Driver Device Identification Using In-Vehicle Telematics Positioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current GPS technology lacks precision to accurately identify the driver of a vehicle, leading to inefficiencies in usage-based insurance policies and telematics data analysis due to the inability to distinguish between the driver and passengers.
Innovation Solution
Utilizing a driver identification computing device that enhances geographic location measurements through Kalman filtering and machine learning techniques, combining GPS data with accelerometer and gyroscope data to determine the relative position of user devices within a vehicle, thereby identifying the driver based on predicted positions and telematics patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If GPS measurements are used to identify the driver, then the system can provide driver identification capability, but the location precision is insufficient to accurately distinguish driver from passengers
Solution Approach 1:
The patent combines multiple data sources including GPS location data, accelerometer data, gyroscope data, and barometer data into an integrated sensor fusion system. This merging of multiple measurement systems allows the device to overcome the limitations of individual sensors and achieve sufficient precision for driver identification by analyzing both location patterns and motion characteristics
Solution Approach 2:
The patent introduces intermediate processing layers including Kalman filtering and machine learning algorithms that act as mediators between raw sensor data and driver identification decisions. These intermediary systems process and refine the imprecise GPS measurements by combining them with other sensor data and applying computational models to extract meaningful driver behavior patterns
2Quantity of substance
If multiple user devices are monitored in a vehicle, then the system can capture comprehensive telematics data, but it becomes difficult to distinguish which device belongs to the driver versus passengers
Solution Approach 1:
The patent applies local quality analysis by examining the specific spatial and temporal characteristics of each device's motion patterns. By analyzing the local quality of acceleration profiles, turning patterns, and location changes specific to each device, the system can identify which device exhibits driver-like behavior patterns distinct from passenger devices, even when multiple devices are present in the vehicle
Solution Approach 2:
The patent employs dynamic analysis of device motion characteristics, using machine learning models to recognize temporal patterns in accelerometer and gyroscope data. The system continuously monitors how each device moves and changes orientation over time, identifying the driver device through its unique dynamic signature that reflects active driving maneuvers rather than passive passenger movement
3Reliability
If GPS error estimates are used to account for atmospheric and artificial interference, then the system acknowledges measurement uncertainties, but the precision remains insufficient for accurate driver identification
Solution Approach 1:
The patent creates a composite measurement system that combines multiple types of sensor data (GPS, accelerometer, gyroscope, barometer) into an integrated dataset. This composite approach is analogous to composite materials in engineering, where combining different materials with complementary properties creates a system that overcomes the weaknesses of individual components. The fused sensor data provides more reliable and precise driver identification than GPS alone, even when GPS error estimates account for atmospheric and artificial interference
Data Source
AI summary
A computer-implemented method to determine a device of a driver including: receiving, by a computing device and from a first user device, first telematics data; receiving, by the computing device and from a second user device, second telematics data; determining, using a trained machine learning model, patterns in the first and second telematics data; determining, using the patterns in the first and second telematics data, relative positions of the first user device and the second user device; determining a predicted position of the first user device relative to the second user device; and identifying that the first user device or the second user device is associated with the driver of a vehicle based at least in part upon the predicted position of the first user device relative to the second user device. Other embodiments are described.