Vehicle Occupant Sensor Fusion Using Signal Quality Weighting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle safety systems fail to adequately address driver medical conditions, such as cardiac events, due to complex algorithms and processing latency, overlooking the need for efficient and timely detection of physiological abnormalities.
Innovation Solution
A method for determining a fused sensor measurement using a signal quality index to weight sensor data based on variance and Kalman filtering, optimizing processing by assigning higher weightage to consistent measurements and ignoring noisy ones, enabling faster and more accurate detection of physiological conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If complex algorithms are used to analyze health data, then measurement precision is improved, but processing time increases causing latency
Solution Approach 1:
The patent segments the complex analysis into distinct stages: signal acquisition from multiple sensors, quality assessment of each sensor signal, selective fusion of high-quality signals, and medical condition determination. This segmentation allows efficient processing at each stage rather than applying complex algorithms to all data uniformly.
Solution Approach 2:
The patent applies local quality assessment by evaluating the quality of individual sensor signals separately using quality metrics. Only signals meeting quality thresholds are selected for fusion, ensuring that local signal quality determines which data contributes to the final analysis, thereby maintaining precision while reducing processing burden.
2Reliability
If multiple sensors are used to detect physiological data, then reliability is improved, but device complexity increases
Solution Approach 1:
The patent merges data from multiple sensors through a fusion process that combines signals meeting quality criteria. This merging approach maintains the reliability benefits of multiple sensors while managing complexity through systematic integration rather than independent processing of each sensor.
Solution Approach 2:
The patent extracts only the necessary information from multiple sensors by applying quality assessments and selecting only high-quality signals for fusion. This extraction approach reduces the complexity burden of processing all sensor data while retaining the reliability advantages of multi-sensor detection.
3Measurement precision
If all sensor measurements are processed, then measurement precision is improved, but productivity decreases due to unnecessary processing
Solution Approach 1:
The patent applies partial action by processing only those sensor measurements that meet quality thresholds rather than all available measurements. This selective processing maintains measurement precision for qualified signals while improving productivity by avoiding unnecessary processing of low-quality data.
Solution Approach 2:
The patent changes the parameter of signal quality through quality assessment metrics, using these quality parameters to determine which measurements warrant further processing. This parameter-based selection optimizes the balance between precision and productivity by processing only measurements above quality thresholds.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach allows for more reliable and timely detection of abnormal physiological conditions, facilitating appropriate vehicle responses to ensure safety, including alerts and automated actions.
Implementation Method 1
determining a signal quality index of each sensor based on a variance of the sensor measurements
Implementation Method 2
a Kalman filtering is performed on the sensor measurements obtained from the sensors
Data Source
Figure 1~2
AI summary
There is provided a method of determining a fused sensor measurement, the method comprising: a. obtaining, by a processor, a number of sensor measurements from each of a plurality of sensors detecting a same type of physiological measurement; b. determining, by the processor, a signal quality index of each sensor, wherein the signal quality index comprises determining an extent to which a sensor measurement differs from the others among the number of sensor measurements obtained from each sensor; c. determining, by the processor, a weightage of each sensor based on the signal quality index of each sensor; and d. determining, by the processor, a fused sensor measurement from the plurality of sensors based on the weightage of each sensor and filtered sensor measurements of each sensor obtained from a Kalman filter operation. There is also provided a vehicle safety system comprising: a plurality of sensors detecting a same type of physiological sensor measurement from an occupant in the vehicle; and a vehicle electronic control unit comprising at least one processor, the at least one processor configured: obtain the sensor measurements from the plurality of sensors, determine an extent to which a sensor measurement differs from the others among the number of sensor measurements obtained from each sensor, to determine a signal quality index of each sensor, determine a weightage of each sensor based on the signal quality index of each sensor, determine a fused sensor measurement from the plurality of sensors based on the weightage of each sensor and filtered sensor measurements of each sensor obtained from a Kalman filter operation, determine a physiological condition of the occupant based on the fused sensor measurement, and if the physiological condition is abnormal, perform at least one vehicle operation in response to the abnormal physiological condition.