Conceptualization of Cloud-Based Motion Analysis and Navigation for Wearable Robotic Applications

Area of Science:

• Robotics
• Human-Computer Interaction
• Computer Science

Background:

• Pedestrian behavior in unconstrained environments is unpredictable.
• Wearable robotics, including lower-limb exoskeletons and active orthoses, require adaptable support for diverse walking activities like level walking and stair climbing.
• Predicting and supporting transitions between different gait activities is a significant challenge for current wearable robotic systems.

Purpose of the Study:

• To propose a novel cloud software system for wearable robotics.
• To enhance the adaptability of lower-limb exoskeletons and active orthoses by providing contextual awareness of pedestrian behavior.
• To address the challenge of unpredictable gait transitions in non-constrained environments.

Main Methods:

• Development of a cloud software system integrating geographical mapping techniques.
• Implementation of Human Activity Recognition (HAR) algorithms to interpret pedestrian behavior.
• Partial implementation and testing of the proposed system in a real-world context.

Main Results:

• The developed system provides hindsight information to give context to surrounding pedestrians.
• Initial testing indicates the system's concept is viable.
• The system demonstrates significant extensibility prospects for future development.

Conclusions:

• The proposed cloud-based system offers a promising approach to enhance the functionality of wearable robotics.
• Integrating geographical mapping and HAR can improve the adaptability of lower-limb exoskeletons and active orthoses.
• The system's ability to provide contextual information shows potential for safer and more intuitive human-robot interaction in dynamic environments.