3D Occupant Position Tracking for Adaptive Airbag Deployment

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current algorithms for determining occupant position in vehicles rely on indirect estimates based on seat track and weight sensors, leading to inaccurate restraint deployment strategies that may increase the risk of injury from airbag deployment.

Innovation Solution

A method and system using a 3D sensor system and a Kalman filter to capture time-delayed image frames, determine precise 3D positions of body points, and calculate filtered head and chest positions, ensuring accurate occupant positioning for adaptive restraint deployment.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If indirect estimates based on seat track and weight sensors are used to determine occupant position, then device complexity is reduced, but measurement precision deteriorates

Engineering Contradiction:

Improvesensor system complexityVSAvoidoccupant position accuracy

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent replaces mechanical sensors (seat track and weight sensors) with a 3D sensor system that uses optical fields to directly measure occupant position. This substitution of measurement methodology enables direct observation of occupant location and posture without relying on indirect mechanical estimates, thereby improving measurement precision while maintaining acceptable system complexity

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent introduces a Kalman filter as an intermediary processing layer that fuses data from multiple 3D sensor measurements over time. This intermediary component processes raw sensor data to produce accurate, filtered estimates of occupant position, enhancing measurement precision through temporal integration and noise reduction

Inventive Principle:

Principle #24Intermediary (Mediator)

2Measurement precision

If 3D sensor system with Kalman filter is used to capture time-delayed image frames, then measurement precision is improved, but device complexity increases

Engineering Contradiction:

Improveoccupant position accuracyVSAvoidsignal processing complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent performs preliminary actions by capturing multiple time-delayed image frames before the actual crash event occurs. This allows the system to establish baseline occupant position data and predict future positions, improving measurement precision for the critical deployment moment while distributing processing load over time

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The Kalman filter implements a feedback mechanism where each new sensor measurement is continuously compared with predicted occupant position based on previous frames. The filter adjusts its estimates based on this feedback loop, improving measurement precision through iterative refinement while managing processing complexity through efficient recursive calculations

Inventive Principle:

Principle #23Feedback

3Measurement precision

If filtered head position and chest position are calculated from multiple body points, then measurement precision is improved, but loss of time increases

Engineering Contradiction:

Improveoccupant position accuracyVSAvoidprocessing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent calculates positions for multiple body points (head, chest, and intermediate points) rather than just the final required positions. This excessive action provides redundant data that improves measurement precision through cross-validation and error reduction, while the Kalman filter efficiently processes this additional information

Inventive Principle:

Principle #16Partial or excessive action

4Reliability

If 3D sensor system is used to capture occupant position data, then reliability is improved, but use of energy increases

Engineering Contradiction:

Improverestraint deployment accuracyVSAvoidsensor system power consumption

Core Design Contradiction:

ReliabilityVSUse of energy by moving object

Solution Approach 1:

The 3D sensor system operates periodically, capturing image frames at specific time intervals rather than continuously. This periodic operation maintains reliability by gathering sufficient position data at critical moments while reducing overall energy consumption by keeping the sensor in low-power states between measurements

Inventive Principle:

Principle #19Periodic action

Data Source

PatentEP4420937B1Method for determining occupant position to adapt a restraint deployment of a vehicle and system for determining an occupant position in a vehicle to be used in a restraint deployment adaption in a vehicle

Publication Date: 2025.11.26 VEONEER SWEDEN SAFETY SYST AB

Figure 1
Figure 2~3C
Figure 4

AI summary

In a method and a system for determining positions of an occupant in a vehicle to adapt a restraint deployment of the vehicle (1), at least one 3D sensor system (21) is applied to capture a plurality of time-delayed image frames of the at least one occupant (3) in the vehicle (1). A Kalman filter (40) determines a filtered head position (33F), a filtered right shoulder position (34F), a filtered left shoulder position (35F), and a filtered middle spine position (37F) from the plurality of body points (32) and predicts state (st) evolution in case of occlusion or transitory low measurement quality. A calculation section (50) calculates a current 3D head position (33) and a current 3D chest position (36) which are used for the adaption of the restraint deployment of vehicle (1).