Dynamic Object Tracking on Autonomous UAV System for Surveillance Applications

Abstract

The ever-burgeoning growth of autonomous unmanned aerial vehicles (UAVs) has demonstrated a promising platform for utilization in real-world applications. In particular, a UAV equipped with a vision system could be leveraged for surveillance applications. This paper proposes a learning-based UAV system for achieving autonomous surveillance, in which the UAV can be of assistance in autonomously detecting, tracking, and following a target object without human intervention. Specifically, we adopted the YOLOv4-Tiny algorithm for semantic object detection and then consolidated it with a 3D object pose estimation method and Kalman filter to enhance the perception performance. In addition, UAV path planning for a surveillance maneuver is integrated to complete the fully autonomous system. The perception module is assessed on a quadrotor UAV, while the whole system is validated through flight experiments. The experiment results verified the robustness, effectiveness, and reliability of the autonomous object tracking UAV system in performing surveillance tasks. The source code is released to the research community for future reference.

A Path Planning Method with Perception Optimization Based on Sky Scanning for UAVs

“…Actually, the traditional grid method plays the most basic role among the various modeling methods [ 36 , 37 ]. When modeling the environment, the scenarios are transformed into the traditional grids by which information is beneficial to be preserved in a computer.…”

Proactive Guidance for Accurate UAV Landing on a Dynamic Platform: A Visual–Inertial Approach

“…For instance, YOLOv4 was applied to develop an autonomous UAV system for electrical and mechanical devices in [ 26 ], and a support vector machine (SVM) was utilized in [ 27 ] for a quadrotor to track and follow a moving target. Additionally, the deep learning algorithm-based tracker was further adopted to conduct a surveillance mission on a UAV system [ 28 ]. However, severe motion blur could still be seen in most of the UAV’s landings, induced by the installation of cameras on the UAV.…”

Design of Gas Monitoring Terminal Based on Quadrotor UAV

“…However, in the future of rapid development, mobile terminals have been widely used in actual production and daily life [ 17 ], such as UAVs for intelligent agricultural irrigation [ 18 ] and real-time traffic monitoring [ 19 ] to strengthen the support of the environmental protection industry and accelerate the improvement of the technical equipment level of the environmental protection industry. In addition, Pretto et al from the Swiss Federal Institute of Technology in Zurich carried out research on the use of UAV systems for fine planting and proposed a plant protection scheme for aerial monitoring of vegetation using rotary-wing UAVs, which greatly reduces labor costs [ 20 ]. Some scholars have created a learning-based UAV system for autonomous surveillance, in which UAVs can be used to autonomously detect and track target objects without human intervention [ 21 ].…”