• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.43.6-43
• /
• 2001

In this paper, we describe an actor-critic method as a kind of temporal difference (TD) algorithms. The value function is regarded as a current estimator, in which two value functions have different inputs: one is an actual experience; the other is a simulated experience obtained through a predictive model. Thus, the parameter´s updating for the actor and critic parts is based on actual and simulated experiences, where the critic is constructed by a radial-basis function neural network (RBFNN) and the actor is composed of a kinematic-based controller. As an example application of the present method, a tracking control problem for the position coordinates and azimuth of a nonholonomic mobile robot is considered. The effectiveness is illustrated by a simulation.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.121.2-121
• /
• 2001

Many of today´s robot are required to perform tasks which demand a high level of accuracy in end-effector positioning. Those rigid robots are very inefficient and slow because its have large and heavy links, In an attempt to solve these problems, a robots using flexible beam were created. But the single-link flexible beam is infinite-dimensional system. Many researchers have proposed controlling such a beam an approximated model consisting of a finite a number of models. In this paper, we start by deriving the analytic model for the dynamics of general single-link beam, and a controller is designed for flexible beam with integral type servo system bases of the linear matrix inequality (LM) technique. To the end, simulation results show that a designed controller guarantees affective vibration control the single-link flexible beam.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2002년도 ICCAS
• /
• pp.108.4-108
• /
• 2002

This paper aims to derive a mathematical model of the dynamics of handling tasks in robot fingers which stably grasps and manipulates a rigid object with some dexterity. Firstly, a set of differential equation describing dynamics of the manipulators and object together with geometric constraint of tight area-contacts is formulated by Lagrange's equation. Secondly, problems of controlling both the internal force and the rotation angle of the grasped object under the constraints of tight area-contacts are discussed. The effect of geometric constraints of area-contacts on motion of the overall system is analyzed and a method of computer simulation for differential-algebraic equations of overall...

• Transactions on Control, Automation and Systems Engineering
• /
• 제1권1호
• /
• pp.22-31
• /
• 1999

In this paper, a robust observer design method for nonlinear multi input multi-output(MINO) plants is presented. This method enables the extension of the time delay observer (TDO) for nonlinear SISO plants in the phase variable form to MIMO plants. The designed TDO reconstructs the states of the plant expressed in the generalized observability canonical form (GOBCF), yet requiring neither the transformation of a plant, nor the real time computation coordinates, the observer turned out to be computationally efficient and easy to design for nonlinear MIMO plants. In a simulation of a two-link manipulator with flexible joints, the control performances using TDO appeared to be similar to those using actual states and superior to those using numerical differentiation. Finally, in an experiment with a robot, it was confirmed that the TDO reconstructs the states reliability and TDO can be effectively used in a real closed-loop system.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1991년도 한국자동제어학술회의논문집(국내학술편); KOEX, Seoul; 22-24 Oct. 1991
• /
• pp.1022-1027
• /
• 1991

A new method to control a robot manipulator by neural networks is proposed. The controller is composed of both a PD controller and a neural network-based feedforward controller. MLP(multi-layer perceptron) neural network is used for the feedforward controller and trained by BP(back-propagation) learning rule. Error terms for BP learning rule are composed of the outputs of a PD controller and the acceleration errors of manipulator joints. We compare the proposed method with existing ones and contrast performances of them by simulation. Also, We discuss the real application of the proposed method in consideration of the learning time of the neural network and the time required for sensing the joint acceleration.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1988년도 한국자동제어학술회의논문집(국내학술편); 한국전력공사연수원, 서울; 21-22 Oct. 1988
• /
• pp.37-40
• /
• 1988

It is well-known that the redundancy can be exploited to avoid the singular regions of the redundant manipulator by increasing the manipulability. The method, however, requires exprecessive energy and gives rather large tracking errors since the manipulability is increased repidly so that the manipulator avoid the singular region quickly. In this paper, a new method is proposed in which the increasing speed of the manipulability is confined to a certain bound. Therefore, in the proposed method, the movement energy and the tracking errors are reduced. The computer simulation studies are performed to show the validity of the method.

• 제어로봇시스템학회논문지
• /
• 제14권4호
• /
• pp.392-398
• /
• 2008

We propose a new path planning method for autonomous mobile robots. To maximize the utility of mobile robots, the collision-free shortest path should be generated by on-line computation. In this paper, we develop an effective and practical method to generate a good solution by lower computation time. The initial path is obtained from skeleton graph by Dijkstra's algorithm. Then the path is improved by changing the graph and path dynamically. We apply the dynamic programming algorithm into the stage of improvement. Simulation results are presented to verify the performance of the proposed method.

• 대한기계학회논문집A
• /
• 제21권5호
• /
• pp.809-818
• /
• 1997

An efficient non-recursive formulation of dynamic force analysis has been developed for serially connected multibody systems. Although derivation of equations of motion is based on a recursive dynamic formulation with joint relative coordinates, in the proposed formulation, dynamic forces such as joint reaction forces and driving force are computed non-recursively for specified joints. The efficiency of the proposed formulation has been proved by the operational count and the CPU time measure, comparing with that of the conventional recursive Newton-Euler formulation. A simulation of 7-DOF RRC robot arm has been carried out to validate solutions of reaction forces by comparing with those from a commercial dynamic analysis program DADS.

• 한국기계가공학회지
• /
• 제11권5호
• /
• pp.35-41
• /
• 2012

Liquified hydrogen peroxide as a monopropellant is drawing an attention as a power generating energy source for a machine requiring simple and light weight structure. The liquified hydrogen peroxide is attracted due to its outstanding applicability because it doesn't require an oxidizer and discharge a hazardous product. For the further industrial applications, however, a feasibility study should be carried out carefully in the aspect of the specific power density. In this study, a prototype of vane motor driven by the liquified hydrogen peroxide with high density of upper 95% was developed and its performance characteristic such as a specific power density was estimated via measuring pressure and rotation speed of the vane motor. The specific power density obtained by numerical simulation using FSI analysis supported by experimental results was up to $0.02kW/kg_{f}$, which reaches at the level of the latest developed fuel cell.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2003년도 춘계학술대회 논문집
• /
• pp.988-991
• /
• 2003

In this paper, we try to develop the controller which uses the block diagrams of SIMTool and internal functions CEMTool for planning the global driving controller for high efficiency AGV. We acquire the control efficiency by controlling the motor used each part of AGV driving controller. The block diagram structures provided with SIMTool is easily designed by the controller, and the monitoring and analysis of the results is researched by simulation. We expect to control AGV. robot and various plant using Ziegler-Nichols auto-tuning method and external I/O board