• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.720-723
• /
• 2005

Ship motion is a complex controlled process with several hydrodynamic parameters that vary in wide ranges with respect to ship load condition, speed and surrounding conditions (such as wind, current, tide, etc.). Therefore, to effectively control ships in a designed track is always an important task for ship masters. This paper presents an effective adaptive autopilot ships that ensure the optimal accuracy, economy and stability characteristics. The PID control methodology is modified and parameters of a PID controller is designed to satisfy conditions for an optimal objective function that comprised by heading error, resistance and drift during changing course, and loss of surge velocity or fuel consumption. Designing of the controller for course changing process is based on the Model Reference Adaptive System (MRAS) control theory, while as designing of the automatic course keeping process is based on the Self Tuning Regulator (STR) control theory. Simulation (using MATLAB software) in various disturbance conditions shows that in comparison with conventional PID autopilots, the designed autopilot has several notable advantages: higher course turning speed, lower swing of ship bow even in strong waves and winds, high accuracy of course keeping, shorter time of rudder actions smaller times of changing rudder direction.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권6호
• /
• pp.2824-2837
• /
• 2019

Container technologies are widely used in infrastructures to deploy and manage applications in cloud computing environment. As containers are light-weight software, the cluster of cloud applications can easily scale up or down to provide Internet-based services. Container-based applications can well deal with fluctuate workloads by dynamically adjusting physical resources. Current works of scheduling applications often construct applications' performance models with collected historical training data, but these works with static models cannot self-adjust physical resources to meet the dynamic requirements of cloud computing. Thus, we propose a self-adaptive automatic container scheduling framework AutoScale for cloud applications, which uses a feedback-based approach to adjust physical resources by extending, contracting and migrating containers. First, a queue-based performance model for cloud applications is proposed to correlate performance and workloads. Second, a fuzzy Kalman filter is used to adjust the performance model's parameters to accurately predict applications' response time. Third, extension, contraction and migration strategies based on predicted response time are designed to schedule containers at runtime. Furthermore, we have implemented a framework AutoScale with container scheduling strategies. By comparing with current approaches in an experiment environment deployed with typical applications, we observe that AutoScale has advantages in predicting response time, and scheduling containers to guarantee that response time keeps stable in fluctuant workloads.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권9호
• /
• pp.4665-4683
• /
• 2019

Previous methods build image annotation model by leveraging three basic dependencies: relations between image and label (image/label), between images (image/image) and between labels (label/label). Even though plenty of researches show that multiple dependencies can work jointly to improve annotation performance, different dependencies actually do not "work jointly" in their diagram, whose performance is largely depending on the result predicted by image/label section. To address this problem, we propose the adaptive attention annotation model (AAAM) to associate these dependencies with the prediction path, which is composed of a series of labels (tags) in the order they are detected. In particular, we optimize the prediction path by detecting the relevant labels from the easy-to-detect to the hard-to-detect, which are found using Binary Cross-Entropy (BCE) and Triplet Margin (TM) losses, respectively. Besides, in order to capture the inforamtion of each label, instead of explicitly extracting regional featutres, we propose the self-attention machanism to implicitly enhance the relevant region and restrain those irrelevant. To validate the effective of the model, we conduct experiments on three well-known public datasets, COCO 2014, IAPR TC-12 and NUSWIDE, and achieve better performance than the state-of-the-art methods.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2002년도 ICCAS
• /
• pp.120.2-120
• /
• 2002

In this paper auto-tuning scheme of weight function in the neural networks has been suggested by immune algorithm for nonlinear process. A number of structures of the neural networks are considered as learning methods for control system. A general view is provided that they are the special cases of either the membership functions or the modification of network structure in the neural networks. On the other hand, since the immune network system possesses a self organizing and distributed memory, it is thus adaptive to its external environment and allows a PDP (parallel distributed processing) network to complete patterns against the environmental situation. Also. It can provi..

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1989년도 한국자동제어학술회의논문집; Seoul, Korea; 27-28 Oct. 1989
• /
• pp.1113-1119
• /
• 1989

Neural network control has many innovative potentials for intelligent adaptive control. Among many, it promises real time adaption, robustness, fault tolerance, and self-learning which can be achieved with little or no system models. In this paper, a dynamic robot controller has been developed based on a backpropagation neural network. It gradually learns the robot's dynamic properties through repetitive movements being initially trained with a PD controller. Its control performance has been tested on a simulated PUMA 560 demonstrating fast learning and convergence.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2002년도 ITC-CSCC -3
• /
• pp.1491-1494
• /
• 2002

The paper introduces a novel system of two hands real-time tracking based on the unrestricted hand skin segmentation by multi color systems. After corer-based segmentation and pre-processing operation, a label set of regions is created to locate the two hands automatically. By the normalization, template matching is used to find out the left or right hand. An improved fast self-adaptive tracking algorithm is applied and Canny filter is used for hand detection.

