디스크립터(Descriptor) 표현은 상태공간 표현을 확장시킨 일반적인 시스템의 기술방법으로 파라미터의 물리적인 구조를 보존하고 있으며, 임펄스를 포함한 현상을 기술할 수 있다. 또한, 대규모 시스템의 모델화가 용이하며, 프로퍼(Proper)이지 않은 시스템을 기술할 수 있고 구속조건을 가진 동적시스템의 기술이 가능하는 등의 특징을 지니고 있다. 그러나, 디스크립터 시스템의 해는 미분항을 포함하고 있으므로 노이즈 및 파라미터 변동에 대해서 매우 민감하여, 관측기설계시 강인(Robust)한 관측기설계가 요구된다. 본 논문에서는 디스크립터 시스템에 대해 로바스트 관측기의 일종인 비례적분(PI) 관측기의 설계에 대해서 논한다. 먼저, 디스크립터 시스템에 대한 PI 관측기의 존재조건을 유도하였으며, 체계적인 설계법을 보였다. 또한, PI 관측기의 존재조건이 Rosenbrock의 가관측 조건하에서 간단히 보여지며, 이것은 시스템 행렬의 Rank 조건에 의해 쉽게 구해질 수 있다.

본 논문에서는 상태와 입력에 시간지연과 구조화된 불확실성이 있는 시스템을 위한 강인 H/sub .inf./ 제어기를 제안한다. 제안된 제안기는 시간지연의 크기에 관계없이 항상 불확실한 시스템을 안정화시키고, 또한 제한된 크기의 어떤 구조화된 불확실성에 대해서도 항상 폐루프 전달함수의 H/sub .inf./ 노옴의 크기를 주어진 레벨 이하로 줄인다. 제어기는 볼록 최적화 알고리즘을 이용한 LMI 문제를 풀어서 구한다.

In a lightly damped cantilever beam, most of the vibration energy is found around natural frequencies. Based on this, a phase delay control for suppressing vibration of the beam is proposed in this paper. This controller is designed to behave like a velocity feedback controller at the frequencies of modes to be controlled. Also, this controller is designed in consideration with uncontrolled modes for robust stability and improving of the sensitivity function of the control system. This phase delay control is applied to vibration suppression of a cantilever beam with a pair of a piezoelectric actuator and a piezoelectric sensor. Experimental results showed that the phase delay control functions efficiently.

It is desirable for autonomous robot systems to possess the ability to behave in a smooth and continuous fashion when interacting with an unknown environment. Although Q-learning requires a lot of memory and time to optimize a series of actions in a continuous state space, it may not be easy to apply the method to such a real environment. In this paper, for continuous state space applications, to solve problem and a triangular type Q-value model\ulcorner This sounds very ackward. What is it you want to solve about the Q-value model. Our learning method can estimate a current Q-value by its relationship with the neighboring states and has the ability to learn its actions similar to that of Q-learning. Thus, our method can enable robots to move smoothly in a real environment. To show the validity of our method, navigation comparison with Q-learning are given and visual tracking simulation results involving an 2-DOF SCARA robot are also presented.

This paper suggests an automatic calibration technique for piezoelectric low pressure transducer measuring a pressure blow 500psi. The present calibration system embedded with error correction algorithm generates it's best you don't cut parts of wards like so dynamic pressure and compensates offset voltage and pressure error. It is shown via experimental results that the instrumentation accuracy has been improved and mean time between calibrations has been shortened.

In this paper, the design and the implementation of the adaptive learning rate neural network controller for an articulate robot, which is being developed (or) has been developed in our Automatic Control Laboratory, are mainly discussed. The controller reduces software computational load via distributed processing method using multiple CPU's, and simplifies hardware structures by the time-division control with TMS32OC31 DSP chip. Proposed neural network controller with adaptive learning rate structure using expert's heuristics can improve learning speed. The proposed controller verifies its superiority by comparing response characteristics of conventional controller with those of the proposed controller that are obtained from the experiments for the 5 axis vertical articulated robot. We, also, present the generalization property of proposed controller for unlearned trajectory and the change of load through experimental data.

