Occupant Interest Detection for Adaptive Autonomous Driving Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional automatic driving control methods do not consider the occupant's level of interest, making it impossible to perform driving control that accurately reflects the occupant's intention.
Innovation Solution
A control method and device that detect the occupant's level of interest using an interest level detection unit, which includes an eyeball state detection unit, switch operation detection unit, and conversation determination unit, to adjust automatic driving characteristics based on the occupant's attention and engagement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If automatic driving control is performed without considering occupant's level of interest, then the control system is simple and easy to operate, but the driving control cannot accurately reflect the occupant's intention
Solution Approach 1:
The interest level detection unit is divided into multiple independent detection modules: eyeball state detection, switch operation detection, and conversation determination. Each module independently detects specific aspects of occupant engagement, and their results are integrated to comprehensively determine interest level, thereby improving detection precision while keeping individual modules manageable in complexity
Solution Approach 2:
The control device integrates multiple detection functions (eyeball detection, switch operation detection, conversation analysis) into a single interest level detection unit that serves both monitoring and control adjustment purposes, reducing overall system complexity while enhancing the precision of occupant intention detection
2Reliability
If multiple detection methods are used to determine occupant's interest level, then the accuracy of reflecting occupant's intention is improved, but the device complexity increases
Solution Approach 1:
Multiple detection methods (eyeball state detection, switch operation detection, conversation determination) are merged into a unified interest level detection unit. The detection results from different methods are combined and processed together to determine the occupant's interest level, improving reliability through multi-modal detection while managing complexity through integrated processing
Solution Approach 2:
The system continuously monitors occupant state through multiple detection methods and uses the collected information to adjust automatic driving characteristics in real-time. This feedback loop ensures reliable reflection of occupant intention while the systematic approach to processing multiple detection streams manages overall complexity
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An interest level detection unit (1) configured to detect a level of interest of an occupant in a travel state of an automatic driving vehicle, a manual driving characteristic learning unit (7) configured to learn manual driving characteristics based on the travel state of the automatic driving vehicle, and an automatic driving characteristic setting unit (8) configured to set automatic driving characteristics based on a surrounding state of the automatic driving vehicle are provided. Moreover, an interest level determination unit (12) configured to determine the occupant's level of interest in the vehicle travel and a driving characteristic setting unit (13) configured to set the driving characteristics based on the manual driving characteristics learned in the manual driving learning unit (7) when the level of interest is determined to be high and set the automatic driving characteristics set by the automatic driving characteristic setting unit (8) when the level of interest is determined to be low are provided.