Vehicle Occupant Severity Estimation Using Multi-Sensor Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle-related incident response systems lack an effective approach to determine appropriate remedial actions based on the severity of the incident and its impact on vehicle occupants.
Innovation Solution
A severity estimation system that combines audio-based, scene recognition, restraint-based, and motion-based inputs using a weighted formula to determine an incident severity indicator, which informs the selection of remedial actions such as contacting emergency personnel, unlocking doors, or sending an SOS signal.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a simple collision response system is used, then the device complexity is reduced, but the measurement precision of incident severity is insufficient
Solution Approach 1:
The patent combines multiple independent detection systems (audio-based indicator system, occupant monitoring system, restraint system, motion-based indicator system) into a unified severity estimation system. Each system contributes specific data (audio inputs, scene recognition inputs, restraint-based inputs, motion-based inputs) that are fused together through a weighted formula to achieve comprehensive and accurate incident severity assessment.
Solution Approach 2:
The severity estimation system serves multiple functions simultaneously: it assesses incident severity, determines occupant status, selects appropriate remedial actions, and coordinates vehicle response systems. This multi-functionality allows a single integrated system to replace multiple separate systems, improving measurement precision while managing device complexity through functional consolidation.
2Reliability
If multiple detection systems are integrated, then the incident severity assessment accuracy is improved, but the device complexity increases
Solution Approach 1:
The system implements feedback mechanisms where the fused incident severity indicator feeds back into the remedial action selection process. The controller continuously monitors inputs from all detection systems, updates the severity assessment, and adjusts remedial actions accordingly. This feedback loop ensures reliable action selection by constantly comparing actual system state with desired outcomes.
Solution Approach 2:
The patent employs parameter changes by using a weighted formula that assigns different importance levels to various input parameters from different detection systems. The weights can be adjusted based on incident type, occupant position, and system configuration, allowing the system to adapt its assessment criteria and maintain reliability across diverse scenarios while managing complexity through parameter optimization.
Data Source
AI summary
A severity estimation system for one or more occupants located in an interior cabin of a vehicle involved in a vehicle-related incident includes one or more microphones located within the interior cabin of the vehicle that capture audio-based inputs, an occupant monitoring system (OMS) that collects scene recognition inputs indicative of a status of the occupants, a restraint system that collects a plurality of restraint-based inputs indicative of one or more restraint-based mechanisms associated with an occupant of the vehicle being activated during the vehicle-related incident, a motion-based indicator system that collect the plurality of motion-based inputs indicative of the motion of the vehicle during the vehicle-related incident, and one or more controllers. The one or more controllers fuse together the audio-based indicator, the OMS indicator, the restraint-based indicator, and the motion-based indicator to determine an incident severity indicator corresponding to the one or more occupants.