Multi-agent Robot Research Articles

1A1-N-030 開発環境を統合したマルチロボットシステムアーキテクチャ : 協調動作を考慮したマルチロボットプログラミング環境(マルチエージェントロボットシステム,生活を支援するロボメカ技術のメガインテグレーション)

1A1-N-030 開発環境を統合したマルチロボットシステムアーキテクチャ : 協調動作を考慮したマルチロボットプログラミング環境(マルチエージェントロボットシステム,生活を支援するロボメカ技術のメガインテグレーション)

2P1-3F-B1 冗長多脚ロボットの制御システム

本研究では, 柔軟な関節機構を有するクモヒトデの工学的実現を考慮し, その関節機構を, セル構造を持つエージェント群として表現することで, マルチエージェントシステムによる制御システムについて検討する。

V-Lab-a virtual laboratory for autonomous agents-SLA-based learning controllers

In this paper, we present the use of stochastic learning automata (SLA) in multiagent robotics. In order to fully utilize and implement learning control algorithms in the control of multiagent robotics, an environment for simulation has to be first created. A virtual laboratory for simulation of autonomous agents, called V-Lab is described. The V-Lab architecture can incorporate various models of the environment as well as the agent being trained. A case study to demonstrate the use of SLA is presented.

Multi-Agent Robotic Systems

From the Publisher: Providing a guided tour of the pioneering work and major technical issues, Multiagent Robotic Systems addresses learning and adaptation in decentralized autonomous robots. Its systematic examination demonstrates the interrelationships between the autonomy of individual robots and the emerged global behavior properties of a group performing a cooperative task. The author also includes descriptions of the essential building blocks of the architecture of autonomous mobile robots with respect to their requirement on local behavioral conditioning and group behavioral evolution.After reading this book you will be able to fully appreciate the strengths and usefulness of various approaches in the development and application of multiagent robotic systems. It covers:oWhy and how to develop and experimentally test the computational mechanisms for learning and evolving sensory-motor control behaviors in autonomous robotsoHow to design and develop evolutionary algorithm-based group behavioral learning mechanisms for the optimal emergence of group behaviorsoHow to enable group robots to converge to a finite number of desirable task states through group learningoWhat are the effects of the local learning mechanisms on the emergent global behaviorsoHow to use decentralized, self-organizing autonomous robots to perform cooperative tasks in an unknown environmentEarlier works have focused primarily on how to navigate in a spatially unknown environment, given certain predefined motion behaviors. What is missing, however, is an in-depth look at the important issues on how to effectively obtain such behaviors in group robots and how to enable behavioral learning and adaptation at the group level.Multiagent Robotic Systems examines the key methodological issues and gives you an understanding of the underlying computational models and techniques for multiagent systems.

マルチエージェントロボットにおける協調行動学習のための進化シミュレーション

本研究では、マルチエージェントロボットシステムを例題として各エージェントに進化アルゴリズムをもたせることにより、自律分散エージェントにおいて、どんな知性がどのような創発行動を引き起こすかを検証することを目的としている。ここでは、他のエージェントとの衝突回避行動と個体エネルギー回復のための目標到達行動がともに与えられているマルチエージェントロボットに対し、さらに環境マップ生成という集団行動目標を与え、協調行動学習のための進化シミュレーションを行った。各エージェントロボットは、自己保存のための局所的な個体行動と協調作業のための大局的な集団行動の相反する2つの行動目的を与えられており、さらにエージェントが死滅した場合に最優良個体からのパラメータコピーを行う集団進化のためのアルゴリズムも有している。筆者らは、環境マップ生成を行うマルチエージェントロボットのための進化型行動学習シミュレータを作成し、これを用いて自己保存のための個体行動と集団としての協調行動の割合がエージェントの集団としての行動に与える影響について検証シミュレーションを行った。この結果、均質な行動特性を持つエージェント集団では、協調と単独の各パラメータのバランスに重点をおいた設定をすることによって比較的能力の高い協調行動が発現することが確認できた。

Hierarchic Social Entropy: An Information Theoretic Measure of Robot Group Diversity

As research expands in multiagent intelligent systems, investigators need new tools for evaluating the artificial societies they study. It is impossible, for example, to correlate heterogeneity with performance in multiagent robotics without a quantitative metric of diversity. Currently diversity is evaluated on a bipolar scale with systems classified as either heterogeneous or homogeneous, depending on whether any of the agents differ. Unfortunately, this labeling doesn't tell us much about the extent of diversity in heterogeneous teams. How can it be determined if one system is more or less diverse than anotherq Heterogeneity must be evaluated on a continuous scale to enable substantive comparisons between systems. To enable these types of comparisons, we introduce: (1) a continuous measure of robot behavioral difference, and (2) hierarchic social entropy, an application of Shannon's information entropy metric to robotic groups that provides a continuous, quantitative measure of robot team diversity. The metric captures important components of the meaning of diversity, including the number and size of behavioral groups in a society and the extent to which agents differ. The utility of the metrics is demonstrated in the experimental evaluation of multirobot soccer and multirobot foraging teams.

Smart office robot collaboration based on multi-agent programming

As a new Artificial Intelligence (AI) application to our everyday life, we designed and implemented a smart office environment in which various information appliances work collaboratively to support our office activities. In this environment, many cameras and infrared sensors allow handling robots and mobile robots to perform complex tasks such as printing and delivering document. The delivery task is a typical example of an important class of tasks supporting humans in the smart office. In this paper, such robots are modeled as robotic agents, and collaboration between the agents is realized using multi-agent programming. We have developed a multi-agent robot language (MRL) as an evolution of concurrent logic programming. MRL provides synchronous and asynchronous control of agents based on guarded Horn clauses. It also supports describing an advanced negotiation protocol using broadcast and incomplete messages, and making decisions using a set of logical rules. These features are unified within an MRL framework, yielding an intelligent integration of the robotic agents. We view the smart office environment as a human assistant system through agent collaboration, and this view is novel and extendable as AI for everyday functions.

