Wearable Robot Hazard Detection Using Safety Templates and User Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Wearable robots like GEMS require a safety function to predict and react to dangerous situations in the user's surroundings, but existing technologies lack effective methods to identify and respond to potential hazards in real-time.
Innovation Solution
An electronic apparatus with a communication interface and processor that receives sensing data from user devices, identifies context information, and controls the wearable robot or user device based on safety template information to prevent dangerous situations, using AI models for accurate hazard detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If wearable robot directly couples to user's body to provide gait enhancement services, then user mobility assistance is improved, but user safety risk increases due to potential dangerous situations in surrounding environment
Solution Approach 1:
The patent introduces user device (smartphone, smartwatch, or PC) as an intermediary between the wearable robot and the surrounding environment. The user device captures sensing data (images, audio, sensor data) and transmits it to the wearable robot, enabling the robot to perceive dangerous situations without directly sensing the environment itself. This intermediary approach enhances safety while maintaining the direct body coupling for mobility assistance.
Solution Approach 2:
The system performs preliminary safety assessment by continuously analyzing sensing data from the user device before dangerous situations occur. The processor identifies potential hazards (obstacles, unsafe environments) in advance and prepares control information to prevent accidents. This preliminary detection and response mechanism ensures user safety while the robot provides gait enhancement services.
2Measurement precision
If wearable robot uses multiple sensing data from user devices to identify dangerous situations, then safety detection accuracy is improved, but system complexity increases
Solution Approach 1:
The patent employs a universal processing architecture where a single processor in the wearable robot handles multiple types of sensing data (image data, audio data, sensor data) from various user devices. The processor universally applies safety template information to different data types, identifying dangerous situations across multiple sensing modalities without requiring separate processing systems for each device, thus managing complexity while maintaining high detection accuracy.
Solution Approach 2:
The user device acts as an intermediary that consolidates and transmits multiple sensing data streams to the wearable robot. Instead of the robot directly integrating sensors in the wearable device, the user device collects and processes data from various sources (camera, microphone, sensors) and relays it to the robot's processor, simplifying the robot's structure while maintaining comprehensive hazard detection capabilities.
3Speed
If wearable robot responds quickly to dangerous situations by controlling user device, then user safety reaction time is improved, but control system complexity increases
Solution Approach 1:
The system implements a feedback mechanism where the processor continuously monitors sensing data, compares it against safety template information, and automatically generates control information when dangerous situations are detected. The control information is immediately transmitted to the user device or wearable robot to execute safety actions (stopping movement, alerting user, adjusting gait). This closed-loop feedback ensures rapid response while maintaining relatively simple control logic based on pre-defined safety templates.
Data Source
AI summary
An electronic apparatus includes a communication interface and at least one processor configured to obtain safety template information. The at least one processor may, based on sensing data being received from a user device, identify the type of the context information corresponding to the at least one sensing data, identify the information about the dangerous situations based on the safety template information and the type of the context information, identify whether a surrounding environment of a wearable robot corresponds to a dangerous situation based on the at least one sensing data; based on identifying that the surrounding environment of the wearable robot corresponds to the dangerous situation, obtain the control information mapped to each dangerous situation based on the safety template information, and control at least one of the wearable robot or the at least one user device based on the obtained control information.