• 전자공학회논문지B
• /
• 제32B권9호
• /
• pp.1137-1153
• /
• 1995

It is not yet possible to solve the optimal number of neurons in hidden layer at neural networks. However, it has been proposed and proved by experiments that there is a limit in increasing the number of neuron in hidden layer, because too much incrememt will cause instability,local minima and large error. This paper proposes a module neural controller with pattern recognition ability to solve the above trade-off problems and to obtain fast learning convergence speed. The proposed neural controller is composed of several module having Multi-layer Perrceptron(MLP). Each module have the less neurons in hidden layer, because it learns only input patterns having a similar learning directions. Experiments with six joint robot manipulator have shown the effectiveness and the feasibility of the proposed the parallel module neural controller with pattern recognition perceptron.

• 한국정밀공학회지
• /
• 제21권11호
• /
• pp.130-139
• /
• 2004

The Casing Oscillator is a bore file Equipment for the all-casing process. All-casing process is a method of foundation work in construction yard to oscillate steel Casing in the ground. The existing Casing Oscillator has some problem like not boring horizontally with disturbance and not driving Casing othor angle except horizon. To solve problem, the new structure Casing Oscillator is presented and studied. The performance of Casing Oscillator is improved by kinematics analysis. The Casing Oscillator is similar to the parallel manipulator in structure. So we obtain Inverse kinematics solution of Casing Oscillator easily. But it is difficult to solve forward kinematics of Casing Oscillator. T his paper presents a novel pose description corresponding to the structure characteristics of parallel manipulators. Through analysis on geometry theory, we obtain a new method of the closed-form solution to the forward kinematics using Kinematic Inversion. The closed-form solution contains two different meanings -analytical and real-time. So we reach the goal of practical application and control. Closed-form forward kinematics solution is verified by an inverse kinematics analysis. It shows that the method has a practical value for real -time control and inverse kinematics servo control.

• 한국해양공학회:학술대회논문집
• /
• 한국해양공학회 2000년도 추계학술대회 논문집
• /
• pp.90-95
• /
• 2000

This paper presents the vergence control of a parallel stereo camera and its application to underwater stereo camera to enhance the working efficiency of underwater vehicles that equips with manipulators in seabed operation. The stereo camera consists of two parallel lenses mounted on a lateral moving base and two CCD cameras mounted on a longitudinal moving base, which is embedded in a small pressure canister for underwater application. Because the lateral shift is related to the backward shift with a nonlinear relation, only one control input is needed to control the vergence and focus of the camera with a special driving device. We can get a clear stereo vision with the camera for all the range of objects in air and in water, especially in short range objects. The control system of the camera is so simple that we are able to realize a small stereo camera system and to apply it to a stereo vision system for underwater vehicles. This paper also shows how to acquire the distance information of an underwater object with this stereo camera. Whenever we focus on an underwater object with the camera, we can obtain the three-dimensional images and the distance information in real-time.

• 한국해양공학회지
• /
• 제15권1호
• /
• pp.79-84
• /
• 2001

This paper present the control of the image disparity of a parallel stereo camera and its application to an underwater stereo camera to enhance the working efficiency of underwater vehicles that are equiped with manipulators in seabed operation. The stereo camera consists of two parallel lenses mounted on a lateral moving base and two CCD cameras mounted on a longitudinal moving base, which is embedded in a small pressure canister for underwater application. Because the lateral shift is related to the backward shift with a nonlinear relation, only one control input is needed to control the vergence and focus of the camera with a special driving device. We can get clear stereo vision with the camera for all the range of objects in air and in water, especially in short range object. The control system of the camera is so simple that we are able to realize a small stereo camera system and apply it to a stereo vision system for underwater vehicles. This paper also shows how to acquire the distance information of an underwater object with this stereo camera. Whenever we focus on an underwater object with the camera, we can obtain three-dimensional images and distance information in real-time.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2004년도 ICCAS
• /
• pp.825-830
• /
• 2004

Kinematic calibration of Gough-Stewart platform using a new measurement device is presented in this paper. The device simultaneously measures components of position and orientation using commercially available gadgets. Additional kinematic parameters are defined to model the sources of inaccuracies for the proposed measurement device. Computer simulations reveal that all kinematic parameters of the Gough-Stewart platform and the additional kinematic parameters of the measurement device can be identified with the partial pose measurements of the device. Results also show that identification is robust for the initial errors and the noise in measurements. The device also facilitates the automation of easurement procedure.

