MULTIPLE-STRATEGY CRASH DISCRIMINATION SYSTEM

Patent №

US 5,490,069

Granted

1996-02-06

Filed 1994

Owner

AUTOMOTIVE SYSTEMS LABORATORY, INC.

Lab

AI components

1

ml

Assignment

Recorded

Dataset

AIPD

2023_r1 edition

Application

08181616

A system (10) and method for actuating a vehicle air bag (38) uses an optical detector (12) to generate an output (22) representative of the distance between a vehicle occupant and a fixed structure within the vehicle. The output (22) is subsequently used by a signal processor/discrimination unit (26) to generate data representative of actual occupant conditions, such as transitory occupant position, velocity and/or acceleration. The occupant condition data is used to select the temporally optimal one of a plurality of different predetermined parameter-based crash discrimination strategies (R,S,T,X,Y,Z), and perhaps further used as a decisional criterion in at least one strategy. Each strategy employs different decisional criteria to provide a different range of actual times to fire. The different decisional criteria of the available strategies may be the result of uniquely-different parameter-based crash discrimination algorithms, or otherwise-identical parameter-based algorithms employing different thresholds, or both. The air bag is actuated when the decisional criteria of the selected strategy are satisfied to generate an actual time to fire. The present system and method thus optimize crash discrimination analysis by customizing the analysis in real time to match actual rather than assumed occupant conditions.

Machine learningB60R 21/013B60R 21/01534B60R 2021/01027B60R 2021/01315

AI classification

Machine learning0.99
Knowledge representation0.07
Planning0.02
AI hardware0.02
Vision0.00
Natural language0.00
Evolutionary computation0.00
Speech0.00

Ownership

AUTOMOTIVE SYSTEMS LABORATORY, INC.

assignment · 68770349

Assignors

GIOUTSOS, TONY, GILLIS, EDWARD J.

On an employer assignment, the assignors are typically the inventors.

© 2026 NYSGPT2525 LLC