Driver Model Estimation for Personalized Vehicle Behavior
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current information processing systems for vehicles, such as those described in PTL 1, face challenges in estimating a driving operation suited to the driver, leading to difficulties in providing an appropriate driving experience during autonomous driving, including issues with notification of intended behaviors and manual driving transitions.
Innovation Solution
An information processing system that includes a history obtainer, modeler, and behavior estimation unit to build a driver model based on personal driving environment histories, allowing for the estimation of behaviors suited to individual drivers by analyzing past driving habits and environments, and an information notification device that detects vehicle surroundings and states to notify drivers of upcoming behaviors during autonomous driving.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a driving control device notifies the driver of autonomous control activation, then the driver can visually recognize the control state, but it becomes difficult to estimate a driving operation suited to the driver
Solution Approach 1:
The system segments the driving history data into multiple dimensions (driving environment parameters, vehicle state parameters, driver operation parameters) and processes them separately through dedicated analysis modules before integrating them for comprehensive driver behavior estimation
Solution Approach 2:
The system introduces a driver model as an intermediary that learns and represents typical driver behavior patterns from historical data, mediating between raw driving history and real-time driving operation estimation, enabling the system to predict driver intentions even when current input data is limited
2Adaptability or versatility
If the system uses only the target driver's personal driving history, then the estimation is specific to that driver, but the estimation accuracy is insufficient when personal history is limited
Solution Approach 1:
The system merges multiple data sources including the target driver's personal driving history, general driving history from other drivers, and learned driver models into a unified estimation framework, allowing the system to leverage collective driving patterns to compensate for limited individual data
Solution Approach 2:
The system dynamically adjusts the weight and contribution of different data sources (personal history vs. general history vs. driver models) based on the availability and quality of personal driving data, transitioning between data sources as conditions change
3Measurement precision
If the system builds a driver model from multiple drivers' histories, then the model can estimate behavior for drivers with limited history, but the system complexity increases
Solution Approach 1:
The system creates simplified driver models that copy and represent essential driving behavior patterns from historical data, using these models to estimate driver operations without requiring complex real-time analysis of complete driving histories
Solution Approach 2:
The system transforms complex driving history data into standardized parameter representations (driving environment parameters, vehicle state parameters, driver operation parameters), enabling efficient model building and comparison across multiple drivers while reducing computational complexity
Data Source
AI summary
An information processing system capable of estimating a driving conduct suited to a driver includes: a history obtainer that obtains a personal driving environment history of each of a plurality of drivers, each of the personal driving environment histories indicating one or more vehicle behaviors selected by the driver, and a driving environment associated with each of the one or more behaviors, the driving environment being a driving environment of the vehicle at a point in time of selection of the behavior it is associated with; and a modeler that models, from a driving environment history including the personal driving environment histories of the plurality of drivers, the personal driving environment history of at least one of the plurality of drivers, to build a driver model indicating a relationship between a behavior and a driving environment for the vehicle of the at least one of the plurality of drivers.