A. Mihailidis

Sequential image classification of human-robot walking environments using temporal neural networks

The paper compares static and sequential image classification of walking environments using temporal neural networks, finding that 3D models outperform 2D ones but have higher computational demands.

AI-powered smart glasses for sensing and recognition of human-robot walking environments

This paper develops AI-powered smart glasses for real-time environment sensing and recognition in human-robot walking scenarios, achieving 93.6% accuracy with low latency.

Efficient Visual Perception of Human-Robot Walking Environments Using Semi-Supervised Learning

This paper introduces a semi-supervised learning system for visual perception in human-robot walking environments, achieving high accuracy with fewer labeled images.

Preliminary Development of a Robotic Hip-Knee Exoskeleton with 3D-Printed Backdrivable Actuators

This paper describes the preliminary development of a modular robotic hip-knee exoskeleton with 3D-printed backdrivable actuators, focusing on design, integration, and potential applications in rehabilitation.

Development and Mobile Deployment of a Stair Recognition System for Human-Robot Locomotion

The paper develops and deploys a lightweight deep learning model for real-time stair recognition on mobile devices with high accuracy and low latency.

Continuous Prediction of Leg Kinematics during Walking using Inertial Sensors, Smart Glasses, and Embedded Computing

The paper develops KIFNet, a deep learning model that continuously predicts leg kinematics during walking using sensor fusion and embedded computing, achieving high accuracy and speed.

Simulation of Energy Regeneration in Human Locomotion for Efficient Exoskeleton Actuation

This study simulates energy regeneration in human locomotion to improve exoskeleton efficiency, finding that slower speeds and decline slopes significantly extend battery life.

Stair Recognition for Robotic Exoskeleton Control using Computer Vision and Deep Learning

The paper develops a deep learning system for stair recognition in robotic exoskeleton control, achieving 98.4% classification accuracy using a new dataset called StairNet.