• 한국지능정보시스템학회:학술대회논문집
• /
• 한국지능정보시스템학회 2001년도 The Pacific Aisan Confrence On Intelligent Systems 2001
• /
• pp.59-64
• /
• 2001

The robot soccer simulation game is a dynamic multi-agent environment. In this paper we suggest a new reinforcement learning approach to each agent's dynamic positioning in such dynamic environment. Reinforcement learning is the machine learning in which an agent learns from indirect, delayed reward an optimal policy to chose sequences of actions that produce the greatest cumulative reward. Therefore the reinforcement learning is different from supervised learning in the sense that there is no presentation of input pairs as training examples. Furthermore, model-free reinforcement learning algorithms like Q-learning do not require defining or learning any models of the surrounding environment. Nevertheless it can learn the optimal policy if the agent can visit every state- action pair infinitely. However, the biggest problem of monolithic reinforcement learning is that its straightforward applications do not successfully scale up to more complex environments due to the intractable large space of states. In order to address this problem. we suggest Adaptive Mediation-based Modular Q-Learning (AMMQL)as an improvement of the existing Modular Q-Learning (MQL). While simple modular Q-learning combines the results from each learning module in a fixed way, AMMQL combines them in a more flexible way by assigning different weight to each module according to its contribution to rewards. Therefore in addition to resolving the problem of large state effectively, AMMQL can show higher adaptability to environmental changes than pure MQL. This paper introduces the concept of AMMQL and presents details of its application into dynamic positioning of robot soccer agents.

• 한국시뮬레이션학회:학술대회논문집
• /
• 한국시뮬레이션학회 2001년도 The Seoul International Simulation Conference
• /
• pp.321-324
• /
• 2001

The robot soccer simulation game is a dynamic multi-agent environment. In this paper we suggest a new reinforcement learning approach to each agent's dynamic positioning in such dynamic environment. Reinforcement Beaming is the machine learning in which an agent learns from indirect, delayed reward an optimal policy to choose sequences of actions that produce the greatest cumulative reward. Therefore the reinforcement loaming is different from supervised teaming in the sense that there is no presentation of input-output pairs as training examples. Furthermore, model-free reinforcement loaming algorithms like Q-learning do not require defining or loaming any models of the surrounding environment. Nevertheless it can learn the optimal policy if the agent can visit every state-action pair infinitely. However, the biggest problem of monolithic reinforcement learning is that its straightforward applications do not successfully scale up to more complex environments due to the intractable large space of states. In order to address this problem, we suggest Adaptive Mediation-based Modular Q-Learning(AMMQL) as an improvement of the existing Modular Q-Learning(MQL). While simple modular Q-learning combines the results from each learning module in a fixed way, AMMQL combines them in a more flexible way by assigning different weight to each module according to its contribution to rewards. Therefore in addition to resolving the problem of large state space effectively, AMMQL can show higher adaptability to environmental changes than pure MQL. This paper introduces the concept of AMMQL and presents details of its application into dynamic positioning of robot soccer agents.

• 한국정밀공학회지
• /
• 제29권9호
• /
• pp.978-985
• /
• 2012

Starfish are a critical problem for fishermen since they eat every farming product including shellfish. The number of starfish is increasing dramatically because they have no natural enemy underwater. We consider the concept of capturing starfish using a semi-autonomous robot. A new underwater robot design to capture starfish is proposed using cooperation between humans and the robot. A requirements list for the robot is developed and two conceptual designs are proposed. Each robot is designed as a modular platform. The kinematic and dynamic performance of each robot is analyzed and compared. This study is a starting point for developing a starfish capture robot and designing underwater robots for other applications. In the near future, a prototype will be assembled and tested in a marine environment.

• 한국산학기술학회논문지
• /
• 제21권5호
• /
• pp.337-343
• /
• 2020

지난 수십 년간, 소형 로봇 가운데 동물과 곤충을 모방한 생체모방 로봇은 인간이 신체적으로 접근할 수 없는 영역에서 특별한 임무를 위해 개발되어 왔다. 최근 들어, 사람의 접근이 제한되는 공간(예 : 고농축 방사능 보관지역, 바이러스 지역, 대테러 위험지역 등)이 늘어나면서 로봇의 활용범위가 더욱 다양해지고 있으며, 과거에는 사람만 가능했던 많은 행위들이 소형 로봇으로의 대체가 시도 되고 있다. 그 중에서도 보행 로봇의 최적 움직임은 이동하는 표면의 특성(예: 거칠기, 곡률, 경사, 재료 등)에 의해 결정될 수 있다. 본 연구에서는 소형 보행 로봇에 적용하기 위한 구조가 간단하고 효율적으로 구동 가능한 압전세라믹 벤더 엑츄에이터를 제안하였다. 유한요소 해석법을 활용한 동적 모델링을 통해 구동원의 형상을 최적화하여 로봇의 이동 성능을 극대화 하였고, 제작과 실험을 통하여 그 결과를 검증하였다. 제작된 엑츄에이터는 무부하 조건에서 최대속도 236mm/s로 이동 하였고, 5g의 부하를 적재하고 156mm/s의 속도로 이동 가능함을 확인 하였다. 제안된 다족형 액추에이터는 수행해야 할 임무와 요구 성능에 따라 모듈식으로 추가가 가능한 장점이 있다.