A Review on Autonomous Drone : Object Detection And Avoidance

Authors

Kalrav Gediya, Pooja Bhatt, Swati Sharma

Abstract

Autonomous drones have become critical in modern
applications such as surveillance, search and rescue, and
logistics. A key challenge to their deployment lies in
ensuring safe navigation through object detection and
collision avoidance. This review presents an in-depth
analysis of recent advancements in autonomous drone
object detection and avoidance systems, focusing on
machine learning techniques, particularly deep learning,
and their integration with sensor data for real-time
decision-making. The review synthesizes findings from
25 significant studies published between 2020 and 2023,
covering both algorithmic developments and sensor
technologies. Key developments include the use of
convolutional neural networks (CNNs), reinforcement
learning (RL), and hybrid sensor systems that enhance
obstacle detection and path planning. The paper
highlights current limitations such as computational
constraints, small object detection, and real-time
processing challenges. Finally, the review explores
emerging trends such as 3D object detection and the role
of 6G networks in enhancing UAV(Unmanned Aerial
Vehicle) communication for collision avoidance. This
comprehensive review serves as a foundation for further
research, emphasizing the potential of AI-driven UAVs
in complex, dynamic environments.

Keywords

Drone, UAV (Unmanned Aerial Vehicle), CNN (convolutional neural networks), Reinforcement Learning, Sensors, Object Detection, Deep Learning

A Review on Autonomous Drone : Object Detection And Avoidance. Kalrav Gediya, Pooja Bhatt, Swati Sharma. 2024. IJIRCT, Volume 10, Issue 5. Pages 1-7. https://www.ijirct.org/viewPaper.php?paperId=2410001