• Title/Summary/Keyword: gradient algorithm

Search Result 1,165, Processing Time 0.024 seconds

ON A VORTICITY MINIMIZATION PROBLEM FOR THE STATIONARY 2D STOKES EQUATIONS

  • KIM HONGCHUL;KWON OH-KEUN
    • Journal of the Korean Mathematical Society
    • /
    • v.43 no.1
    • /
    • pp.45-63
    • /
    • 2006
  • This paper is concerned with a boundary control problem for the vorticity minimization, in which the flow is governed by the stationary two dimensional Stokes equations. We wish to find a mathematical formulation and a relevant process for an appropriate control along the part of the boundary to minimize the vorticity due to the flow. After showing the existence and uniqueness of an optimal solution, we derive the optimality conditions. The differentiability of the state solution in regard to the control parameter shall be conjunct with the necessary conditions for the optimal solution. For the minimizer, an algorithm based on the conjugate gradient method shall be proposed.

Numerical Analysis of Shallow Water Equation with Fully Implicit Method (음해법을 이용한 천수방정식의 수치해석)

  • Kang, Ju Whan;Park, Sang Hyun;Lee, Kil Seong
    • KSCE Journal of Civil and Environmental Engineering Research
    • /
    • v.13 no.3
    • /
    • pp.119-127
    • /
    • 1993
  • Recently, ADI scheme has been a most common tool for solving shallow water equation numerically. But ADI models of tidal flow is likely to cause so called ADI effect in such a region of the Yellow Sea which shows complex topography and has submarine canyons especially. To overcome this, a finite difference algorithm is developed which adopts fully implicit method and preconditioned conjugate gradient squared method. Applying the algorithm including simulation of intertidal zone to Sae-Man-Keum. velocity fields and flooding/drying phenomena are simulated well in spite of complex topography.

  • PDF

Discriminative Training of Sequence Taggers via Local Feature Matching

  • Kim, Minyoung
    • International Journal of Fuzzy Logic and Intelligent Systems
    • /
    • v.14 no.3
    • /
    • pp.209-215
    • /
    • 2014
  • Sequence tagging is the task of predicting frame-wise labels for a given input sequence and has important applications to diverse domains. Conventional methods such as maximum likelihood (ML) learning matches global features in empirical and model distributions, rather than local features, which directly translates into frame-wise prediction errors. Recent probabilistic sequence models such as conditional random fields (CRFs) have achieved great success in a variety of situations. In this paper, we introduce a novel discriminative CRF learning algorithm to minimize local feature mismatches. Unlike overall data fitting originating from global feature matching in ML learning, our approach reduces the total error over all frames in a sequence. We also provide an efficient gradient-based learning method via gradient forward-backward recursion, which requires the same computational complexity as ML learning. For several real-world sequence tagging problems, we empirically demonstrate that the proposed learning algorithm achieves significantly more accurate prediction performance than standard estimators.

Optical Recognition of Credit Card Numbers (신용카드 번호의 광학적 인식)

  • Jung, Min Chul
    • Journal of the Semiconductor & Display Technology
    • /
    • v.13 no.1
    • /
    • pp.57-62
    • /
    • 2014
  • This paper proposes a new optical recognition method of credit card numbers. Firstly, the proposed method segments numbers from the input image of a credit card. It uses the significant differences of standard deviations between the foreground numbers and the background. Secondly, the method extracts gradient features from the segmented numbers. The gradient features are defined as four directions of grayscale pixels for 16 regions of an input number. Finally, it utilizes an artificial neural network classifier that uses an error back-propagation algorithm. The proposed method is implemented using C language in an embedded Linux system for a high-speed real-time image processing. Experiments were conducted by using real credit card images. The results show that the proposed algorithm is quite successful for most credit cards. However, the method fails in some credit cards with strong background patterns.

Nonlinear Adaptive PID Controller Desist based on an Immune Feedback Mechanism and a Gradient Descent Learning (면역 피드백 메카니즘과 경사감소학습에 기초한 비선형 적응 PID 제어기 설계)

  • 박진현;최영규
    • Proceedings of the Korean Institute of Intelligent Systems Conference
    • /
    • 2002.12a
    • /
    • pp.113-117
    • /
    • 2002
  • PID controllers, which have been widely used in industry, have a simple structure and robustness to modeling error. But it is difficult to have uniformly good control performance in system parameters variation or different velocity command. In this paper, we propose a nonlinear adaptive PR controller based on an Immune feedback mechanism and a gradient descent teaming. This algorithm has a simple structure and robustness to system parameters variation. To verify performances of the proposed nonlinear adaptive PID controller, the speed control of nonlinear DC motor Is peformed. The simulation results show that the proposed control systems are effective in tracking a command velocity under system parameters variation

Wavelet Neural Network Based Indirect Adaptive Control of Chaotic Nonlinear Systems

