Elevator Audio-Visual Floor Name Detection for Indoor Positioning

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing indoor positioning systems struggle to consistently provide vertical position as floor names across different buildings and venues due to inconsistent naming conventions, making manual entry of floor names impractical.

Innovation Solution

Utilizing audio and visual samples to identify floor names through machine learning models, and integrating these into positioning maps to provide vertical location as floor names, with error handling for inconsistent estimates.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual determination and entry of floor names is performed for individual buildings and venues, then positioning accuracy can be maintained, but the process becomes untenable due to the large number of buildings and venues

Engineering Contradiction:

Improvepositioning accuracyVSAvoidfloor name entry efficiency

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The system enables automatic self-determination of floor names through audio and visual sample analysis. User devices capture audio samples (elevator announcements) and visual samples (floor indicators), which are then processed by machine learning models to automatically identify and assign floor names without manual intervention, making the system self-sufficient for large-scale deployment

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

Manual mechanical processes of floor name entry are replaced with automated electronic processing. Machine learning models analyze audio and visual data electronically to determine floor names, substituting the manual mechanical act of typing and entering floor names into databases with automated computational analysis

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Adaptability or versatility

If floor names are consistently provided across different buildings and venues, then user experience is improved, but inconsistent naming conventions make this difficult to achieve

Engineering Contradiction:

Improvefloor name consistencyVSAvoidnaming convention management

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system changes the parameter of floor name identification from manual convention-based assignment to automated pattern recognition. Machine learning models detect and standardize floor naming patterns from diverse sources (elevator announcements, visual indicators), transforming inconsistent local naming conventions into consistent standardized floor names across different buildings and venues

Inventive Principle:

Principle #35Parameter changes

3Extent of automation

If audio and visual samples are analyzed using machine learning models, then automated floor name determination is achieved, but system complexity increases

Engineering Contradiction:

Improvefloor name determination automationVSAvoidsystem architecture

Core Design Contradiction:

Extent of automationVSDevice complexity

Solution Approach 1:

The complex automated floor name determination system is segmented into distinct functional modules: audio sample capture and processing module, visual sample capture and processing module, machine learning analysis module, and floor name assignment module. This segmentation allows each component to be developed, tested, and maintained independently, managing overall system complexity

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Machine learning models serve as intermediary components that bridge the gap between raw audio/visual data and standardized floor names. These intermediaries process and interpret unstructured data from elevators and visual indicators, transforming them into consistent floor name identifiers without requiring direct complex processing of all input data

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS12473174B2Determining floor name based on audio and/or visual samples

Publication Date: 2025.11.18 HERE GLOBAL BV

AI summary

A processor obtains an audio sample captured by an audio sensor associated with a device and/or image data captured by an image sensor associated with the device. The image data comprises a representation of one or more elevator buttons. The processor analyzes the audio sample to identify a floor name indicator in the audio sample and/or the image data to identify an activated elevator button from among the one or more elevator buttons represented by the image data; and determines a floor name based on the floor name indicator identified in the analyzed audio sample and/or a floor name indicator associated with the identified activated elevator button in accordance with the analyzed image data. The processor causes the determined floor name to be stored and/or provided for positioning purposes.