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.
Highlights
Introduction and Enric CerveraUnmanned aerial vehicles (UAVs) have revealed their unprecedented potential for commercial, military, and civil-government utilization in a wide range of applications such as infrastructure inspection [1], aerial photography [2], logistics [3], and so forth
A unmanned aerial vehicles (UAVs) is capable of assisting the surveillance activities by its agile maneuverability to approach confined areas of low accessibility and its visual functionality to capture the remote scene in real-time
Real-time, learning-based object detection algorithm is integrated with the UAV embedded system to autonomous locate the desired object without human interference; a 3D pose tracking algorithm with object detection, stereo reconstruction techniques, and Kalman filter is implemented in a low-cost UAV system to recognize, locate and track the target object autonomously; whilst an UAV path planning is included for surveillance mission, which obeys the dynamic constraints for UAV to track and follow the target object movement; system experiments include both dynamic object and dynamic sensor, and the results validated good performance of the proposed system
Summary
Unmanned aerial vehicles (UAVs) have revealed their unprecedented potential for commercial, military, and civil-government utilization in a wide range of applications such as infrastructure inspection [1], aerial photography [2], logistics [3], and so forth. Chung et al [4] implemented the standard but relatively old techniques based on background subtraction and frame differencing to detect objects from an aerial robot These methods work poorly with a moving UAV that has high-frequency vibrations in the camera motion. A 3D pose tracking algorithm with object detection, stereo reconstruction techniques, and Kalman filter is implemented in a low-cost UAV system to recognize, locate and track the target object autonomously; whilst an UAV path planning is included for surveillance mission, which obeys the dynamic constraints for UAV to track and follow the target object movement; system experiments include both dynamic object and dynamic sensor, and the results validated good performance of the proposed system. The video footage of experiments and implementation codes are attached in the supplementary material
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
