• Journal of the Korea Institute of Military Science and Technology
• /
• v.21 no.2
• /
• pp.147-157
• /
• 2018

There are increasing demands to provide early warning against intruding drones and cope with potential threats. Commercial anti-drone systems are mostly based on simple target detection by radar reflections. In real scenario, however, it becomes essential to obtain drone radar signatures so that hostile targets are recognized in advance. We present experimental test results that micro-Doppler radar signature delivers partial information on multi-rotor platforms and exhibits limited performance in drone recognition and classification. Afterward, we attempt to generate high resolution profile of flying drone targets. To this purpose, wide bands radar signals are employed to carry out inverse synthetic aperture radar(ISAR) imaging against moving drones. Following theoretical analysis, experimental field tests are carried out to acquire real target signals. Our preliminary tests demonstrate that high resolution ISAR imaging provides effective measures to detect and classify multiple drone targets in air.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.14 no.4
• /
• pp.1-7
• /
• 2018

The floating sculptures in the form of ad-ballon commonly used ropes in order to hold on. Relatively air flow is much less indoor than outdoor. Users of buoyancy sculptures hope to be able to maintain their desired posture without being fixed. This study applied drone technology to buoyancy sculptures. The drones can be moved vertically and horizontally, and the posture can be maintained, so buoyancy sculptures are easy to apply. Therefore, we have studied the control system of buoyancy sculpture using drone technology. Also, a control system that can maintain the desired posture at a constant height was studied. The overall shape was a light fiber material and helium gas for zero buoyancy to support the sculpture. The system configuration was STM32F103CB from ARM. In addition, the gyro and acceleration, geomagnetic sensors and motors are composed of small and medium size BLDC motors. The scheduling of the control system in the configuration of the control device was carefully considered. Because the role of the whole component becomes very important. The communication between the components is divided into the sensor fusion and the interface communication with the whole controller. Each communication technology is designed to expand. This study was implemented to actively respond from the viewpoint of posture control using the drone technology.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.12
• /
• pp.4795-4815
• /
• 2020

Drone detection can be considered as a specific sort of small object detection, which has always been a challenge because of its small size and few features. For improving the detection rate of drones, we design a Deeper SSD network, which uses large-scale input image and deeper convolutional network to obtain more features that benefit small object classification. At the same time, in order to improve object classification performance, we implemented the up-sampling modules to increase the number of features for the low-level feature map. In addition, in order to improve object location performance, we adopted the down-sampling modules so that the context information can be used by the high-level feature map directly. Our proposed Deeper SSD and its variants are successfully applied to the self-designed drone datasets. Our experiments demonstrate the effectiveness of the Deeper SSD and its variants, which are useful to small drone's detection and recognition. These proposed methods can also detect small and large objects simultaneously.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.14 no.6
• /
• pp.461-466
• /
• 2021

Drone can be controlled by the method such as Bluetooth communication for close distance and can be controlled through network communication for long distance. Especially, the coordinate is set using GPS and drone is controlled using network communication and video communication when the activity range is long distance. However, the drone should be controlled by receiving control authority accordingly in response about it appropriately when the drone leaves the control area after arriving at the destination if there is a problem with network communication and video communication. So, this study proposes a method to control a drone with a simple mutually promised simple gesture and the drone can be controlled in the proposed method even if the drone leaves from the control authority in above situation. The reference posture was established for mutually promised simple gesture algorithm and automatically handed over the control authority of drone to a person who takes the reference posture when the drone recognizes it to implement this. And all the movements of the drone could be controlled by starting the beginning of all commands from the reference posture (The hovering posture of the drone). Lastly, the control authority of the drone should be returned after achieving the purpose, and the algorithm was implemented to make the drone can perform next action of its own, and it was confirmed that the drone was operating normally by the mapped instruction.

• The Journal of Korea Robotics Society
• /
• v.12 no.4
• /
• pp.425-431
• /
• 2017

As drones gain more popularity these days, drone detection becomes more important part of the drone systems for safety, privacy, crime prevention and etc. However, existing drone detection systems are expensive and heavy so that they are only suitable for industrial or military purpose. This paper proposes a novel approach for training Convolutional Neural Networks to detect drones from images that can be used in embedded systems. Unlike previous works that consider the class probability of the image areas where the class object exists, the proposed approach takes account of all areas in the image for robust classification and object detection. Moreover, a novel loss function is proposed for the CNN to learn more effectively from limited amount of training data. The experimental results with various drone images show that the proposed approach performs efficiently in real drone detection scenarios.

• Journal of the Korean Society of Industry Convergence
• /
• v.26 no.4_2
• /
• pp.687-696
• /
• 2023

In recent construction sites, digital maps obtained through drone photogrammetry have garnered increasing attention as indispensable tools for effective construction site management. the strategic placement of Ground Control Points (GCPs) is crucial in drone photogrammetry. Nevertheless, the manual labor and time-intensive nature of GCP surveying pose significant challenges. The purpose of this study is to design the concept of automated GCPs survey technology for enhancing drone photogrammetry efficiency in construction sites. As a result, the productivity of the automated method was analyzed as 118,894.7㎡/hr. It is over 25% productivity improvement compared to traditional methods. In future studies, economic analysis of automated methods should be studied.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2017.11a
• /
• pp.134-137
• /
• 2017

Follow-me function of drones is new and attractive for selfie drone users, where the drone autonomously follows and capture the user. Currently the products use the difference between GPS's in the drone and user side mobile GCS, but the targeting accuracy is not satisfactory owing to the low accuracy of GPS data, often the order of ten meters. We designed a new follow-me mode algorithm that utilizes the accuracy of visual tracking algorithm and the reliability of GPS-based. The experiment shows that proposed follow-me can capture much accurately the target user in the center of video content than GPS-only methods, and recover the vision algorithm failure quickly in 5-10 seconds.

• Journal of Institute of Convergence Technology
• /
• v.6 no.1
• /
• pp.43-47
• /
• 2016

In this paper, we report a work-in-progress about development of multi-drone system for smart agriculture. The multi-drone system is controlled via a haptic teleoperation by a human operator. The purpose of the multi-drone system is that let the human operator to easily handle the multiple drones which are maintaining a fixed formation using ZigBee communication network.

• Journal of Advanced Engineering and Technology
• /
• v.11 no.4
• /
• pp.273-277
• /
• 2018

In the field of forest disaster prevention, drones are expected to save higher human resources than the existing manpower has, and produce high-efficiency results over time. However, operational limitations brought by short flight times have brought the environment of limited use of the various capabilities of the drone, and the existing development systems operating the multi-drone are mainly for performance purpose, so it is a difficult to use for forest disaster prevention. The purpose of this paper is to design the concept based on multi-drone operation procedure through analysis of mission of ground control system for forest disaster prevention.

• Journal of the Semiconductor & Display Technology
• /
• v.20 no.3
• /
• pp.117-119
• /
• 2021

With the development of the industry, the use of drones in specific mission flight is being actively studied. These drones fly a specified path and perform repetitive tasks. if the drone system will detect objects in real time, the performance of these mission flight will increase. In this paper, we implement object detection system and mount GPU acceleration to maximize the efficiency of limited device resources with drone video using Tensorflow Lite which enables in-device inference from a mobile device and Mobile SDK of DJI, a drone manufacture. For performance comparison, the average processing time per frame was measured when object detection was performed using only the CPU and when object detection was performed using the CPU and GPU at the same time.