• 대한전기학회논문지
• /
• 제39권12호
• /
• pp.1306-1316
• /
• 1990

A controller using a multilayer neural network is proposed to the dynamic control of a PUMA 560 robot arm. This controller is developed based on an error back-propagation (BP) neural network. Since the neural network can model an arbitrary nonlinear mapping, it is used as a commanded feedforward torque generator. A Proportional Derivative (PD) feedback controller is used in parallel with the feedforward neural network to train the system. The neural network was trained by the current state of the manipulator as well as the PD feedback error torque. No a priori knowledge on system dynamics is needed and this information is rather implicitly stored in the interconnection weights of the neural network. In another experiment, the neural network was trained with the current, past and future positions only without any use of velocity sensors. Form this thim window of position values, BP network implicitly filters out the velocity and acceleration components for each joint. Computer simulation demonstrates such powerful characteristics of the neurocontroller as adaptation to changing environments, robustness to sensor noise, and continuous performance improvement with self-learning.

• International Journal of Precision Engineering and Manufacturing
• /
• 제2권1호
• /
• pp.76-80
• /
• 2001

A binary parallel robot manipulator uses actuators that have only two stable states being built by stacking variable geometry trusses on top of each other in a long serial chain. Discrete characteristics of the binary manipulator make it impossible to analyze an inverse kinematic problem in conventional ways. We therefore introduce new definitions of workspace and inverse kinematic solution, and the apply a genetic algorithm to the newly defied inverse kinematic problem. Numerical examples show that our genetic algorithm is very efficient to solve the inverse kinematic problem of binary robot manipulators.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1991년도 한국자동제어학술회의논문집(국내학술편); KOEX, Seoul; 22-24 Oct. 1991
• /
• pp.147-151
• /
• 1991

In this paper a time-optimal path planning scheme for the multiple robot manipulators will be proposed by using hopfield neural network. The time-optimal path planning, which can allow multiple robot system to perform the demanded tasks with a minimum execution time and collision avoidance, may be of consequence to improve the productivity. But most of the methods proposed till now suffers from a significant computational burden and thus limits the on-line application. One way to avoid such a difficulty is to rearrange the problem as MTSP(Multiple Travelling Salesmen Problem) and then apply the Hopfield network technique, which can allow the parallel computation, to the minimum time problem. This paper proposes an approach for solving the time-optimal path planning of the multiple robots by using Hopfield neural network. The effectiveness of the proposed method is demonstrated by computer simulation.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1992년도 한국자동제어학술회의논문집(국내학술편); KOEX, Seoul; 19-21 Oct. 1992
• /
• pp.632-637
• /
• 1992

Recently many dynamics control algorithms using robot dynamic equation have been proposed. One of them, Kawato's feedback error learning scheme requires neither an accurate model nor parameter estimation and makes the robot motion closer to the desired trajectory by repeating operation. In this paper, the feedback error learning algorithm is implemented to control a robot system, 5 DOF revolute type movemaster. For this purpose, an actuator dynamic model is constructed considering equivalent robot dynamics model with respect to actuator as well as friction model. The command input acquired from the actuator dynamic model is the sum of products of unknown parameters and known functions. To compute the control algorithm, a parallel processing computer, transputer, is used and real-time computing is achieved. The experiment is done for the three major link of movemaster and its result is presented.

• 제어로봇시스템학회논문지
• /
• 제12권9호
• /
• pp.918-925
• /
• 2006

Medical robotics and computer aided surgery in general, and robotic telesurgery in particular, are promising applications of robotics. In this paper, we shows a novel single-master (PHANTOM based single-master multi-slave telerobotic system) multi-slave system using two parallel mechanism micromanipulators as a slave device. After a general introduction to the systems structure and configuration of telerobotic system, a manipulation control strategy to build the system that human and both manipulators perform the cooperative manipulation, is introduced, followed by its kinematic analysis, mapping method, and experimental results.