• Proceedings of the KIEE Conference
• /
• 1990.07a
• /
• pp.77-80
• /
• 1990

This paper presents a continuous-time robust direct adaptive algorithm in the presence of bounded disturbances and / or unmodeled dynamics. In the new algorithm, Narendra's adaptation law is adapted. And a term, proportional to the product of tracking error and normalizing signal, is added to the conventional control law. It is shown that the performance of the adaptive schemes is improved if a proportional adaptation tera is added to the control law. The scalar case is only discussed in the stability analysis. Computer simulation is presented to complement the theoretical result.

• International Journal of Aeronautical and Space Sciences
• /
• v.6 no.2
• /
• pp.56-63
• /
• 2005

Some methods have been presented to avoid collisions among satellites for satellite formation flying mission. The potential function method based on Lyapunov's theory is known as a powerful tool for collision avoidance in the robotic system because of its robustness and flexibility. During the last decade, a potential function has also been applied to UAV's and spacecraft operations, which consists of repulsive and attractive potential. In this study, the controller is designed using a potential function via sliding mode technique for the configuration of satellite formation flying. The strategy is based on enforcing the satellite to move along the gradient of a given potential function. The new scalar velocity function is introduced such that all satellites reach the goal points simultaneously. Simulation results show that the controller drives the satellite toward the desired point along the gradient of the potential function and is robust against external disturbances.

• Journal of Institute of Control, Robotics and Systems
• /
• v.8 no.11
• /
• pp.916-921
• /
• 2002

We presents a new method of constructing an equivalent T-S fuzzy model by using the sum of products of linearly independent scalar functions from nonlinear dynamics. This method exactly expresses nonlinear systems and automatically determines the number of rules. We design a stabilizing controller f3r ul inverted pendulum system by using the concep of parallel distributed compensation (PDC) and linear matrix inequalities (LMIs) based on the proposed T-S fuzzy modeling method. We show effectiveness of a systematically designed fuzzy controller based on the proposed T-S fuzzy modeling method through the simulation and experiment of an inverted pendulum system.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.47 no.7
• /
• pp.78-86
• /
• 2010

This paper discusses the programming model of SODA-II that is a baseband processor for software defined radio (SDR) systems. Signal processing On-Demand Architecture Ⅱ (SODA-II) is an on-chip multiprocessor architecture consisting of four processor cores and each core has both an wide SIMD datapath and a scalar datapath. This architecture is appropriate for baseband processing that is a mixture of vector computations and scalar computations. The programming model of the SODA-II is based on C library routines. Because the library routines hide the details of complex SIMD datapath control procedures, end users can easily program the SODA-II without deep understanding on its architecture. In this paper, we discuss the details of library routines and how these routines are exploited in the implementation of baseband signal processing algorithms. As application examples, we show the implementation result of W-CDMA multipath searcher and OFDM demodulator on the SODA-II.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.46 no.5
• /
• pp.1-13
• /
• 2009

Filter cache has been introduced as one solution of reducing cache power consumption. More than 50% of the power reduction results from the filter cache, whereas more than 20% of the performance is compromised. To minimize the performance degradation of the filter cache, the predictive filter cache has been proposed. In this paper, we review the previous filter cache predictors and analyze the problems of the solutions. As a result, we found main problems that cause prediction misses in previous filter cache schemes and, to resolve the problems, this paper proposes a new prediction policy. In our scheme, some reference bit entries, called MSBs, are inserted into filter cache and BTB, to adaptively control the filter cache access. In simulation parts, we use a modified SimpleScalar simulator with MiBench benchmark programs to verify the proposed filter cache. The simulation result shows in average 5% performance improvement, compared to previous ones.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.48 no.3
• /
• pp.6-12
• /
• 2011

Electric vehicles and robots are real-time distributed control systems composed of multiple drive subsystems using micro controller units. Each control subsystem should be modular, compact, power saving, interoperable and fault tolerable in order to be incorporated into the networked real-time distributed control system. Under the networked real-time distributed control the synchronization problem can be occurred to the position and orientation tracking control due to the load variance, mismatch and time delay between the multiple drive subsystems. This paper suggests two types of position synchronization control of the single link manipulators. One of them is composed of cross controller, Kalman filter and disturbance observer, and the other uses the generation of target trajectories to minimize the gradient vector of the scalar function which is composed of the sum of square errors between the reference input vector and the output vectors. The availability of the proposed control schemes is shown through the control experiments.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.142-145
• /
• 2003

