Seat-Back Position Sensing for Accurate Passenger Body-Type Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional passenger determining devices using cameras for imaging face challenges in accurately distinguishing between adult and child passengers due to obstruction by objects and difficulty in discerning the slight disparity between booster and small adult passengers, leading to potential misidentification and increased injury risk from airbag deployment.
Innovation Solution
A system utilizing a seat detector, passenger detector, and controller to learn a body-type classification model based on seat and passenger position, angle, and body part parameters, correcting abnormal sitting states, and compensating for body size distortions to ensure accurate passenger classification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a camera is used to capture passenger images for identification, then passenger type differentiation is enabled, but measurement precision deteriorates when passengers are obstructed by objects or not sitting normally
Solution Approach 1:
The patent introduces a seat position sensor as an intermediary device that detects seat back position and angle, providing auxiliary data to compensate for measurement errors caused by passenger obstruction or abnormal sitting postures. This mediator enables the system to infer body dimensions more accurately even when direct camera measurement is compromised.
Solution Approach 2:
The system changes measurement parameters by combining multiple data sources (camera images, seat position, seat angle) rather than relying solely on direct image measurement. It dynamically adjusts classification thresholds based on seat configuration parameters to maintain measurement precision under varying conditions.
2Measurement precision
If traditional camera-based passenger determination is used, then device complexity is low, but measurement precision deteriorates due to inability to discern slight body size disparities
Solution Approach 1:
The patent makes the detection system multi-functional by enabling it to perform both direct image-based passenger detection and indirect inference based on seat position data. The same system can adaptively switch between measurement modes depending on sitting posture quality, thereby improving precision without requiring entirely separate detection systems.
Solution Approach 2:
The system creates a virtual model of passenger body dimensions by combining camera imagery with seat position data. This copied or inferred body model allows for more precise differentiation of slight size disparities compared to relying on raw camera measurements alone.
3Object-affected harmful factors
If airbag deployment is regulated based on passenger classification, then passenger safety is improved, but object-affected harmful factors increase due to misidentification risks
Solution Approach 1:
The system performs preliminary verification of passenger identification by cross-checking multiple data sources (camera images, seat position, seat angle) before final classification. This preliminary action reduces misidentification risks and ensures more reliable airbag deployment decisions, thereby reducing harmful effects.
Solution Approach 2:
The system incorporates feedback mechanisms where seat position sensor data continuously validates and refines passenger classification results. This feedback loop enhances identification reliability by detecting inconsistencies that might indicate misclassification, thereby reducing airbag-related injury risks.
Data Source
AI summary
A system capable of determining a type of a passenger sitting on a seat according to a body type is disclosed. The system includes a seat detecting means detecting a position of a seat and an angle of a seat back, a passenger detecting means detecting a passenger based on a state where a specific passenger normally sits on the seat, and a controller detecting a passenger's body part along with a seat-back area where the seat back is positioned and a passenger area where the passenger is positioned using information detected by the seat detecting means and the passenger detecting means, and learning a parameter value that is input as a relationship of detected information and a classification result for a body type of the passenger that is output by the parameter value, thus setting a body-type classification model.