• 대한임베디드공학회논문지
• /
• 제16권5호
• /
• pp.163-169
• /
• 2021

In this paper, we implemented a deep learning-based automatic object tracking and handy motion control drone system and analyzed the performance of the proposed system. The drone system automatically detects and tracks targets by analyzing images obtained from the drone's camera using deep learning algorithms, consisting of the YOLO, the MobileNet, and the deepSORT. Such deep learning-based detection and tracking algorithms have both higher target detection accuracy and processing speed than the conventional color-based algorithm, the CAMShift. In addition, in order to facilitate the drone control by hand from the ground control station, we classified handy motions and generated flight control commands through motion recognition using the YOLO algorithm. It was confirmed that such a deep learning-based target tracking and drone handy motion control system stably track the target and can easily control the drone.

• 한국정밀공학회지
• /
• 제26권1호
• /
• pp.89-96
• /
• 2009

This paper describes the intelligent robot's hand with three finger sensors for a humanoid robot. In order to grasp an unknown object safely, the intelligent robot's hand should measure the mass of the object, and determine the grasping force using the mass, finally control the grasping force using the finger sensors and the controller. In this paper, the intelligent robot's hand for a humanoid robot was developed. First, the six-axis force/moment sensor was manufactured. second, three finger force sensors were designed and fabricated, third, the high-speed controller was manufactured using DSP(digital signal processor), finally, the characteristic test for determining a grasping force and for grasping an unknown object safely It is confirmed that the hand could grasp an unknown object safely.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권1호
• /
• pp.272-287
• /
• 2017

This paper proposes new methods, named Derivative Code (DerivativeCode) and Derivative Code Pattern (DCP), for object recognition. The discriminative derivative code is used to capture the local relationship in the input image by concatenating binary results of the mathematical derivative value. Gabor based DerivativeCode is directly used to solve the palmprint recognition problem, which achieves a much better performance than the state-of-art results on the PolyU palmprint database. A new local pattern method, named Derivative Code Pattern (DCP), is further introduced to calculate the local pattern feature based on Dervativecode for object recognition. Similar to local binary pattern (LBP), DCP can be further combined with Gabor features and modeled by spatial histogram. To evaluate the performance of DCP and Gabor-DCP, we test them on the FERET and PolyU infrared face databases, and experimental results show that the proposed method achieves a better result than LBP and some state-of-the-arts.

• 한국측량학회지
• /
• 제8권1호
• /
• pp.15-22
• /
• 1990

본 연구에서는 근접사진측량에 의하여 대상물의 전면에 대한 동시 정밀 3차원 위치해석의 가능성을 제시하는데 목적을 두었으며, 이를 위하여 network의 imaging geometry, 기준점 배치 등 기하학적 배치를 고려하여 대상물의 전면을 피복하는 스트립과 블럭을 형성, 번들조정기법으로 해석을 시도하였다. 그 결과, 양호한 정학도로 전면해석의 가능성을 입증할 수 있었고, network의 기하학적 배치에 따른 정확도의 특성을 도출하였다. 또한 국부적으로 배치된 소수의 기준점으로도 소기의 정학도를 얻을 수 있었으므로 전면해석을 위한 기준점 측량의 난제를 크게 줄일 수 있을 것으로 사료된다.

• 전자공학회논문지SC
• /
• 제47권2호
• /
• pp.55-59
• /
• 2010

현재 움직이는 물체의 운동을 제어하기 위해 하중이라는 개념의 사용이 시도되고 있다. 하중이라는 제어 개념으로 불확실성으로 인한 비선형성을 포함한 시스템에서 보다 신뢰성 있는 제어값을 찾기 위해, 기존의 선형화 방법에 의한 상태방정식에서 제어값을 찾는 문제와 달리 비선형 하중 모델에서 직접 제어 값을 찾는 방법이다. 본 논문에서는 동력이 없고 제어되지 않는 자유 낙하 물체를 비선형 하중모델로 구현해보고, 외란에 따라 낙하하는 물체에 작용하는 하중값을 분석해 본다. 또한 이러한 외란에 의해 발생한 하중값에 따른 낙하 경로를 구한다. 본 논문의 연구결과는 비선형 모델 예측 제어 시스템에 적용하여 다 연장로켓, 추진력을 갖지 않은 GPS 유도탄 그리고 우주비행체의 지상귀환 제어시스템의 정밀도 향상을 위해 이용될 수 있다.

