Human Tracking Robot Using Deep Learning for Real-Time Obstacle-Aware Navigation and Path Optimization

DOI

10.2991/978-94-6463-858-5_175How to use a DOI?

Keywords

NVIDIA; Accuracy; Sensor; Outdoor

Abstract

Human-following mobile robots have gained significant attention in robotics research due to their applications in assistance, surveillance, and industrial automation. This paper presents an embedded-based human-following mobile robot that leverages an RGB-D camera and deep learning-based human detection for real-time tracking and navigation. The proposed system utilizes YOLOv5 for human detection, offering high accuracy and low latency, making it suitable for embedded platforms. An NVIDIA Jetson Xavier NX serves as the processing unit, enabling efficient execution of deep learning models. The system integrates depth-based distance estimation to maintain a safe following distance while implementing obstacle avoidance mechanisms using ultrasonic sensors and depth data. The robot’s motion is controlled via ROS (Robot Operating System), ensuring smooth trajectory adjustments. Experimental results demonstrate that the proposed system achieves high detection accuracy (93%), fast processing time (80ms), and low power consumption (12W), making it an efficient and robust solution for real-world human-following applications. Future improvements will focus on multi-sensor fusion, reinforcement learning-based adaptive navigation, and outdoor environment adaptation.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - B. Sujitha
AU  - S. Rajamohan
AU  - D. Saffiyulla
AU  - S. Sanjeev
AU  - R. Vignesh
PY  - 2025
DA  - 2025/11/04
TI  - Human Tracking Robot Using Deep Learning for Real-Time Obstacle-Aware Navigation and Path Optimization
BT  - Proceedings of International Conference on Computer Science and Communication Engineering (ICCSCE 2025)
PB  - Atlantis Press
SP  - 2097
EP  - 2111
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-858-5_175
DO  - 10.2991/978-94-6463-858-5_175
ID  - Sujitha2025
ER  -