• 한국지능시스템학회논문지
• /
• 제16권1호
• /
• pp.8-17
• /
• 2006

군집 로봇시스템에서 로봇들 간의 협조행동 및 군 행동을 하기 위해서는 로봇의 센싱과 통신 기능은 필수적이다. 일반적으로 대역적 통신 시스템에서 로봇의 대수가 증가하면 통신자원의 제한과 정보의 범람이 발생한다. 따라서 이 경우 지역적 통신방법이 유리하다. 본 논문에서는 지역적 통신 방식을 제안하고, 정보의 전파(propagation) 해석을 통하여 최적의 통신 반경을 찾는 방법들을 제시한다. 또한 로봇이 정보를 획득하고 소실할 경우 발생하는 군(group)의 카오스 행동을 피하기 위한 최적의 조건을 제시한다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2003년도 ICCAS
• /
• pp.2670-2675
• /
• 2003

A human-friendly interactive system, based on the harmonious symbiotic coexistence of human and robots, is explored. Based on interactive technology paradigm, a robotic cane is proposed for blind or visually impaired travelers to navigate safely and quickly among obstacles and other hazards faced by blind pedestrians. Robotic aids, such as robotic canes, require cooperation between humans and robots. Various methods for implementing the appropriate cooperative recognition, planning, and acting, have been investigated. The issues discussed include the interaction of human and robot, design issues of an interactive robotic cane, and behavior arbitration methodologies for navigation planning.

• 대한기계학회논문집A
• /
• 제24권12호
• /
• pp.3072-3079
• /
• 2000

Assembly tasks are often performed by one robot with fixtures. This type of assembly system has low flexibility in terms of the variety of parts and the part-presentation the system can handle. This paper addresses assembly without fixtures using two-manipulator robot. An active method using force feedback is proposed for the peg-in-hole assembly in highly uncertain environment. Assembly states are defined as status having unique motion constraints and events are modeled as variation of the environmental force. The states are recognized through identification of the events using two 6-d. o. f. force/moment sensors. The proposed method is verified and evaluated by experiments with round peg-in-hole assembly.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 1997년도 추계학술대회 논문집
• /
• pp.507-512
• /
• 1997

This paper presents a method of finding optitnal joint torques of two robots when they hold an object simultaneously. Although the importance of the multiple cooperating robot system increases for more flcviblc ni;mufacturing automation, dynamic solutions to multi-robot system forming closcd kinematic chain is not easy to find. Newton-Eulcr approach is used for the dynamic formulation of two robots fonn~ng closcd kincmatic chains gmsping a rigid body object. The nrcthodology to optimize the joint torques to satisfy given criterta and obtain bettcr control of the system is discussed. The scheme is illustrated by an example.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2000년도 제15차 학술회의논문집
• /
• pp.73-73
• /
• 2000

Assembly is usually performed by one robot and fixtures. This type of assembly system has Low flexibility in terms of variety of parts and part-presentation that the system can handle. This paper addresses assembly without fixtures using two-manipulator robot. An active method using force feedback is proposed for the peg-in-hole assembly in highly uncertain environment. Assembly states are described by extended contact relations. Qualitative templates for events are easily derived from the token vector of the Petrinet model. The states are recognized through identification of the events using two 6-d.o.f force/moment sensors. The proposed method is verified and evaluated through experiments with round peg- in-hole assembly.

• 제어로봇시스템학회논문지
• /
• 제19권12호
• /
• pp.1132-1137
• /
• 2013

In swarm robot systems, it is vital to maintain network connectivity to ensure cooperative behavior between robots. This paper deals with the behavior control algorithm of the swarm robots for maintaining network connectivity. To do this, we divide swarm robots into search-robots, base-robots, and relay-robots. Using these robots, we propose behavior control algorithm to maintain network connectivity. The behavior control algorithms to maintain network connectivity are proposed for the local path planning using virtual force and global path planning using the Delaunay triangulation, respectively. Finally, we demonstrate the effectiveness and applicability of the proposed method through some simulations.

