Human-State Estimation Using Thermal Sensation Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current human-state estimating methods face challenges in accurately distinguishing between stress, sleepiness, and thermal sensation influences on physiological measures like skin temperature and heartbeat intervals, leading to false detections and reduced accuracy.
Innovation Solution
A method that involves obtaining a thermal sensation index and determining if it falls within a predetermined range, allowing for more accurate estimation of human states such as stress or sleepiness by isolating the influence of thermal sensation and using nasal skin temperature or heartbeat intervals as physiological measures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If human state is estimated based on physiological amounts like skin temperature and heartbeat intervals, then sleepiness and stress detection is enabled, but false detections occur due to thermal sensation influences
Solution Approach 1:
The patent segments the estimation process into two distinct stages: first estimating thermal sensation from skin temperature, then using that thermal sensation estimate to correct the interpretation of physiological amounts for human state detection. This segmentation isolates the thermal sensation variable that causes false detections, allowing the system to distinguish between temperature changes due to thermal sensation versus those due to stress or sleepiness.
Solution Approach 2:
The patent introduces thermal sensation as an intermediary variable that mediates between raw skin temperature measurements and final human state estimation. By first determining thermal sensation from skin temperature and then using this intermediary information to adjust the interpretation of heartbeat intervals and other physiological signals, the system eliminates false detections caused by thermal influences.
2Measurement precision
If continuous monitoring of physiological amounts is performed to improve detection accuracy, then human state estimation precision increases, but processing load and power consumption increase
Solution Approach 1:
The patent performs preliminary estimation of thermal sensation from skin temperature before proceeding to human state estimation. This preliminary action allows the system to pre-process the thermal sensation information and use it to guide subsequent, more computationally intensive physiological amount analysis, thereby reducing overall processing load by avoiding unnecessary calculations when thermal sensation is stable or when physiological signals clearly indicate a specific state.
Data Source
AI summary
A human-state estimating method includes: obtaining a thermal sensation index obtained by indexing, in a specified range, an estimated thermal sensation of a person; determining whether or not the obtained thermal sensation index is in a predetermined range in the specified range; and estimating a human state, which is a state of the person, based on a physiological amount in which an activity of an autonomic nervous system of the person is reflected, the physiological amount being obtained when it is determined that the obtained thermal sensation index is in the predetermined range.