• The Journal of Korea Robotics Society
• /
• v.17 no.3
• /
• pp.264-272
• /
• 2022

Human-robot cooperative tasks are increasingly required in our daily life with the development of robotics and artificial intelligence technology. Interactive reinforcement learning strategies suggest that robots learn task by receiving feedback from an experienced human trainer during a training process. However, most of the previous studies on Interactive reinforcement learning have required an extra feedback input device such as a mouse or keyboard in addition to robot itself, and the scenario where a robot can interactively learn a task with human have been also limited to virtual environment. To solve these limitations, this paper studies training strategies of robot that learn table balancing tasks interactively using deep reinforcement learning with human's facial expression feedback. In the proposed system, the robot learns a cooperative table balancing task using Deep Q-Network (DQN), which is a deep reinforcement learning technique, with human facial emotion expression feedback. As a result of the experiment, the proposed system achieved a high optimal policy convergence rate of up to 83.3% in training and successful assumption rate of up to 91.6% in testing, showing improved performance compared to the model without human facial expression feedback.

• International Journal of Control, Automation, and Systems
• /
• v.1 no.2
• /
• pp.236-242
• /
• 2003

This paper presents a cooperation strategy between a human operator and autonomous robots for an interactive robot soccer game, The interactive robot soccer game has been developed to allow humans to join into the game dynamically and reinforce entertainment characteristics. In order to make these games more interesting, a cooperation strategy between humans and autonomous robots on a team is very important. Strategies can be pre-programmed or learned by robots themselves with learning or evolving algorithms. Since the robot soccer system is hard to model and its environment changes dynamically, it is very difficult to pre-program cooperation strategies between robot agents. Q-learning - one of the most representative reinforcement learning methods - is shown to be effective for solving problems dynamically without explicit knowledge of the system. Therefore, in our research, a Q-learning based learning method has been utilized. Prior to utilizing Q-teaming, state variables describing the game situation and actions' sets of robots have been defined. After the learning process, the human operator could play the game more easily. To evaluate the usefulness of the proposed strategy, some simulations and games have been carried out.

• Journal of Institute of Control, Robotics and Systems
• /
• v.21 no.6
• /
• pp.565-570
• /
• 2015

Supervised machine learning has become popular in discovering context descriptions from sensor data. However, collecting a large amount of labeled training data in order to guarantee good performance requires a great deal of expense and time. For this reason, semi-supervised learning has recently been developed due to its superior performance despite using only a small number of labeled data. In the existing semi-supervised learning algorithms, unlabeled data are used to build a graph Laplacian in order to represent an intrinsic data geometry. In this paper, we represent the unlabeled data as the spatial-temporal dataset by considering smoothly moving objects over time and space. The developed algorithm is evaluated for position estimation of a smartphone-based robot. In comparison with other state-of-art semi-supervised learning, our algorithm performs more accurate location estimates.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.2
• /
• pp.279-284
• /
• 2009

In swarm robot systems, each robot must behaves by itself according to the its states and environments, and if necessary, must cooperates 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 using many 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 Cascade SVM is adopted in this paper.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.4
• /
• pp.19-29
• /
• 2012

According to the result of PISA and TIMSS, it was reported that interest for Math and Science was far lower compared to high achievement of Them. The purpose of this study is to investigate effects of robot based STEAM education on elementary school students' Math learning behavior and Science motivation. Robot based STEAM education integrated science, mathematics and art with a theme of 'Energy' was practiced for test group and For control group, those three subjects were taught separately in order to achieve this purpose. Curriculum of fourth grade second semester's science, mathematics and art was analysed to teach a robot based STEAM class and STEAM class Model with the theme 'Energy was designed and applied to elementary students. In science class, heat transfer experiment was conducted with robots and the result was related to drawing polygonal lines in mathematics. In art class, robot components were used to describe the heat energy in shapes and colors. The research shows that students' Math learning behavior and Science motivation were improved more with robot based STEAM education than with traditional lessons(p<.05). It proves that robot based STEAM class can be effective for improving interest in elementary Math and Science.