  • Choi, Yoon-Ho;Choi, Jong-Tae;Park, Jin-Bae
    • Journal of the Korean Institute of Intelligent Systems
    • /
    • v.14 no.1
    • /
    • pp.118-124
    • /
    • 2004
  • In this paper, we present a indirect adaptive control method using a wavelet neural network (WNN) for the control of chaotic nonlinear systems without precise mathematical models. The proposed indirect adaptive control method includes the off-line identification and on-line control procedure for chaotic nonlinear systems. In the off-line identification procedure, the WNN based identification model identifies the chaotic nonlinear system by using the serial-parallel identification structure and is trained by the gradient-descent method. And, in the on-line control procedure, a WNN controller is designed by using the off-line identification model and is trained by the error back-propagation algorithm. Finally, the effectiveness and feasibility of the proposed control method is demonstrated with applications to the chaotic nonlinear systems.

Variable Structure Controller with Time-Varying Switching Surface under the Bound of Input using Evolution Strategy (진화전략과 입력제약조건에 의한 시변스위칭면의 가변구조제어기 설계)

  • Lee, Min-Jeong;Kim, Hyeon-Sik;Choe, Yeong-Gyu;Jeon, Seong-Jeup
    • The Transactions of the Korean Institute of Electrical Engineers A
    • /
    • v.48 no.4
    • /
    • pp.402-409
    • /
    • 1999
  • Variable structure control law is well known to be a robust control algorithm and evolution strategy is used as an effective search algorithm in optimization problems. In this paper, we propose a variable structure controller with time-varying switching surface. We calculate the maximum value of seitching surface gradient that is of the 3rd order polynomial form. Evolution strategy is used to optimize the parameters of the switching surface gradient. Finally, the proposed method is applied to position tracking control for BLDC motor. Experimental results show that the proposed method is more useful than the conventional variable structure controller.

  • PDF

Mobile Robot Path Planner for Environment Exploration (효율적 환경탐사를 위한 이동로봇 경로 계획기)

  • Bae, Jung-Yun;Lee, Soo-Yong;Lee, Beom-Hee
    • The Journal of Korea Robotics Society
    • /
    • v.1 no.1
    • /
    • pp.9-16
    • /
    • 2006
  • The Mobile robots are increasingly being used to perform tasks in unknown environments. The potential of robots to undertake such tasks lies in their ability to intelligently and efficiently search in an environment. An algorithm has been developed for robots which explore the environment to measure the physical properties (dust in this paper). While the robot is moving, it measures the amount of dust and registers the value in the corresponding grid cell. The robot moves from local maximum to local minimum, then to another local maximum, and repeats. To reach the local maximum or minimum, simple gradient following is used. Robust estimation of the gradient using perturbation/correlation, which is very effective when analytical solution is not available, is described. By introducing the probability of each grid cell, and considering the probability distribution, the robot doesn't have to visit all the grid cells in the environment still providing fast and efficient sensing. The extended algorithm to coordinate multiple robots is presented with simulation results.

  • PDF

Static Analysis of a Small Scale Ducted-Fan UAV using Wind Tunnel Data

  • Choi, Youn-Han;Suk, Jin-Young;Hong, Sang-Hwee
    • International Journal of Aeronautical and Space Sciences
    • /
    • v.13 no.1
    • /
    • pp.34-42
    • /
    • 2012
  • This paper discusses the mathematical modeling of a small scale ducted-fan UAV and examines its results in comparison to the wind tunnel test. A wind tunnel test is first performed, producing a substantial amount of test data. The acquired set of wind tunnel test data is then categorized and approximated as mathematical functions. Finally, the mathematically modeled forces and moments acting on the UAV are compared with the acquired wind tunnel data. The analysis involves a gradient-based algorithm and is applied to extract trim states with respect to various flight conditions. Consequently, a numerical analysis demonstrates that there exists a reasonable flight status with respect to airspeed.

Residual Learning Based CNN for Gesture Recognition in Robot Interaction

  • Han, Hua
    • Journal of Information Processing Systems
    • /
    • v.17 no.2
    • /
    • pp.385-398
    • /
    • 2021
  • The complexity of deep learning models affects the real-time performance of gesture recognition, thereby limiting the application of gesture recognition algorithms in actual scenarios. Hence, a residual learning neural network based on a deep convolutional neural network is proposed. First, small convolution kernels are used to extract the local details of gesture images. Subsequently, a shallow residual structure is built to share weights, thereby avoiding gradient disappearance or gradient explosion as the network layer deepens; consequently, the difficulty of model optimisation is simplified. Additional convolutional neural networks are used to accelerate the refinement of deep abstract features based on the spatial importance of the gesture feature distribution. Finally, a fully connected cascade softmax classifier is used to complete the gesture recognition. Compared with the dense connection multiplexing feature information network, the proposed algorithm is optimised in feature multiplexing to avoid performance fluctuations caused by feature redundancy. Experimental results from the ISOGD gesture dataset and Gesture dataset prove that the proposed algorithm affords a fast convergence speed and high accuracy.