In order to implement the real-time operating systems for robot controller, this paper proposes a systematic method for implementing the real-time kernel under the DOS environment. So far, we designed the robot control software and its own operating system simultaneously. Though robot operating systems have simple structure, it allows the developer to have a surplus time and effort to implement complete robot systems. In addition to this, in most cases of this type, operating systems does not support multitasking function, thus, low level hardware interrupts are used for real-time execution. Subsequently, some kinds of real-time tasks are hard to implement under this environment. Nowadays, the operating systems for robot controller requires multitasking functions, intertask communication and task synchronization mechanism, and rigorous real-time responsiveness. Thus, we propose an effective and low costs real-time systems for robot controller satisfying the various real-time characteristics. The proposed real-time systems are verified through real implementation.

This paper proposes a heuristic algorithm to efficiently perform line balancing in the PCB assembly line including multiple surface mounters efficiently. Generally, the problems in line balancing are classified into two kinds. Firstly, is the determining of the minimum number of machines required for achieving the desired production rate. Secondly, is the assign of jobs to multiple machines in order to minimize the cycle time which is defined as a maximum among the working times of machines when the number of machines is fixed. In this paper, we deal with the latter. We consider a PCB assembly line, including the multiple surface mounters arranged serially as a target system. Also, the conveyor is assumed to move at a constant speed and have no buffer. Considering that the minimum number of machines required for the desired production rate is a discrete nonincreasing function which is inversely proportional to the cycle time, we propose an optimization algorithm for line balancing by using the binary search method. The algorithm is validated through computer simulation, the results of which show that their shapes coincide nearly with those of optimal line balancing efficiency graphs regardless of the number of components, the performance of surface mounters, and the structure of assembly line.

In this paper, we propose a new method of contour error modeling for cross-coupled control of CNC machines. This modeling method is based on the information that the interpolator of a CNC machine generates knot points per each sampling time in order to approximate a given curved path as a series of small straight-line segments. The merits of the proposed method are : (1) its applicability for arbitrary curved contours and (2) its ability to calculate contour errors more accurately than the other conventional methods. The proposed method is evaluated and compared with the conventional methods using the three typical curved trajectories by computer simulations. Furthermore, it is shown that the cross-coupled controller based on this proposed error model improves contouring accuracy more effectively than the other methods.

This article presents a new model which is called Periodic Square-Wave(PSW) to describe the material flow of the periodic processes involving intermediate buffer. The material flows incoming into and outgoing from the intermediate buffer are assumed to be periodic square shaped. PSW model gives the same result as that of Economic Production Quantity(EPQ) model for determining optimal lot size of single stage batch storage system. However, for batch storage serial train system, PSW model gives a different optimal solution of about 6 % reduced total cost. PSW model provides the more accurate information on inventory and production system than the classical approach by maintaining simplicity and increasing computational burden.

A real-time inspection system is developed using line scan cameras. Several improved algorithms are proposed for real-time detection of defects in this automated inspection system. The major improved algorithms include the preprocessing, the threshold decision, and the clustering algorithms. The preprocessing algorithms are for exact binarization and the threshold decision algorithm is for fast detection of defects in 1-D binary images. The clustering algorithm is also developed for fast classifying of the defects. The system is applied to PCBs(Printed Circuit Boards) inspection. The typical defects in PCBs are pits, dent, wrinkle, scratch, and black spots. The results show that most defects are detected and classified successfully.

A car driving trainer for beginners developed under PC-based environment is described in this paper. For this trainer, a hardware is implemented as a practice car, and a trainer program is designed by computer image generation method to display 3-dimensional images on a CRT monitor. The trainer program consists of 3 main parts, that is, a speed estimate part, a wheel trace calculation part and a driving image generation part. Furthermore, a map editor is also installed for taking any test drive. After comparing this driving trainer to specify it was verified that the developed car driving trainer showed has good performances, such as lower cost, higher resolution and better image display speed.

In this paper, a two-stage kinematic optimal design for a 3 degree of-freedom (DOF) excavator subsystem, which consists of boom, arm and bucket, is performed. The objective of the first stage is to find the optimal parameters of the joint-actuating mechanisms which maximize the force-torque transmission ratio between the hydraulic actuator and the rotating joint. The objective of the second stage is to find the optimal link parameters which maximize the isotropic characteristic of the excavator subsystem throughout the workspace. It is illustrated that kinematic/dynamic performances of the kinematically optimized excavator subsystem have improved compared to those of original HE280 excavator, with respect to three performance indices such as maximum load handling capacity, maximum velocity capability, and acceleration capability.