• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.85.4-85
• /
• 2001

A learning controller is presented for repetitive walking motion of biped robot. The learning control scheme learns the approximate inverse dynamics input of biped walking robot and uses the learned input pattern to generate an input profile of different walking motion from that learnt. In the learning controller, the PID feedback controller takes part in stabilizing the transient response of robot dynamics while the feedforward learning controller plays a role in computing the desired actuator torques for feedforward nonlinear dynamics compensation in steady state. It is shown that all the error signals in the learning control system are bounded and the robot motion trajectory converges to the desired one asymptotically. The proposed learning control scheme is ...

• The Transactions of the Korean Institute of Electrical Engineers P
• /
• v.55 no.2
• /
• pp.83-88
• /
• 2006

This paper proposes a method that minimizes the consumed energy by searching the optimal locations of the mass centers of the biped robot's links using Genetic Algorithm. This paper presents a learning controller for repetitive gait control of the biped robot. The learning control scheme consists of a feedforward learning nile and linear feedback control input for stabilization of learning system. The feasibility of learning control to the biped robotic motion is shown via computer simulation and experimental results with 24 DOF biped walking robot.

• Proceedings of the KIEE Conference
• /
• 2000.07d
• /
• pp.2834-2836
• /
• 2000

This paper introduces the behavior-based learning controller for mobile robot using topological map. When the mobile robot navigates to the goal position, it utilizes given information of topological map and its location. Under navigating in unknown environment, the robot classifies its situation using ultrasonic sensor data, and calculates each motor schema multiplied by respective gain for all behaviors, and then takes an action according to the vector sum of all the motor schemas. After an action, the information of the robot's location in given topological map is incorporated to the learning module to adapt the weights of the neural network for gain learning. As a result of simulation, the robot navigates to the goal position successfully after iterative gain learning with topological information.

• Journal of Fisheries and Marine Sciences Education
• /
• v.27 no.2
• /
• pp.430-440
• /
• 2015

The purpose of this study was to analyze student's metacognition interaction based on a robot lesson. For this research as an analytical metacognition interaction tool was utilized. The results of this study revealed that, first, elementary school students had more metacognition interaction in middle learning levels but middle school students had more in the low learning level. Second, in the low learning level, middle school students revised the initiated goal strategy of the robot lesson. Third, in all learning levels, students showed much diagnosis and assesment metacognition interaction in the robot lesson. According to this study's results, the robot lesson has a positive effect in facilitating diagnosis meta cognition for processing of task performance. These results could provide effective cues and information on how to improve the robot lesson.

• Journal of the Semiconductor & Display Technology
• /
• v.19 no.1
• /
• pp.1-5
• /
• 2020

In this paper, a predictive maintenance of the robot trouble using the machine learning method, so called MT(Mahalanobis Taguchi), was studied. Especially, 'MD(Mahalanobis Distance)' was used to compare the robot arm motion difference between before the maintenance(bearing change) and after the maintenance. 6-axies vibration sensor was used to detect the vibration sensing during the motion of the robot arm. The results of the comparison, MD value of the arm motions of the after the maintenance(bearing change) was much lower and stable compared to MD value of the arm motions of the before the maintenance. MD value well distinguished the fine difference of the arm vibration of the robot. The superior performance of the MT method applied to the prediction of the robot trouble was verified by this experiments.

• 제어로봇시스템학회:학술대회논문집
• /
• 1996.10a
• /
• pp.197-200
• /
• 1996

This paper considers feedback error learning neural networks for robot manipulator control. Feedback error learning proposed by Kawato [2,3,5] is a useful learning control scheme, if nonlinear subsystems (or basis functions) consisting of the robot dynamic equation are known exactly. However, in practice, unmodeled uncertainties and disturbances deteriorate the control performance. Hence, we presents a robust feedback error learning scheme which add robustifying control signal to overcome such effects. After the learning rule is derived, the stability is analyzed using Lyapunov method.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.18 no.5
• /
• pp.712-717
• /
• 2008

In swarm robot systems, each robot must act by itself according to the its states and environments, and if necessary, must cooperate with other robots in order to carry out a given task. Therefore it is essential that each robot has both learning and evolution ability to adapt the dynamic environments. In this paper, reinforcement learning method with SVM based on structural risk minimization and distributed genetic algorithms is proposed for behavior learning and evolution of collective autonomous mobile robots. By distributed genetic algorithm exchanging the chromosome acquired under different environments by communication each robot can improve its behavior ability. Specially, in order to improve the performance of evolution, selective crossover using the characteristic of reinforcement learning that basis of SVM is adopted in this paper.

• Journal of the Korea Society of Computer and Information
• /
• v.24 no.1
• /
• pp.247-255
• /
• 2019

In this paper, we propose the effect of project-based robot education program on the interest toward robots and the computational thinking of elementary school students. Software education is being actively carried out around the world in order to cultivate software talents in accordance with the 4th industrial revolution. As a result, the importance of robots in education has increased, and education using robots has been actively introduced. However, the activities of simply assembling and repeating robots in schools were not effective in enhancing elementary school students' interest toward robots and computational thinking. Therefore, it is necessary to overcome traditional teaching-learning methods and to develop robot education. So, in this study, the robot education program that introduces project-based learning was developed for improvement of interest toward robots and computational thinking of elementary school students. In order to verify the developed education program, 114 elementary six grade students were selected as research subjects and the traditional teaching-learning method and project-based learning were applied to the experimental and control group. As a result, project-based learning was more effective for elementary school students' interest toward robot than traditional teaching-learning method. In computing thinking, the experimental group showed a significant improvement, but there was no statistically significant difference in the post-test.

• The Journal of Korea Robotics Society
• /
• v.14 no.2
• /
• pp.81-86
• /
• 2019

Robots are used in various industrial sites, but traditional methods of operating a robot are limited at some kind of tasks. In order for a robot to accomplish a task, it is needed to find and solve accurate formula between a robot and environment and that is complicated work. Accordingly, reinforcement learning of robots is actively studied to overcome this difficulties. This study describes the process and results of learning and solving which applied reinforcement learning. The mission that the robot is going to learn is bottle flipping. Bottle flipping is an activity that involves throwing a plastic bottle in an attempt to land it upright on its bottom. Complexity of movement of liquid in the bottle when it thrown in the air, makes this task difficult to solve in traditional ways. Reinforcement learning process makes it easier. After 3-DOF robotic arm being instructed how to throwing the bottle, the robot find the better motion that make successful with the task. Two reward functions are designed and compared the result of learning. Finite difference method is used to obtain policy gradient. This paper focuses on the process of designing an efficient reward function to improve bottle flipping motion.

• 제어로봇시스템학회:학술대회논문집
• /
• 1997.10a
• /
• pp.552-555
• /
• 1997

Recently, increasing attention has been paid to the application of learning control method to robot manipulator control. Because the learning control method does not require an exact dynamic model, it is flexible and easy to implement. In this paper, we implement a learning control scheme which consists of a unique feedforward learning controller and a linear feedback controller. The learning control method does not require acceleration terms that are sensitive to noise and has the capability of rejecting unknown disturbances and adapting itself to time-varying system parameters. The feasibility of the learning control scheme is soon by implementing the control scheme to a commercial robot manipulator and the performance of which is also compared with the conventional linear PID control method.