• The Journal of Korea Robotics Society
• /
• v.13 no.3
• /
• pp.151-156
• /
• 2018

Military robots are expected to play an important role in the future battlefield, and will be actively engaged in dangerous, repetitive and difficult tasks. During the robots perform the tasks a human operator controls the robots in a supervisory way. The operator recognizes battlefield situations from remote robots through an interface of the operator control center, and controls them. In the meantime, operator workload, controller interface, robot automation level, and task complexity affect robot operability. In order to assess the robot operability, we have developed ROSim (Robot Operational Simulator) incorporating these operational factors. In this paper, we introduce the results of applying ROSim experimentally to the assessment of reconnaissance robot operability in a battle field. This experimental assessment shows three resulting measurements: operational control workload, operational control capability, mission success rate, and discuss its applicability to the defense robot research and development. It is expected that ROSim can contribute to the design of an operator control center and the design analysis of a human-robot team in the defense robot research and development.

• The Journal of Korea Robotics Society
• /
• v.14 no.1
• /
• pp.40-49
• /
• 2019

Reinforcement learning has been applied to various problems in robotics. However, it was still hard to train complex robotic manipulation tasks since there is a few models which can be applicable to general tasks. Such general models require a lot of training episodes. In these reasons, deep neural networks which have shown to be good function approximators have not been actively used for robot manipulation task. Recently, some of these challenges are solved by a set of methods, such as Guided Policy Search, which guide or limit search directions while training of a deep neural network based policy model. These frameworks are already applied to a humanoid robot, PR2. However, in robotics, it is not trivial to adjust existing algorithms designed for one robot to another robot. In this paper, we present our implementation of Guided Policy Search to the robotic arms of the Baxter Research Robot. To meet the goals and needs of the project, we build on an existing implementation of Baxter Agent class for the Guided Policy Search algorithm code using the built-in Python interface. This work is expected to play an important role in popularizing robot manipulation reinforcement learning methods on cost-effective robot platforms.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.1121-1125
• /
• 2003

In this paper we deal with visual servoing that can control a robot arm with a camera using information of images only, without estimating 3D position and rotation of the robot arm. Here it is assumed that the robot arm is calibrated and the camera is uncalibrated. We use a pinhole camera model as the camera one. The essential notion can be show, that is, epipolar geometry, epipole, epipolar equation, and epipolar constrain. These play an important role in designing visual servoing. For easy understanding of the proposed method we first show a design in case of the calibrated camera. The design is constructed by 4 steps and the directional motion of the robot arm is fixed only to a constant direction. This means that an estimated epipole denotes the direction, to which the robot arm translates in 3D space, on the image plane.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.2 no.4
• /
• pp.221-226
• /
• 2007

This paper propose a control network system based on IEEE1394 for a multi body robot control. The IEEE1394 has the characteristic of high speed(400Mbps), real-time, stability and plug&play. And IEEE1394 also supports freeform daisy chaining, branching and hot plugging, which reduce cabling complexity and make a system simple. Especially, multi host and broad casting support network data sharing method which is suitable for control network for multi body robot. Through experiment, we show that the proposed control network can interface 48 joints (BLDC motors, gears, and encoders) and four 6-axis force/torque sensors with 4Khz communication bandwidth, which is adequate for a multi body robot.

• The Journal of Korea Robotics Society
• /
• v.1 no.2
• /
• pp.203-211
• /
• 2006

In this research, a comprehensive study is performed upon the design of a quadruped walking robot. In advance, the walking posture and skeletal configuration of the vertebrate are analyzed to understand quadrupedal locomotion, and the roles of limbs during walking are investigated. From these, it is known that the forelimbs just play the role of supporting their body and help vault forward, while most of the propulsive force is generated by hind limbs. In addition, with the study of the stances on walking and energy efficiency, design criteria and control method for a quadruped walking robot are derived. The proposed controller, though it is simple, provides a useful framework for controlling a quadruped walking robot. In particular, introduciton of a new rhythmic pattern generator relieves the heavy computational burden because it does not need any computation on kinematics. Finally, the proposed method is validated via dynamic simulations and implementing in a quadruped walking robot, called AiDIN(Artificial Digitigrade for Natural Environment).

• ETRI Journal
• /
• v.38 no.6
• /
• pp.1229-1239
• /
• 2016

For efficient interaction between humans and robots, robots should be able to understand the meaning and intention of human behaviors as well as recognize them. This paper proposes an interactive human intention reading method in which a robot develops its own knowledge about the human intention for an object. A robot needs to understand different human behavior structures for different objects. To this end, this paper proposes a hierarchical behavior knowledge network that consists of behavior nodes and directional edges between them. In addition, a human intention reading algorithm that incorporates reinforcement learning is proposed to interactively learn the hierarchical behavior knowledge networks based on context information and human feedback through human behaviors. The effectiveness of the proposed method is demonstrated through play-based experiments between a human and a virtual teddy bear robot with two virtual objects. Experiments with multiple participants are also conducted.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2013.10a
• /
• pp.131-134
• /
• 2013

Nuri process is a common process of childcare and education for infants to 5 years old in 2012 has been carried out. In order to operate a paint program, teachers not only easy to take advantage in the field, education robot, taking into account developmental compatibility in accordance with stable, physical stability, age psychological infant, the diversity, education robot that can be worn thinking, creativity, scientific thinking is required. In this paper, I will study the applications and Internet services to infants through your favorite smartphones. To the study of development in the smartphone educational insect robot application with the educational environment that infants are environmentally friendly. Utilizing the game smartphone, Nari proceed to step 15, place the questions and answers for enhancing the education effect on the way, to enhance the creative effect of the robot and education. In this paper, I will continue to contribute to the development of content for the education of taste creative.

• Journal of the Semiconductor & Display Technology
• /
• v.22 no.4
• /
• pp.113-118
• /
• 2023

Maintenance of semiconductor equipment processes is crucial for the continuous growth of the semiconductor market. The process must always be upheld in optimal condition to ensure a smooth supply of numerous parts. Additionally, it is imperative to monitor the status of the robots that play a central role in the process. Just as many senses of organs judge a person's body condition, robots also have numerous sensors that play a role, and like human joints, they can detect the condition first in the joints, which are the driving parts of the robot. Therefore, a normal state test bed and an abnormal state test bed using an aging reducer were constructed by simulating the joint, which is the driving part of the robot. Various sensors such as vibration, torque, encoder, and temperature were attached to accurately diagnose the robot's failure, and the test bed was built with an integrated system to collect and control data simultaneously in real-time. After configuring the user screen and building a database based on the collected data, the characteristic values of normal and abnormal data were analyzed, and machine learning was performed using the KNN (K-Nearest Neighbors) machine learning algorithm. This approach yielded an impressive 94% accuracy in failure diagnosis, underscoring the reliability of both the test bed and the data it produced.

• 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.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.835-838
• /
• 2003

Computer vision and natural-language dialogue play an important role in friendly human-machine interfaces for service robots. In this paper we describe an integrated face detection and face recognition system for a welfare robot, which has also been combined with the robot's speech interface. Our approach to face detection is to combine neural network (NN) and genetic algorithm (GA): ANN serves as a face filter while GA is used to search the image efficiently. When the face is detected, embedded Hidden Markov Model (EMM) is used to determine its identity. A real-time system has been created by combining the face detection and recognition techniques. When motivated by the speaker's voice commands, it takes an image from the camera, finds the face inside the image and recognizes it. Experiments on an indoor environment with complex backgrounds showed that a recognition rate of more than 88% can be achieved.