کنفرانس مهندسی برق ایران

صفحه اصلی / سی و سومین کنفرانس بین المللی مهندسی برق

Human detection and following by a mobile robot using YOLO structured convolutional neural network

نویسندگان :

Yasan Majidi¹ Amir Hossein Hassanabadi²

1- Department of Electrical Engineering, Qazvin Branch, Islamic Azad University 2- Department of Electrical Engineering, Qazvin Branch, Islamic Azad University

کلمات کلیدی :

Convolutional neural networks،YOLO،Object tracking،Target Recovery،Mobile Robot،Visual Similarity Matching،Real-time computer vision

چکیده :

In mobile robot applications within home environments, losing track of a target can result in permanent loss, significantly affecting performance. This research enhances the YOLO (You Only Look Once) algorithm by integrating a target recovery mechanism to address temporary target loss. When the robot loses sight of the target, a visual similarity matching algorithm is used to recover the target once it returns to the robot's field of view. The proposed method was implemented and tested on a TurtleBot3 robot. Results demonstrate that this approach successfully adds a recovery capability to the YOLO algorithm while requiring fewer computational resources compared to other recovery methods. This makes it a practical solution for real-time applications in resource-constrained environments.