Many control applications using Neural Network need a priori information about the objective system. But it is impossible to get exact information about the objective system in real world. To solve this problem, several control methods were proposed. Reinforcement learning control using neural network is one of them. Basically reinforcement learning control doesn't need a priori information of objective system. This method uses reinforcement signal from interaction of objective system and environment and observable states of objective system as input data. But many methods take too much time to apply to real-world. So we focus on faster learning to apply reinforcement learning control to real-world. Two data types are used for reinforcement learning. One is reinforcement signal data. It has only two fixed scalar values that are assigned for each success and fail state. The other is observable state data. There are infinitive states in real-world system. So the number of observable state data is also infinitive. This requires too much learning time for applying to real-world. So we try to reduce the number of observable states by classification of states with Self-Organizing Map. We also use neural dynamic programming for controller design. An inverted pendulum on the cart system is simulated. Failure signal is used for reinforcement signal. The failure signal occurs when the pendulum angle or cart position deviate from the defined control range. The control objective is to maintain the balanced pole and centered cart. And four states that is, position and velocity of cart, angle and angular velocity of pole are used for state signal. Learning controller is composed of serial connection of Self-Organizing Map and two Multi-layer Feed-Forward Neural Networks.

• Journal of the Korea Computer Graphics Society
• /
• v.12 no.1
• /
• pp.1-7
• /
• 2006

본 논문에서는 다양체 (manifold) 구조와 스칼라 변위함수 (scalar displacement function)에 기반한 새로운 변위곡면 (displaced surface)의 표현 기법을 제안한다. 변위곡면은 제어메쉬 (control mesh)의 각 정점에서 변위된 국소적 패치들 (displaced local patches)을 블렌딩 (blending)함으로써 생성된다. 제안된 변위곡면은 점 군 (point cloud)의 형태로 주어진 기하학적 모델을 근사하기 위해서 사용된다. 점 군의 데이터로터 제어메쉬가 생성되고, 점 군의 점들이 제어메쉬의 국소적 패치들 (local patches)에 사영 (projection)되어 각 패치들로 부터의 스칼라 변위함수가 구해지고, 이러한 변위함수들을 최적화 하여 높은 정밀도를 갖는 최종적인 곡면을 생성된다. 점 군의 형태로 주어진 다양한 모델에 대한 실험 결과를 통해서 제안된 근사기법의 효율성과 정밀도가 입증된다. 본 논문에서 제안된 표현기법은 다 단계 (multi-level) 변위함수를 통해 다중해상도 표현 (multi-resolution representations)과 골격기반 형상변형 (skeleton-driven deformation)등과 같은 다양한 응용들에 효율적으로 사용된다.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.6 no.3
• /
• pp.282-290
• /
• 2001

The accurate identification of the motor parameters is crucially important to achieve high dynamic performance of induction motors. In this paper, th motor parameters such as stator(rotor) resistance, stator(rotor) leakage inductance, mutual inductance, and rotor inertia are measured in off-line. Stator(rotor) resistance and stator(rotor) leakage inductance are measured based on the stationary coordinate equations of induction motors. On the other hand, mutual inductance are measured under the scalar control. Finally, the inverse rotor time constant is identified in on-line using an extended kalman filter algorithm. To demonstrate the practical significance of the results, Some experimental results are presented.

• Journal of Korean Society for Atmospheric Environment
• /
• v.12 no.5
• /
• pp.577-585
• /
• 1996

A screening study was performed in order to resolve the flow, combustion and emission characteristics of the gas furmace with co-axial diffusion flane burner. A control-valume based finite-difference method with the power-law scheme was employed for discretization. Numerical procedure for the differential equation was used by SIMPLEST to enclosute rapid converge. A k-.varepsilon. model was incorporated for the closure of turbulence. The mass fraction and mixture fraction were calculated by cinserved scalar method. An equilibrium analysis was employed to determine the concentration of radicals in the product stream and conserbation equations were them solved for N amd NO by Zelovich reaction scheme. The method was exercised in a simple one-dimensional case first, to determine the effects of air ratio, temperature and residence time on NO formation and applied to a furnace with co-axial diffusion flame burner.