Deadlock avoidance of a multi-agent robot based on a network of chaotic elements

Deadlock avoidance of a multi-agent robot based on a network of chaotic elements

Chaotic Evolutionary Parallel Computation on Intelligent Agents

This paper proposes evolutionary computation using chaotic dynamics rather than the conventional genetic algorithm (GA) for such intelligent agents as welfare robots. Proposed evolutionary computation applies chaotic retrieval to associative memory. We applied evolutionary computation to multiagent robots moving side by side in step. Evolutionary computation is basically parallel processing, so we implement its parallel processing algorithm on A-NET (Actors NETwork) parallel objectoriented computer to show usefulness of parallel processing in evolutionary computation.

MULTI-AGENT ROBOT ARCHITECTURES: THE DECOMPOSITION ISSUE AND A CASE STUDY

In this paper, some fundamental issues of modern multi-agent robot architectures are discussed. It is argued that the multi-agent approach provides the necessary flexibility and adaptivity for such architectures, and that the primary issue in designing a multi-agent robot architecture is the selection of the granularity level, i.e., the decision on decomposing the overall desired functionality physically or across tasks. It is explained why at the various system levels different decomposition grains are needed; physical components, tasks or hybrid. This granularity decision is made on the basis of specific criteria of control localization, knowledge decoupling and interaction minimization so as to identify the decision points of the overall functionality. The above criteria lead to a dual composition-decomposition relation, which provides a good basis for system scaling. The paper specializes the discussion to a proposed neuro-fuzzy multi-agent architecture, which is then applied to design the local path planning system of an indoor mobile robot.

Deadlock avoidance of a multi-agent robot based on a network of chaotic elements

(1998). Deadlock avoidance of a multi-agent robot based on a network of chaotic elements. Advanced Robotics: Vol. 13, No. 3, pp. 249-251.

Integrating skills into multi-agent systems

Currently, an important topic of robotic research is the design and development of multi-agent robot systems (MASs). In these a number of autonomous robots cooperate and coordinate themselves in order to pursue given goals. The agents of an MAS not only have to work autonomously or in cooperation with other agents, but in dynamic, relatively unstructured environments. Therefore, the agents require agent-specific but flexible skills to cope with their tasks and the environment's variability. On the other hand, the actions to be performed by agents in an MAS have to meet certain requirements imposed by the MAS's structure. The representation of actions has to support planning, inter-agent communication, task negotiation etc. In this paper, we describe a method of combining the agent-specific nature of skills with the requirements for a general action knowledge representation inherent to MASs, by presenting elementary operations (EOs) that provide an appropriate interface.

A distributed control architecture for autonomous mobile robots - implementation of the Karlsruhe Multi-Agent Robot Architecture (KAMARA)

The main advantage of distributed controlled robots and subsystems is the decentralized task execution by the system components. In this way, properties for the design of flexible control architectures like modularity, fault-tolerance, integrability and extendibility are easily obtained; furthermore, it is possible to use the concepts of distributed knowledge and decentralized world representation. On the other hand, coordination between the components, e.g. path planning for collision avoidance between both manipulators in a two-arm system, is very difficult to guarantee. To explain these concepts, the Karlsruhe Autonomous Mobile Robot (KAMRO), which is being developed at our institute, is considered. The robot system consists of several subcomponents like two manipulators, hand-eye cameras, one overhead camera and a mobile platform. Extensions to the distributed control architecture KAMARA (KArlsruhe Multi-Agent Robot Architecture) are described that are used to overcome coordination problems, e.g. as caused by the independent task execution of both manipulator systems. Furthermore, we explain in which way the decentralized world representation can be used for parallel task execution. The intelligent control architecture described is used to replace the former control architecture of the autonomous robot KAMRO. The advantages of distributed control systems will be summarized through the actual implementation of the architecture to the robot KAMRO.

A taxonomy for multi-agent robotics

A key difficulty in the design of multi-agent robotic systems is the size and complexity of the space of possible designs. In order to make principled design decisions, an understanding of the many possible system configurations is essential. To this end, we present a taxonomy that classifies multi-agent systems according to communication, computational and other capabilities. We survey existing efforts involving multi-agent systems according to their positions in the taxonomy. We also present additional results concerning multi-agent systems, with the dual purposes of illustrating the usefulness of the taxonomy in simplifying discourse about robot collective properties, and also demonstrating that a collective can be demonstrably more powerful than a single unit of the collective.

MARC: A multiagent robot control system

AbstractA system based on multiagents is proposed as an information processing model for robot programming. This model uses object‐oriented representation in which physical entities are described as logical objects. An agent is an active and independent logical entity which can control objects. Agents work cooperatively and concurrently, communicating with each other to perform the task assigned to the robot. An agent starts a new job or changes its behavior when a specific event occurs or when it receives a message from another agent. This multiagent system provides robot programmers with a useful mental model for dividing the task of a robot into several segments and distributing them to several personified processes which control subtasks.This paper presents an overview of an experimental system used for multiagent robot programming and describes its implementation on a Lisp machine which supports multiprogramming and object‐oriented programming. The Tower of Hanoi task is used to test the system. It demonstrates that the multiagent system, using function‐based and behavior‐based agents, provides a high‐level modular programming environment for robot control.