• Proceedings of the KIEE Conference
• /
• 2005.07d
• /
• pp.2815-2817
• /
• 2005

This paper presents an experiment study on the control of a rotary-type inverted pendulum based on the Takagi-Sugeno (T-S) fuzzy model approach. A sufficient condition for stability of the T-S fuzzy control system is given via linear matrix inequalities (LMIs). State-feedback controllers for sub-systems are designed from the sufficient condition via change of variables which is one of the popular LMI techniques. Experimental results on a rotary-type inverted pendulum control show the feasibility of the T-S fuzzy model-based control method.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2018.10a
• /
• pp.705-707
• /
• 2018

Rotary Inverted Pendulum 은 제어분야에서 비선형 제어 시스템을 설명하기 위해 자주 사용되어왔다. 본 논문은 강화학습 에이전트의 환경으로써 Rotary Inverted Pendulum 을 도입하였다. 이를 통해서 강화학습이 실제 세계에서의 복합적인 문제를 해결할 수 있음을 보인다. 강화학습 에이전트의 가상 환경과 실제 환경을 맵핑시키기 위해서 Ethernet 연결 위에 MQTT 프로토콜을 사용하였으며 이를 통해서 경량화된 IoT 분야에서의 강화학습의 활용도를 조명한다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.18 no.2
• /
• pp.236-242
• /
• 2008

In this paper, we propose an optimized fuzzy controller based on Hierarchical Fair Competition-based Genetic Algorithms (HFCGA) for rotary inverted pendulum system. We adopt fuzzy controller to control the rotary inverted pendulum and the fuzzy rules of the fuzzy controller are designed based on the design methodology of Linear Quadratic Regulator (LQR) controller. Simple Genetic Algorithms (SGAs) is well known as optimization algorithms supporting search of a global character. There is a long list of successful usages of GAs reported in different application domains. It should be stressed, however, that GAs could still get trapped in a sub-optimal regions of the search space due to premature convergence. Accordingly the parallel genetic algorithm was developed to eliminate an effect of premature convergence. In particular, as one of diverse types of the PGA, HFCGA has emerged as an effective optimization mechanism for dealing with very large search space. We use HFCGA to optimize the parameter of the fuzzy controller. A comparative analysis between the simulation and the practical experiment demonstrates that the proposed HFCGA based fuzzy controller leads to superb performance in comparison with the conventional LQR controller as well as SGAs based fuzzy controller.

• Proceedings of the Korean Institute of IIIuminating and Electrical Installation Engineers Conference
• /
• 2004.05a
• /
• pp.555-558
• /
• 2004

Due to the asymptotic property, deadbeat control can hardly applied to the continuous time system control. But some delay element method can deal such a problem. Except delay element method, well-known digital deadbeat control can br used with the aid of som smoothing elements. In this paper, and order smoothing element is used for the smoothing of the digital deadbeat controller. And this element is argumented to the plant, and so control problem is to control digitally the argumented system. We simulated this control system using Matlab language and finally apply this algorithm to the rotary inverted pendulum system.

• Proceedings of the KIEE Conference
• /
• 1998.07b
• /
• pp.681-683
• /
• 1998

Fuzzy Inference System is to trans late and be concrete with human expert in to mathematical equation. It is easy to be applied for Nonlinear System and the know ledge can be applied at that. With using the rule according to the Knowledge, when it is realized simulations must be required repeatedly and small vibration is generated in steady state, too. In this paper, we applied the system to the methodology of optimization with self-learn ing by us ing ANFIS(Adaptive Network-based Fuzzy Inference System) which makes use of back-propagation and least square method at a first order Sugeno Fuzzy System. In order to show the effect of Algorithm, we demonstrated it by us ing Rotary Inverted Pendulum.

• Proceedings of the KIEE Conference
• /
• 2001.07d
• /
• pp.2240-2242
• /
• 2001

A rotary double inverted pendulum, the nonlinear system has a regulation problem. In this paper, we linearize the nonlinear system at the upright equilibrium position. The linearized system can be expressed in state space. To maintain the upright position, we design a feedback controller using LQR(Linear Quadratic Regulator) algorithm. Then we simulate the system with third-order Adams Bashforth Moulton Method. The simulated result shows that the applied algorithm is effective for the regulation problem.

• Proceedings of the KIEE Conference
• /
• 2002.07d
• /
• pp.2196-2198
• /
• 2002

Fuzzy controller design consists of intuition, and any other information about how to control system, into a set of rules. These rules can then be applied to the system. It is very important to decide parameters of IF-THEN rules. Because Fuzzy controller can make more adequate force to the plant by means of parameter optimization, which is accomplished by learning procedure. In this paper, we apply adaptive fuzzy controller designed to the Rotary Inverted Pendulum.

• Journal of Institute of Control, Robotics and Systems
• /
• v.7 no.11
• /
• pp.918-926
• /
• 2001

In this paper, intelligent control method using neural network as a nonlinear controller is presented. Simulation studies for three link rotary robot are performed. Neural network controller is implemented on DSP board in PC to make real time computing possible. On-line training algorithms for neural network control are proposed. As a test-bed, a large x-y table was build and interface with PC has been implemented. Experiments such as inverted pendulum control and large x-y table position control are performed. The results for different PD controller gains with neural network show excellent position tracking for circular trajectory compared with those for PD controller only. Neural control scheme also works better for controlling inverted pendulum on x-y table.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2007.04a
• /
• pp.349-352
• /
• 2007

본 논문은 계층적 공정 경쟁 유전자 알고리즘(Hierarchical Fair Competition-based Genetic Algorithms : HFCGAs)을 이용한 회전형 역 진자 시스템의 최적 2-자유도 제어기 설계를 제안한다. 비선형이며 불안정한 운동을 하는 회전형 역 진자 시스템은 회전지렛대(rotating arm) 위의 폴(pole)이 적당한 제어력이 없는 상황에서 중력에 의해 어느 한 쪽 방향으로 넘어지려고 할 때, 외부에서 회전지렛대에 힘을 가하여 회전지렛대의 특정위치와 폴의 각도를 유지시키는 시스템으로, 본 논문에서는 두 개의 피드백 제어루프를 갖는 2-자유도 제어기를 구성한다. 각각의 제어기는 PD 제어기로 구성하고, 조기수렴 문제를 내재하고 있는 기존의 유전자 알고리즘의 해결방안중 하나인 HFCGAs를 이용하여 최적의 제어기 파라미터들을 구한다. 마지막으로 실제 공정에 적용하여 설계된 제어기의 성능을 평가한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.05a
• /
• pp.493-495
• /
• 2019

제어 분야의 가장 기초적인 시스템인 Rotary Inverted Pendulum 을 제어하기 위하여, 본 논문에서는 강화학습에서 Deep Q-Network 과 함께 대표적인 알고리즘으로 알려진 Asynchronous Advantage Actor-Critic 을 활용하여 다중 디바이스 제어를 설계한다. Deep Q-Network 알고리즘을 활용한 기존 연구와 동일한 방식으로 실 세계의 물리 에이전트와 가상 환경을 맵핑시키며, 스위치를 통하여 로컬 에이전트와 글로벌 네트워크 간 통신을 구성한다. 본 논문에서는 분산 Asynchronous Advantage Actor-Critic 을 이용함으로써 실 세계의 다중 에이전트 제어를 위한 강화 학습의 활용 가능성을 조명한다.