Echolocation Sound Nodes for Visually Impaired Navigation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Individuals with no sight or limited sight face challenges in navigating three-dimensional spaces, both physically and virtually, due to the lack of effective auditory cues for spatial awareness.

Innovation Solution

A system that uses unsupervised machine learning to segment a digital representation of a three-dimensional space into depth planes, determines echo sound nodes based on object depth and surface area, and generates spatialized echolocation sounds from these nodes to assist users in navigating through the space.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If traditional navigation methods are used by visually impaired individuals, then they lack spatial awareness, but implementing auditory cue systems increases device complexity

Engineering Contradiction:
Improvespatial awarenessVSAvoidsystem complexity
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The system segments the three-dimensional space into multiple depth planes using unsupervised machine learning algorithms. Each depth plane represents a specific distance range from the user, allowing the system to process and prioritize spatial information at different depths separately. This segmentation enables efficient generation of echolocation sounds that correspond to objects at various distances, improving spatial awareness without requiring overly complex processing of the entire scene simultaneously.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system introduces echo sound nodes as intermediary elements that represent objects in the environment. These nodes are generated based on detected objects and their properties (depth, surface area), and they serve as mediators between the complex environmental data and the simplified auditory cues presented to the user. The nodes enable the system to translate complex spatial information into intuitive sound characteristics without directly exposing the user to the underlying complexity.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If detailed object information is processed to improve echolocation accuracy, then measurement precision increases, but computational requirements and processing time increase

Engineering Contradiction:
Improveobject detection precisionVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system applies partial action by focusing computational resources on processing objects that are most relevant to the user's navigation needs. Rather than processing all detected objects with equal detail, the system prioritizes objects based on their depth, surface area, and potential impact on navigation. This selective processing maintains measurement precision for critical objects while reducing overall computational burden and processing time.

Inventive Principle:
Principle #16Partial or excessive action

Solution Approach 2:

The system changes parameters such as depth plane segmentation and object priority thresholds dynamically based on the user's context and environmental conditions. By adjusting these parameters, the system can optimize the balance between measurement precision and processing speed, allocating more computational resources when high precision is needed and reducing processing when speed is more critical.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3864494B1Locating spatialized sounds nodes for echolocation using unsupervised machine learning
Publication Date: 2023.07.05 MICROSOFT TECHNOLOGY LICENSING LLC
  • EP3864494B1 patent drawingFigure 1
  • EP3864494B1 patent drawingFigure 2
  • EP3864494B1 patent drawingFigure 3

AI summary

Described herein is a system for generating echolocation sounds to assist a user having no sight or limited sight to navigate a three-dimensional space (e.g., physical environment, computer gaming experience, and/or virtual reality experience). Input is received from a user to generate echolocation sounds to navigate a three-dimensional space. Based at least on the received input, a digital representation of the threedimensional space is segmented into one or more depth planes using an unsupervised machine learning algorithm. For each depth plane, object segments are determined for each object within the particular depth plane. Locations of a plurality of echo sound nodes are determined in accordance with the depth level and surface area of each object defined by the determined segments. The echolocation sounds comprising a spatialized sound from each echo sound node originating from the determined location are generated.