maintaining accuracy
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.