Elevator Cab Sensor Fusion for Trapped Passenger Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing elevator systems lack accurate methods to detect trapped passengers and often result in false alarms, and conventional methods fail to distinguish between living beings and inanimate objects, leading to inadequate assistance and potential risks for trapped individuals.

Innovation Solution

A system incorporating jerk detection sensors, microphones, and breath detection sensors, coupled with a central server and machine learning algorithms, to prognostically identify elevator malfunctions and trapped passengers, providing quick assistance and accurate detection.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If weight sensors are used to determine passenger presence, then the system can detect if someone is in the elevator, but it cannot distinguish between a living breathing human being and an inanimate object such as a trolley or luggage

Engineering Contradiction:
Improvepassenger presence detection accuracyVSAvoidability to distinguish living being from object
Core Design Contradiction:
Measurement precisionVSLoss of information

Solution Approach 1:

The detection system is segmented into multiple independent sensor components: weight sensors for mass detection, microphones for sound detection, and breath detection sensors for respiratory analysis. Each sensor segment performs a specific function, and their combined results enable accurate distinction between living passengers and inanimate objects.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Multiple sensing modalities (weight, sound, breath detection) are merged into a unified detection system. The controller integrates data from all sensors to comprehensively determine passenger presence and status, overcoming the limitations of any single sensor type.

Inventive Principle:
Principle #5Merging (Combining)

2Ease of operation

If panic buttons are used for alarm signaling, then passengers can request help, but the system produces many false alarms due to inadvertent pushing or unknowing children

Engineering Contradiction:
Improvealarm activation simplicityVSAvoidalarm accuracy
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The system continuously monitors multiple parameters (weight, sound patterns, breath detection) and uses this feedback to dynamically assess whether an alarm condition is genuine. The controller analyzes the combination of sensor inputs over time to distinguish true emergencies from false alarm scenarios, reducing unreliable panic button activations.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system performs preliminary detection and analysis using multiple sensors before triggering an alarm response. By预先 analyzing sensor data patterns and confirming genuine distress conditions before activation, the system prevents premature or false alarm responses while maintaining ease of legitimate emergency activation.

Inventive Principle:
Principle #10Preliminary action

3Measurement precision

If multiple sensors and machine learning algorithms are integrated for accurate trapped passenger detection, then false alarms are reduced, but the device complexity increases

Engineering Contradiction:
Improvetrapped passenger detection accuracyVSAvoidsystem structure complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The controller serves multiple functions: it processes data from weight sensors, microphones, and breath detection sensors; runs machine learning algorithms for pattern recognition; and coordinates alarm signaling. This multi-functional design consolidates complexity into a single central processing unit rather than requiring separate dedicated systems for each function.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The machine learning algorithms automatically learn and adapt to distinguish genuine trapped passenger scenarios from false alarm conditions without requiring manual programming of every possible scenario. The system self-adjusts its detection parameters based on learned patterns, reducing the need for complex manual configuration and maintenance.

Inventive Principle:
Principle #25Self-service

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

The system effectively detects trapped passengers and elevator malfunctions, ensuring rapid response and rescue, reducing false alarms, and enhancing passenger safety by integrating multiple sensors and centralized monitoring.

Implementation Method 1

The elevator cab (102) includes a jerk detection sensor (106) that detects a jerk in the elevator cab

Methodology Applied
Scientific EffectAcceleration detection: Accelerometer

Implementation Method 2

a microphone (107) that detects a sound generated by a trapped passenger

Methodology Applied
Scientific EffectSound wave detection: Sound

Implementation Method 3

a breath detection sensor (109) that detects a breath of the trapped passenger

Methodology Applied
Scientific EffectBreath detection: Absorption Spectroscopy

Data Source

PatentEP3459888B1System and method for detection of a malfunction in an elevator
Publication Date: 2026.02.25 OTIS ELEVATOR CO
  • EP3459888B1 patent drawingFigure 1
  • EP3459888B1 patent drawingFigure 2
  • EP3459888B1 patent drawingFigure 3

AI summary

A system and method for detection of a malfunction in an elevator. The system includes an elevator cab, a server, at least one sensor, and a controller. The controller receives a signal captured by the at least sensor and transmit to the server. The server processes the transmitted signal and prognostically detects the malfunction in the elevator cab.