• 한국지능시스템학회논문지
• /
• 제16권2호
• /
• pp.131-137
• /
• 2006

군집 로봇시스템에서 개개의 로봇은 스스로 주위의 환경과 자신의 상태를 스스로 판단하여 행동하고, 필요에 따라서는 다른 로봇과 협조를 통하여 어떤 주어진 일을 수행할 수 있어야 한다. 따라서 개개의 로봇은 동적으로 변화하는 환경에 잘 적응할 수 있는 학습과 진화능력을 갖는 것이 필수적이다. 이를 위하여 본 논문에서는 지연된 보상능력이 있는 강화학습과 분산유전알고리즘을 이용한 새로운 자율이동로봇의 행동학습 및 진화방법을 제안한다. 지연 보상능력이 있는 강화학습은 로봇이 취한 행동에 대하여 즉각적인 보상을 가할 수 없는 경우에도 학습이 가능한 방법이다. 또한 개개의 로봇이 통신을 통하여 염색체를 교환하는 분산유전알고리즘은 각기 다른 환경에서 학습한 우수한 염색체로부터 자신의 능력을 향상시킨다. 특히 본 논문에서는 진화의 성능을 향상시키기 위하여 강화학습의 특성을 이용한 선택 교배방법을 채택하였다. 제안된 방법은 협조탐색 문제에 적용하여 컴퓨터 시뮬레이션을 통하여 그 유효성을 검증한다.

• Journal of Semiconductor Engineering
• /
• 제2권2호
• /
• pp.130-135
• /
• 2021

Recently, deep neural networks (DNNs) are actively used for action control so that an autonomous system, such as the robot, can perform human-like behaviors and operations. Unlike recognition tasks, the real-time operation is essential in action control, and it is too slow to use remote learning on a server communicating through a network. New learning techniques, such as reinforcement learning (RL), are needed to determine and select the correct robot behavior locally. In this paper, we propose an energy-efficient DNN processor with a LUT-based processing engine and near-zero skipper. A CNN-based facial emotion recognition and an RNN-based emotional dialogue generation model is integrated for natural HRI system and tested with the proposed processor. It supports 1b to 16b variable weight bit precision with and 57.6% and 28.5% lower energy consumption than conventional MAC arithmetic units for 1b and 16b weight precision. Also, the near-zero skipper reduces 36% of MAC operation and consumes 28% lower energy consumption for facial emotion recognition tasks. Implemented in 65nm CMOS process, the proposed processor occupies 1784×1784 um2 areas and dissipates 0.28 mW and 34.4 mW at 1fps and 30fps facial emotion recognition tasks.

• 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.

• 한국경영과학회:학술대회논문집
• /
• 대한산업공학회/한국경영과학회 1995년도 춘계공동학술대회논문집; 전남대학교; 28-29 Apr. 1995
• /
• pp.869-879
• /
• 1995

The new field of learning control develops controllers that learn to improve their performance at executing a given task, based on experience performing this task. The simplest forms of learning control are based on the same concept as integral control, but operating in the domain of the repetitions of the task. In the previous paper, I had studied the use of such controllers in a decentralized system, such as a robot with the controller for each link acting independently. The basic result of the paper is to show that stability of the learning controllers for all subsystems when the coupling between subsystems is turned off, assures stability of the decentralized learning in the coupled system, provided that the sample time in the digital learning controller is sufficiently short. In this paper, we present two examples. The first illustrates the effect of coupling between subsystems in the system dynamics, and the second studies the application of decentralized learning control to robot problems. The latter example illustrates the application of decentralized learning control to nonlinear systems, and also studies the effect of the coupling between subsystems introduced in the input matrix by the discretization of the system equations. The conclusion is that for sufficiently small learning gain, and sufficiently small sample time, the simple learning control law based on integral control applied to each robot axis will produce zero tracking error in spite o the dynamic coupling in the robot equations. Of course, the results of this paper have much more general application than just to the robotics tracking problem. Convergence in decentralized systems is seen to depend only on the input and output matrices, provided the sample time is suffiently small.