• 전자공학회논문지S
• /
• 제34S권8호
• /
• pp.56-64
• /
• 1997

In distributed autonomous robotic systems, each robot must behaves by itself according to the its states and environements, and if necessary, must cooperates with other orbots in order to carray 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, the new learning and evolution method based on reinforement learning having delayed reward ability and distributed genectic algorithms is proposed for behavior learning and evolution of collective autonomous mobile robots. Reinforement learning having delayed reward is still useful even though when there is no immediate reward. And 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 perfodrmance of evolution, selective crossover using the characteristic of reinforcement learning is adopted in this paper, we verify the effectiveness of the proposed method by applying it to cooperative search problem.

• 한국산업융합학회 논문집
• /
• 제18권2호
• /
• pp.99-109
• /
• 2015

This study proposes a new approach to analyze the impedance and the elasticity of a serial chain of spring-damper system, areal-time collision-free trajectory generation algorithm is proposed. The reference points on a trajectory connected by the spring-damper system have a mechanism for self-position adjustment to solve a collision problem by the impedance, and the local adjustment of each reference point is propagated through the elasticity to a real robot at the end of the spring-damper system. As a result, the overall trajectory consisting of the reference points becomes free of collision with environmental obstacles and efficient having the shortest distance as possible. In this process,, the reference points connected by the spring-damper system take role of virtual robot as global guidance for a real robot, and a cooperative is carried out by the system of robots. A control technology is proposed to implement for mobile robot.

• 제어로봇시스템학회논문지
• /
• 제21권9호
• /
• pp.898-902
• /
• 2015

Alternative radio-navigation technologies aim at providing continuous navigation solution even if one cannot use GNSS (Global Navigation Satellite System). In shadowing region such as indoor environment, GNSS signal is no longer available and the alternative navigation system should be used together with GNSS to provide seamless positioning. For soldiers in battlefield where GNSS signal is jammed or in street battle, the alternative navigation system should work without positioning infrastructure. Moreover, the radio-navigation system should have scalability as well as high accuracy performance. This paper presents a TWR (Two-Way-Ranging)-based cooperative positioning system (CPS) that does not require location infrastructure. It is assumed that some members of CPS can obtain GNSS-based position and they are called mobile anchors. Other members unable to receive GNSS signal compute their position using TWR measurements with mobile anchors and neighboring members. Error propagation in CPS is analytically studied in this paper. Error budget for TWR measurements is modeled first. Next, location error propagation in CPS is derived in terms of range errors. To represent the location error propagation in the CPS, Location Error Propagation Indicator (LEPI) is proposed in this paper. Simulation results show that location error of tags in CPS is mainly influenced by the number of hops from anchors to the tag to be positioned as well as the network geometry of CPS.

• 한국컴퓨터정보학회논문지
• /
• 제29권1호
• /
• pp.11-19
• /
• 2024

멀티에이전트는 전장 교전 상황, 무인 운송 차량 등 다양한 실제 협동 환경에 사용될 수 있다. 전장 교전 상황에서는 도메인 정보의 제한으로 즉각적인 보상(Dense Reward) 설계의 어려움이 있어 명백한 희소 보상(Sparse Reward)으로 학습되는 상황을 고려해야 한다. 본 논문에서는 전장 교전 상황에서의 아군 에이전트 간 협업 가능성을 확인하며, 희소 보상 환경인 Multi-Robot Warehouse Environment(RWARE)를 활용하여 유사한 문제와 평가 기준을 정의하고, 강화학습 라이브러리인 Ray RLlib의 QMIX 알고리즘을 사용하여 학습 환경을 구성한다. 정의한 문제에 대해 QMIX의 Agent Network를 개선하고 Random Network Distillation(RND)을 적용한다. 이를 통해 에이전트의 부분 관측값에 대한 패턴과 시간 특징을 추출하고, 에이전트의 내적 보상(Intrinsic Reward)을 통해 희소 보상 경험 획득 개선이 가능함을 실험을 통해 확인한다.