• Smart Structures and Systems
• /
• 제8권1호
• /
• pp.107-117
• /
• 2011

Biological sensory systems continuously monitor and analyze changes in real-world environments that are relevant to an animal's specific behavioral needs and goals. Understanding the sensory mechanisms and information processing principles that biological systems utilize for efficient sensory data acquisition may provide useful guidance for the design of smart-sensing systems in engineering applications. Weakly electric fish, which use self-generated electrical energy to actively sense their environment, provide an excellent model system for studying biological principles of sensory data acquisition. The electrosensory system enables these fish to hunt and navigate at night without the use of visual cues. To achieve reliable, real-time task performance, the electrosensory system implements a number of smart sensing strategies, including efficient stimulus encoding, multi-scale virtual sensor arrays, task-dependent filtering and online subtraction of sensory expectation.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2004년도 ICCAS
• /
• pp.52-56
• /
• 2004

Unmanned Aerial Vehicles (UAVs) are remotely piloted or self-piloted aircraft by inputted program in advance or artificial intelligence. In this study Aileron and Elevator are used to control the movement of airplane for horizontal and vertical flights about its longitudinal and lateral axis. In an introduction, the drone was linearly modeled by extracting aerodynamic parameter through flight test and simulation, lift and drag coefficient corresponding to angle of attack, changes of pitching moment coefficient. In the main subject, the flight simulation was performed after constructing hardware using TMS320F2812 from TI company and PID with lateral and longitudinal controller for horizontal and vertical flights. Flying characteristics of two system were estimated and compared through real flight test with hardware equipped algorithm and adaptive algorithm that was applied to consider external factors such as turbulence. In conclusion the control performance of the controller with proposed algorithm was streamlined at lateral and longitudinal controller respectively, we will discuss guidance command to pass way point.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.121.1-121
• /
• 2001

Congestion control is a critical issue in TCP networks, Recently, active queue management (AQM) was proposed for congestion control at routers. The random early detection RED algorithm is widely known in the AQM algorithms, We present an adaptive nonlinear RED (NRED) algorithm, which has nonlinear drop probability profile. The proposed algorithm enhanced the performance of the RED algorithm by the self-parameterization based on the traffic load Furthermore, the proposed algorithm can effectively adapt itself between he RED and the drop-tail queue management by adopting proper nonlinearity in the drop probability profile. Through simulation, we show the effectiveness of the proposed algorithm comparing with the drop-tail and the original RED algorithm.

• 정보처리학회논문지D
• /
• 제12D권4호
• /
• pp.647-656
• /
• 2005

본 논문에서는 분산된 관리대상의 시스템자원과 사용자정보, 사용패턴을 Context로 수집하여, 구성 (Configuration)을 수행하는 적응형 자가관리시스템을 제안한다. 본 시스템은 기존에 수동으로 이루어지던 Configuration작업들(Install, Reconfiguration, Update)을 자율적으로 수행하여, 사용자의 시스템관리에 대한 부담을 줄여주게 되며, 많은 비용과 오류를 감소시켜준다. 본 시스템은 수집된 Context정보를 기반으로 사용자의 환경에 맞는 구성요소를 선택하여 설치하게 되며, 사용자의 기존 애플리케이션의 환경설정과 사용패턴을 기반으로, 보다 개인화된 설정을 해준다. 설정 이후에는 사용자의 행동을 암시적 피드백으로 받아, 이를 학습하고 유사한 상황이 다시 발생할 경우, 이를 다음 행동에 반영한다. 그리고 기존에 중앙서버로부터 일률적으로 관련파일을 전송하고 관리하는 중앙집중배포방식의 여러 문제점에 대응하기 위해 Peer-to-Peer방식으로 파일을 복사하고, 이를 통해 중앙서버의 과부하를 줄이는 동시에 빠른 파일의 배포가 가능하도록 하였다. 본 시스템의 평가를 위해 프로토타입을 구현하여, 기존 수동 Configuration작업, MS-IBM의 관련시스템과 비교를 수행하였으며, 기능적 측면과 작업에 소요되는 시간에 대한 비교결과를 통해 본 시스템의 유효성을 증명하였다.