• The Journal of Korean Association of Computer Education
• /
• v.18 no.3
• /
• pp.15-24
• /
• 2015

Appropriate distribution and utilization of learning time in class are regarded as essential and basic conditions for successful education. Nonetheless, among studies about UCR(User Created Robot) based instruction so far is difficult to find the research related to the class. For these reasons, we attempt to analyze the ALT(Actual Learning Time) and unuse of learning time in UCR based instruction. For these purpose, we observed three students who were with third and fourth grade integrated class of elementary school and interviewed the teachers at pre-post class. The result of this study showed the following results: (1) UCR based instruction present lower ALT than traditional classes. (2) Most of the unnecessary time used in their classes tend to be used in preparing and arranging the robot module, a little is used unnecessarily because of the students' unrelated behaviors for their learning, decentralized behaviors and other external influences.

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 1996.10a
• /
• pp.6-10
• /
• 1996

This paper presents an overview of research results obtained by the authors in a series of publications. Methods are developed both for time-varying and time-invariant for linear and nonlinear. for time domain and frequency domain . and for discrete-time and continuous-time systems. Among the topics presented are: 1. Learning control based on integral control concepts applied in the repetition domain. 2. New algorithms that give improved transient response of the indirect adaptive control ideas. 4. Direct model reference learning control. 5 . Learning control based frequency domain. 6. Use of neural networks in learning control. 7. Decentralized learning controllers. These learning algorithms apply to robot control. The decentralized learning control laws are important in such applications becaused of the usual robot decentralized controller structured.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.20 no.3
• /
• pp.441-446
• /
• 2010

Each robot decides and behaviors themselves surrounding circumstances in the swarm robot system. Robots have to conduct tasks allowed through cooperation with other robots. Therefore each robot should have the ability to learn and evolve in order to adapt to a changing environment. In this paper, we proposed learning based on Q-learning algorithm and evolutionary using Harmony Search algorithm and are trying to improve the accuracy using Harmony Search Algorithm, not the Genetic Algorithm. We verify that swarm robot has improved the ability to perform the task.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.616-621
• /
• 2003

Until now, many networked robots have been connected to the Internet for the various applications. With these networked robots, very long distance teleoperation can be possible through the Internet. However, the promising area of the Internet-based teleoperation may be distance learning, because of several reasons such as the unpredictable characteristics of the Internet. In robotics class, students learn many theories about robots, but it is hard to perform the actual experiments for all students due to the rack of the real robots and safety problems. Some classes may introduce the virtual robot simulator for students to program the virtual robot and upload their program to operate the real robot through the off-line programming method. However, the students may also visit the laboratory when they want to use the real robot for testing their program. In this paper, we developed an Internet-based robot education system. The developed system was composed of two parts, the robotics class materials and the web-based Java3d robot simulator. That is, this system can provide two services for distance learning to the students through the Internet. The robotics class materials can be provided to the student as the multimedia contents on the web page. As well, the web-based robot simulator as the real experiment tool can help the students get good understanding about certain subject. So, the students can learn the required robotics theories and perform the real experiments from their web browser when they want to study themselves at any time.

• The Journal of Korea Robotics Society
• /
• v.15 no.2
• /
• pp.169-176
• /
• 2020

This paper studied the collision detection of robot manipulators for safe collaboration in human-robot interaction. Based on sensor-based collision detection, external torque is detached from subtracting robot dynamics. To detect collision using joint torque sensor data, a comparative study was conducted using data-based machine learning algorithm. Data was collected from the actual 3 degree-of-freedom (DOF) robot manipulator, and the data was labeled by threshold and handwork. Using support vector machine (SVM), decision tree and k-nearest neighbors KNN method, we derive the optimal parameters of each algorithm and compare the collision classification performance. The simulation results are analyzed for each method, and we confirmed that by an optimal collision status detection model with high prediction accuracy.