• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.34.3-34
• /
• 2001

The design method of H$\infty$ filter for discrete-time uncertain linear systems with multiple state delays is investigated. The uncertain parameters are assumed to be unknown but belonging to known convex compact set of polytope type less conservative than norm bounded parameter uncertainty. The modified H$\infty$ performance measure is introduced to consider the initial states values which affect the performance of filter. The objective is to design a stable H$\infty$ filter guaranteeing asymptotic stability of filtering error dynamics and minimizing H$\infty$ norm bound. The sufficient condition for the existence of filter and the filter design method are established by LMI (linear matrix inequality) approach.

• 농업생명과학연구
• /
• 제46권6호
• /
• pp.185-195
• /
• 2012

The purpose of this paper is to analyze how strategically the bilaterally monopolistic firms, only-one-seller and only-one-buyer, behave in a situation in which each firm has uncertain information on its opponent firm's cost. Even though the two firms know that seeking integrated profit leads to the optimized profit for both firms, each firm has an incentive to opportunistically behave to increase its share of the integrated profit. These opportunistic behaviors of the firms are analyzed through a game theoretic approach especially finding Nash equilibrium mixed strategies for the strategic profiles such as true-report or not and monitoring or not. The comparative statics to the Nash equilibrium mixed strategies shows that as the profit share increases the probability of monitoring an opponent firm is decreased while the probability increases as the size of the overstated production cost increases. This study also shows that high penalty and low monitoring cost lead to high probability to tell the truth of the production cost.

• International Journal of Reliability and Applications
• /
• 제5권2호
• /
• pp.59-74
• /
• 2004

There are many uncertain parameters associated with vibration components. Their physical parameters, the machining quality of vibration components, and the applied load acting on them are all uncertain. As a result, the natural frequency and the fatigue limits are also uncertain variables. In this paper, we express these parameters of vibration components and the frequency zone of resonance through interval models; this way, the robust reliability of the vibration components is defined. The robust reliability model measures and assesses the reliability of vibration components. The robust reliability of a cantilever beam is evaluated as an example. The results show that this method is reasonable for robust reliability analysis of vibration components because it does not require a large amount of failure data, it avoids the evaluation of the probability density function, and the computation is simple.

• 대한설비공학회:학술대회논문집
• /
• 대한설비공학회 2006년도 하계학술발표대회 논문집
• /
• pp.233-238
• /
• 2006

A new district cooling system using ice slurry for the uncertain cooling load of the future is presented. The chilled water produced by the absorption chillers is used for the base cooling load. The temperature of the chilled water is reduced by mixing of ice slurry depending on increasing of the cooling load. Finally, IPF of the ice slurry is increased up to 10% at the peak load. The transporting mass flow rate is decreased down to 44.7%, and the diameter of the main pipe is decreased down to 66.7%, but the diameter of the branched pipe is designed as the same size of the chilled water.

• 한국컴퓨터정보학회논문지
• /
• 제11권3호
• /
• pp.239-248
• /
• 2006

유비쿼터스 컴퓨팅 환경에서 현재의 상황 인식 어플리케이션은 다루고 있는 상황 정보가 정확하다고 가정하지만, 실제로 센서로 입력되고 해석된 상황 정보들은 종종 모호하거나 불확실하다. 본 논문에서는 상황 정보의 모호성을 해결하기 위하여 베이지안 네트워크를 사용하고 상황 정보를 표현하기 위해 온톨로지 기반 모델을 확장한 확률 모델을 제안한다. 이 논문에서 제시한 확률 확장 온톨로지 기반 상황 인식 미들웨어는 유비쿼터스 환경에서 요구되는 다양한 상황 인식 서비스의 개발 및 운용을 효과적으로 지원 할 수 있다.

• Journal of KIEE
• /
• 제10권1호
• /
• pp.29-34
• /
• 2000

In this paper, a design method of a controller is presented for a class of nonlinear systems which have time-varying parametric uncertainly. Some features of this controller are that it can tackle 1) nonlinear parametrization(i.e. uncertain parameters enter the system in the nonlinear form) and 2) nonvanishing peturbation (i.e. uncertainty need not vanish at the origin). The class of systems considered in this paper has the triangular structure for which the well-known backstepping design can be applied. The uncertain parameter is assumed to be contained in the bounded set whose size can be arbitrarily large. Also, the uncertain system are globally uniformly bounded and converge to a compact set whose size is designable. In particular, the first state of the system can be made arbitrarily small, which can be seen by the presented simulation result.

