Driver Identification Model Using Vehicle And Terminal Position Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in accurately identifying and specifying the user who is driving a vehicle, which is crucial for determining car insurance rates based on driving skill and safety.
Innovation Solution
The system employs a learning model that estimates the group of users on board a vehicle by comparing terminal positions with vehicle positions, and generates a model using identification information, driving attributes, and vehicle attributes as training data to specify the driver.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If terminal position data and vehicle position data are collected and compared to identify on-board users, then user identification accuracy is improved, but system complexity and data processing burden increase
Solution Approach 1:
The system segments the user identification process into distinct functional modules: a terminal position data acquisition unit that collects terminal location information, a vehicle position data acquisition unit that collects vehicle location information, and an on-board user identification unit that compares these datasets to determine which terminals are on-board users. This segmentation allows each module to perform a specific function, improving identification accuracy while managing system complexity through modular design.
2Measurement precision
If comprehensive training data including user identification information, driving attributes, and vehicle attributes is collected to generate learning models, then driver specification accuracy is improved, but data processing time and computational resources increase
Solution Approach 1:
The system performs preliminary data collection and organization by gathering terminal position data, vehicle position data, user identification information, driving attributes, and vehicle attributes in advance. This pre-collected comprehensive dataset is then used to generate learning models, which reduces the computational burden and processing time required during actual driver specification operations, as the model generation can be performed offline or in advance.
3Reliability
If multiple data sources including terminal data and vehicle data are integrated for on-board user estimation, then identification reliability is improved, but information processing load increases
Solution Approach 1:
The system introduces an on-board user identification unit as an intermediary component that receives and processes both terminal position data and vehicle position data. This intermediary unit compares the datasets and generates identification results, thereby integrating multiple data sources to improve reliability while managing the information processing load through a dedicated specialized component rather than distributed processing.
Data Source
Figure 1~3
Figure 4~5
Figure 6~7
AI summary
Learning model generating unit 332 generates a learning model using a user ID of a user estimated by driver estimating unit 331 and driving attributes of the user, as training data. Specifically, learning model generating unit 332 generates a learning model in which the driving attributes are an explanatory variable and the user ID of a user is an objective variable. When various driving attributes are input to this driving behavior model, the user ID of a driver specified by the attributes is output.