Gaze-Guided Wearable Beamforming for Continuous Hazard Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Wearable devices like AR glasses struggle to maintain beamforming on external objects when users shift their gaze, leading to difficulties in identifying hazards such as approaching vehicles, and inefficient current consumption.
Innovation Solution
A wearable device with multiple microphones, a display module, and a processor that continuously performs beamforming on identified external objects, amplifies relevant notifications, and optimizes power usage by activating microphones close to the object.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If beamforming is performed on external objects to amplify sound signals, then the user can identify hazards better, but the beamforming is lost when the user rotates toward the object
Solution Approach 1:
The system uses gaze direction detection as feedback to dynamically adjust beamforming parameters. When the user's gaze is detected to be directed toward an external object, the system maintains or adjusts the beamforming configuration to ensure continuous audio notification delivery to the user, resolving the contradiction between hazard identification reliability and beamforming maintenance during rotation
Solution Approach 2:
The beamforming system is made dynamic by continuously adjusting the beam direction and configuration based on real-time gaze detection. This allows the beamforming to adapt to user rotation and maintain optimal orientation toward external objects of interest, ensuring continuous audio notification delivery while preserving hazard identification reliability
2Reliability
If all microphones are activated to ensure comprehensive audio detection, then audio coverage is improved, but current consumption increases
Solution Approach 1:
Instead of activating all microphones uniformly, the system selectively activates only the microphone(s) closest to the detected external object based on spatial information. This localizes the audio detection capability to the most relevant microphone, maintaining comprehensive audio coverage for the object of interest while significantly reducing overall current consumption
Solution Approach 2:
The system applies partial action by activating only a subset of microphones (specifically, only the microphone closest to the external object) rather than all microphones. This partial activation is sufficient for the intended purpose of detecting audio from the identified object, thereby reducing energy consumption while maintaining adequate audio detection coverage
Data Source
AI summary
Disclosed are a wearable device for performing continuous beamforming on at least one object, and a method of controlling the wearable device. The wearable device includes: a plurality of microphones; a display module; at least one speaker; at least one processor. The wearable device may perform beamforming on at least one object located around a user wearing the wearable device, by using at least one microphone; output, on the basis of performing the beamforming, a notification corresponding to the at least one object through at least one of the at least one speaker or the display module; identify movement of the wearable device rotating to face an object among a plurality of objects; and highlight and output, on the basis of identifying the movement, through at least one of the at least one speaker or the display module, the notification while maintaining the beamforming on the at least one object.