• 한국정밀공학회지
• /
• 제14권7호
• /
• pp.73-81
• /
• 1997

An new class of robust position/force hybrid control law is proposed for uncertain robot manipulators. The uncertainty is nonlinear and (plssibly fast) time-varying. Therefore, the uncertain factors such as imper- fect modeling, friction, payload change, and external disturbance are all addressed. Based on the possible bound of the uncertainty, the controller is constructed and the stability study based on Lyapunov function is presented. To show that the proposed control laws are indeed applicable, the theoretical result is applied to a SCARA-type robot manipulator and simulation result is presented.

• 대한교통학회지
• /
• 제21권1호
• /
• pp.77-90
• /
• 2003

교통정보의 정확도는 정보에 대한 운전자의 반응을 결정하는데 중요한 요소이다. 교통상황의 정확한 정보를 제공하는 것은 현실적으로 어렵기 때문에 대부분의 교통정보는 불확실성을 내재하게 된다. 그러나 기존의 많은 관련 연구에서는 정확한 정보와 그에 대한 운전자의 신뢰를 가정하고 있다. 본 논문의 목적은 불확실한 정보가 운전자의 의사결정에 미치는 영향을 분석하는 것이다. 이를 위해 노선선택에서 불확실한 통행료 정보의 선택상황을 가정해 SP 설문조사를 실시하고 통행료 정보의 불확실정도를 정량화 하기 위해 범위 값으로 표현하였다. 특히 본 연구는 기존 확률효용이론(Random Utility Theory)이 설명하지 못하는 불확실한 정보가 노선선택에 미치는 영향을 분석하기 위하여 Prospect Theory를 적용하여 정보의 불확실성에 따른 운전자의 노선선택행태를 분석하였다. 분석결과, 운전자는 정보가 주어진 노선을 정보가 주어지지 않는 노선보다 선호하는 것으로 나타났다. 또한 운전자는 확실한 정보를 선호해 노선선택을 하는 것으로 나타났다. 그러나 정보의 불확실정도가 증가하면서(범위값이 증가하면서) 노선의 선호도가 감소하는 경향이 있는 것으로 나타났다. 한편 범위값으로 제공된 정보에 대해서는 범위값의 크기와 더불어서 중위값도 운전자의 의사결정에 영향을 미치는 것으로 나타났다. 위 결과에 Prospect Theory를 적용한 결과, 불확실한 정보에 대한 운전자의 의사결정은 선택상황을 이해하고 평가하여 변형, 재형하는 일련의 과정에 따라서 다양한 행태가 나타나는 것으로 분석되었다. 따라서, 불확실한 정보에 대한 운전자의 의사결정을 분석할 때는 운전자가 선택상황을 해석하는 방식에 따라 다양하고 융통성 있게 접근해야 한다. 본 논문은 Prospect Theory의 기본 개념과 의사결정단계에 따른 원칙만을 적용하였으나 향후 불확실한 정보에 대한 운전자의 행태를 설명할 수 있는 좀 더 상세한 분석과 더불어 다양한 분야의 이론이 적용되는 것이 필요하다고 사료된다.

• Transactions on Control, Automation and Systems Engineering
• /
• 제4권2호
• /
• pp.140-146
• /
• 2002

We consider a class of large-scale interconnected time delay systems and investigate a decentralized robust passive control problem. sufficient conditions for unforced interconnected uncertain systems with time delay to be robustly stable with extended strictly passivity is given in terms of algebraic Riccati inequality and linear matrix inequality. The decentralized robust passive control problem for norm-bounded and positive real uncertainty is shown to be converted to extended strictly positive real control problem for a modified system which contains neither time delay nor uncertainty.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1997년도 한국자동제어학술회의논문집; 한국전력공사 서울연수원; 17-18 Oct. 1997
• /
• pp.159-162
• /
• 1997

This paper describes robust stabilization of nonlinear single-input uncertain systems without matching conditions. We consider nonlinear systems with a vector of unknown constant parameters perturbed about a known value. The approach utilizes the generalized controller canonical form to lump the unmatched uncertainties recursively into the matched ones. This can be achieved via nonlinear coordinate transformations which depend not only on the states of the nonlinear system but also on the control input. Then the dynamic robust control law is derived and the stability result is also presented.