• 한국전자통신학회논문지
• /
• 제16권1호
• /
• pp.95-100
• /
• 2021

본 논문에서는 수중 건설 로봇을 제어하기 위한 모델 기반 메타 강화 학습 방법을 제안한다. 모델 기반 메타 강화 학습은 실제 응용 프로그램의 최근 경험을 사용하여 모델을 빠르게 업데이트한다. 다음으로, 대상 위치에 도달하기 위해 매니퓰레이터의 제어 입력을 계산하는 모델 예측 제어로 모델을 전송한다. MuJoCo 및 Gazebo를 사용하여 모델 기반 메타 강화 학습을 위한 시뮬레이션 환경을 구축하였으며 수중 건설 로봇의 실제 제어 환경에서의 모델 불확실성을 포함하여 제안한 방법을 검증하였다.

• 한국정밀공학회지
• /
• 제10권3호
• /
• pp.175-181
• /
• 1993

In this paper, a practical method to generate task strategies applicable to chamferless and high-precision assembly, is proposed. The difficulties in devising reliable assembly strategies result from various forms of uncertainty such as imperfect knowledge on the parts being assembled and functional limitations of the assembly devices. In approach to cope with these problems, the robot is provided with the capability of learning the corrective motion in response to the force signal trrough iterative task execution. The strategy is realized by adopting a learning algorithm and represented in a binary tree type database. To verify the effectiveness of the proposed algorithm, a series of simulations and experiments are carried out under assimilated real production environments. The results show that the sensory signal-to-robot action mapping can be acquired effectively and, consequently, the chamferless assembly can be performed successfully.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2003년도 춘계학술대회 논문집
• /
• pp.1105-1108
• /
• 2003

Adaptation in dynamic environments gains a significant advantage by combining evolution and learning. We propose an on-line, realtime evolutionary learning mechanism to determine the structure and the synaptic weights of a neural network controller for mobile robot navigations. We support our method, based on (1+1) evolutionary strategy, which produces changes during the lifetime of an individual to increase the adaptability of the individual itself, with a set of experiments on evolutionary neural controller for physical robots behaviors.

• 로봇학회논문지
• /
• 제16권3호
• /
• pp.270-275
• /
• 2021

Most of the initial forms of cooperative robots were intended to repeat simple tasks in a given space. So, they showed no significant difference from industrial robots. However, research for improving worker's productivity and supplementing human's limited working hours is expanding. Also, there have been active attempts to use it as a service robot by applying AI technology. In line with these social changes, we produced a mobile manipulator that can improve the worker's efficiency and completely replace one person. First, we combined cooperative robot with mobile robot. Second, we applied speech recognition technology and deep learning based object detection. Finally, we integrated all the systems by ROS (robot operating system). This system can communicate with workers by voice and drive autonomously and perform the Pick & Place task.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제54권5호
• /
• pp.300-306
• /
• 2005

In this paper we present a guideline to evaluate how easy the use of personal robots is and how good their learning abilities are, based on the analysis of their built-in commands, user interfaces, and intelligences. Recently, we are living with robots that can be able to do lots of roles; cleaning, security, pets and education in real life. They can be classified as home robots, guide robots, service robots, robot pets, and so on. There we, however, no standards to evaluate their abilities, so it is not easy to select an appropriate robot when a user wants to buy it. Thus, we present, as a guideline that can be a standard for the evaluation of the personal robots, the standards by means of analyzing existing personal robots and results of the recent research works. We will, also, demonstrate how to apply the evaluation method to the cleaning robot as an example.

• 컴퓨터교육학회논문지
• /
• 제22권1호
• /
• pp.79-86
• /
• 2019

본 연구는 모듈형 로봇 프로그래밍 교육이 중학교 정보교과 학습동기에 미치는 영향에 대하여 알아보고자 한다. 이를 위하여 실험집단 25명, 통제집단 25명을 대상으로 실험을 하였으며 연구도구로 모듈형 로봇 프로그래밍 교육, 학습동기 검사지를 사용하였다. 학습동기 검사지의 결과를 독립표본 t-검정과 대응표본 t-검정으로 자료를 처리한 결과, 실험집단은 통제집단 보다 학습동기가 9.36점 높았고, 실험집단 내에서도 사후검사가 사전검사보다 15.44점 높았다. 특히 정보교과 학습동기의 모든 하위요소 향상에 유의미한 영향을 주며, 그 중 주의집중, 관련성, 만족감 요소 향상에 큰 영향을 주었다. 결론적으로 모듈형 로봇 프로그래밍 교육이 학생들의 정보교과 학습동기 향상에 긍정적인 영향을 주는 것을 알 수 있다.