• 한국정보처리학회논문지
• /
• 제6권11S호
• /
• pp.3340-3349
• /
• 1999

Supporting object group on distributed object system give merits such as load balancing, fault tolerance and high availability. In this paper, we describe a CORBA ORB that has been designed to support object group based on active replication. The ORB supports the operational model in which it uses the IIOP for communication between client and server and total ordered multicast protocol for consistency control among group members. And through extension of ORB, it provides functions required for support of object group. Since it provides transparency of object replication, the ORB is interoperable with the existing CORBA products. It make possible for existing server application to be easily extended to application supporting object group as adding interface functions which should be used for building applications is minimized. A prototype is implemented, and performance of the replicated object group is tested and compared with a single object invocation.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.135-135
• /
• 2001

In recent years there have been increasing interests in real-time object tracking with image information. Since image information is affected by illumination, this paper presents the real-time object tracking method based on neural networks that have robust characteristics under various illuminations. This paper proposes three steps to track the object and the fast tracking method. In the first step the object color is extracted using neural networks. In the second step we detect the object feature information based on invariant moment. Finally the object is tracked through a shape recognition using neural networks. To achieve the fast tracking performance, we have a global search for entire image and then have tracking the object through local search when the object is recognized.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1999년도 제14차 학술회의논문집
• /
• pp.262-265
• /
• 1999

This paper describes a new method for object extraction necessary for image tracking systems. The extraction method which this paper proposes here is that an M-array is set between a camera and the object and the obtained image including the object and M-array is pro-cessed for extracting the object. The image processing utilizes a characteristic of M-array which is robust to noise. When an M-array is overlapped on the object in background image, the object woud have a part of M-array, which is detected by use of partial correlation between the mosaic image of M-array and the standard M-array. Thus the shape and position of the object are extracted by extracting a common domain of width of high correlation value. Experiments are carried out by using an actual photo of Kumamoto city taken from an airplane as background, and by use of a rectangular and circular object. The results of experiment show a wide application of this method for practical image tracking systems.

• 정보기술과데이타베이스저널
• /
• 제7권1호
• /
• pp.17-26
• /
• 2000

To provide highly available services in the distributed object system, it is required to support the object group. The state machine approach and primary-backup approach are proposed as two representative approaches for support of object group. The primary-backup approach does not only give merits such as transparency of object group and non-deterministic execution but also require less resource than state machine approach. This paper describes an extension of CORBA that is required to support of the primary-backup object group. In this paper, the state of backup is synchronized with primary through the atomic multicast protocol whenever the request of client is executed at primary. As a result, it does not require message logging and check pointing. The object group of this paper also provides fast response time in case of failure of the primary since it makes primary election unnecessary. And through an extension of IDL, it makes possible to avoid consistency control depending on characteristic of application. A prototype has been implemented and the performance of object group has been compared with a single object invocation.

• Journal of Advanced Marine Engineering and Technology
• /
• 제20권3호
• /
• pp.110-120
• /
• 1996

The purpose of this study is to build up the control scheme that promptly controls the pressure in a hydraulic cylinder having small control volume, using a PCV(proportional control valve) and a digital computer. Object pressure control system has the character to be unstable easily, because the displacement-flow gain of the PCV is so large considering comparatively small volume of the hydraulic cylinder and the time delay of response of PCV is long. Considering the above-mentioned characteristics of the object pressure control system, in this study, a control system is designed with two degree of freedom scheme that is composed by adding a feed-forward control path to I-PD control system, and the reference model is used to decide control parameters. And through some experiments on FF-I-PD, the validity of this control method is confirmed.