• Proceedings of the KIEE Conference
• /
• 2001.07d
• /
• pp.2102-2104
• /
• 2001

This paper present the strategy of self-tuning nonlinear PID control using neural network. The nonlinear PID controller consists of a conventional PID controller and a neural network compensator. The neural network is trained by back-propagation algorithm. In this paper we propose modified back-propagation algorithm to improve learning speed. The results of simulation show the usefulness of the proposed scheme.

• 제어로봇시스템학회:학술대회논문집
• /
• 1999.10a
• /
• pp.173-176
• /
• 1999

This paper shows a self tuning fuzzy inference method by the genetic algorithm in the fuzzy-sliding mode control for a Polishing robot. Using this method, the number of inference rules and the shape of membership functions are determined by the genetic algorithm. The fuzzy outputs of the consequent part are derived by the gradient descent method. Also, it is guaranteed that .the selected solution become the global optimal solution by optimizing the Akaike's information criterion expressing the quality of the inference rules. It is shown by simulations that the method of fuzzy inference by the genetic algorithm provides better learning capability than the trial and error method.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.7
• /
• pp.3171-3191
• /
• 2016

Due to growing demand on wireless data traffic, a large number of different types of base stations (BSs) have been installed. However, space-time dependent wireless data traffic densities can result in a significant number of idle BSs, which implies the waste of power resources. To deal with this problem, we propose an active state control algorithm based on semi-Markov decision process (SMDP) for a heterogeneous network. A MDP in discrete time domain is formulated from continuous domain with some approximation. Suboptimal on-line learning algorithm with a random policy is proposed to solve the problem. We explicitly include coverage constraint so that active cells can provide the same signal to noise ratio (SNR) coverage with a targeted outage rate. Simulation results verify that the proposed algorithm properly controls the active state depending on traffic densities without increasing the number of handovers excessively while providing average user perceived rate (UPR) in a more power efficient way than a conventional algorithm.

• Proceedings of the KAIS Fall Conference
• /
• 2007.05a
• /
• pp.150-153
• /
• 2007

The intent of this paper is to describe a neural network structure called multi dynamic neural network(MDNN), and examine how it can be used in developing a learning scheme for computing robot inverse kinematic transformations. The architecture and learning algorithm of the proposed dynamic neural network structure, the MDNN, are described. Computer simulations are demonstrate the effectiveness of the proposed learning using the MDNN.

• 제어로봇시스템학회:학술대회논문집
• /
• 2002.10a
• /
• pp.42.2-42
• /
• 2002

We discuss a neural network solver for the inverse optimization problem. The problem is that find functional relations between input and output data, which are include defects. Finding the relations, predictions of the defect parts are also required. The part of finding the defects in the input data is an inverse problem . We consider the meanings to solve the problem on the neural network system at first. Next, we consider the network structure of the system, the learning scheme of the network, and at last, examine the precision on the numerical calculations. In the paper, we proposed the high-precision learning method for plural three-layer neural network system that is series-connect...

• Proceedings of the KAIS Fall Conference
• /
• 2008.11a
• /
• pp.191-196
• /
• 2008

The intent of this paper is to describe a neural network structure called dynamic neural processor(DNP), and examine how it can be used in developing a learning scheme for computing robot inverse kinematic transformations. The architecture and learning algorithm of the proposed dynamic neural network structure, the DNP, are described. Computer simulations are provided to demonstrate the effectiveness of the proposed learning using the DNP.

• Proceedings of the KIEE Conference
• /
• 1998.11b
• /
• pp.533-536
• /
• 1998

The intent of this paper is to describe a neural network structure called dynamic neural processor(DNP), and examine how it can be used in developing a learning scheme for computing robot inverse kinematic transformations. The architecture and learning algorithm of the proposed dynamic neural network structure, the DNP, are described. Computer simulations are provided to demonstrate the effectiveness of the proposed learning using the DNP.

• Proceedings of the KAIS Fall Conference
• /
• 2006.11a
• /
• pp.122-128
• /
• 2006

The intent of this paper is to describe a neural network structure called multi dynamic neural network(MDNN), and examine how it can be used in developing a learning scheme for computing robot inverse kinematic transformations. The architecture and learning algorithm of the proposed dynamic neural network structure, the MDNN, are described. Computer simulations are demonstrate the effectiveness of the proposed learning using the MDNN.

• Journal of Institute of Control, Robotics and Systems
• /
• v.9 no.4
• /
• pp.284-295
• /
• 2003

The servo tracks of disk drives are constructed at the time of manufacture with the equipment of servo track writer. Because of the imperfection of servo track writer, disk vibrations and head fluctuations during servo track writing process, the constructed servo tracks might deviate from perfect circles and take eccentric shapes. The servo track writing error should be corrected because it might cause interference with adjacent tracks and irrecoverable operation error of disk drives. The servo track writing error is repeated every disk rotation and so is periodic time function. In this paper, we propose a new correction algorithm of servo track writing error based on iterative teaming approach. Our correction algorithm can learn iteratively the servo track writing error as accurately as is desired. Furthermore, our algorithm is robust to system model errors, is computationally simple, and has fast convergence rate. In order to demonstrate the generality and practical use of our work, we present the convergence analysis of our correction algorithm and some simulation results.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.41.1-41
• /
• 2001

Since the advanced control technique such as model predictive control has been introduced to industrial plant, there have been many progresses in the process control. As a way to improve the control performance, the on-line process optimizer was integrated with the advance controller. In this study, a control technique which improves the control. As the number of changes by the optimizer is increased, the control performance of the proposed algorithm is improved. Its control performance is shown via an numerical example.