Cooperative Multi-Agent System

This paper introduces a fuzzy decision-making algorithm for robot behavior coordination. The algorithm belongs to the arbitration class of behavior coordination mechanisms, under which only one behavior is running at a time. However, it is possible to use a hierarchical decision mechanism for hierarchical behaviors without interference between hierarchical levels. With this fuzzy decision method it is possible to represent a specific model of the world where the robot evolves. This algorithm consists of defining a set of behaviors, a set of world states, a cost function for behaviors, a set of goals, and a set of constraints. For each behavior and actual world state pair, a cost function is computed. The cost of each pair is evaluated by the overall goals. Goals and constraints are aggregated using a fuzzy operator and the optimal choice is the behavior with the maximum resulting value. This algorithm was tested with success in realistic simulations of a goalkeeper soccer robot.

Integrating Business Intelligence (BI) processes in an information system requires a form of strategic scanning system for which the information is the main source of efficiency and decision support. A process of strategic scanning system network is primarily a cooperative approach to sharing knowledge that actors are "producers" of information. The dynamics of the actor’s interactions allow gradual building of shared knowledge. This paper proposes Multi Agent System (MAS) architecture which facilitates the integration of a process of strategic scanning system network in the information system, to emerge relevant information from simple information while ensuring the quality and safety information. In particular, this approach is geared towards supporting system properties specially focused on cooperative multi-agent system. It gives finally an overview of implementation of a prototype of the proposed solution limited for the moment to the integration of processes most use...

Integrating Business Intelligence (BI) processes in an information system requires a form of strategic scanning system for which the information is the main source of efficiency and decision support. A process of strategic scanning system network is primarily a cooperative approach to sharing knowledge that actors are "producers" of information. The dynamics of the actor's interactions allow gradual building of shared knowledge. This paper proposes Multi Agent System (MAS) architecture which facilitates the integration of a process of strategic scanning system network in the information system, to emerge relevant information from simple information while ensuring the quality and safety information. In particular, this approach is geared towards supporting system properties specially focused on cooperative multi-agent system. It gives finally an overview of implementation of a prototype of the proposed solution limited for the moment to the integration of processes most used in the majority of information systems.

RMAS ArtSapience was founded by the Robotics and Multi-Agent Systems research group (RMAS), the faculty of Media Engineering and Technology (MET), the German University in Cairo (GUC), Egypt. The German University in Cairo participated in the RoboCup in 2009 and ...

Robot soccer is a challenging platform for multi-agent research, involving topics such as real-time image processing and control. A team of robots must work together to put the ball in the opponent’s goal while at the same time defending their own goal. A good strategy for the robot defenders can determine who wins the robotic soccer game. Therefore, the goal of this study is to propose a strategy for the defenders using a production rule based on state diagrams. The rule can facilitate easy and rapid comprehension of certain behaviors with respect to two indicators, such as condition and action. The authors determine five key aspects as conditions, including the positions of two defender robots, the position of the goalkeeper and the ball, and coordination between two defender robots and the goalkeeper robot. Each robot has been set its own defence area and specific actions. They conducted three experiments namely simulator testing, real time testing, and ping pong testing to evalu...

Running time, flexibility and on-line adaptation are important features presenting to the decision making (DM) module of soccer robots. In this paper, the design of DM algorithm based on behaviour trees and fuzzy obstacle avoidance algorithm for control of soccer robots are presented. The basic elements of the behaviour tree (BT) used in decision making process of soccer robots are described. The integration of BT and fuzzy logic is proposed to increase the flexibility of the decision made by DM modules. Given algorithms have been extensively tested in simulation and provided satisfactory results in run time and adaptation to the new situation. Also, the efficient navigation procedure is designed using fuzzy obstacle avoidance algorithm. The obstacle avoidance algorithm uses fuzzy if-then rule base and inference mechanism for finding optimal path for avoidance of obstacles. The presented BT based decision making and fuzzy obstacle avoidance algorithms are used for control of the holonomic 4wheel-driven soccer robots. The obtained results demonstrate the effectiveness of proposed algorithms in soccer robot control.

En este documento presentamos el estado actual del equipo de STOX para participar en la competencia LARC / CBR Small Size Robot League 2018, en Joao Pessoa, Brasil. Inicialmente, mostramos la estructura de nuestro equipo en terminos de mecanica y hardware. A continuacion, describimos el funcionamiento de nuestros robots en el nivel inferior. Luego, presentamos nuestro nuevo enfoque de una estrategia mas inteligente para la seleccion tactica y la asignacion dinamica de roles, basada en maquinas jerarquicas de estado finito. Finalmente, creemos que la competencia LARC / CBR Small Size League es un evento adecuado para validar nuestro nuevo enfoque, con la ayuda de otros equipos de grupos de investigacion reconocidos en robotica a nivel regional.

We present our Small Size League (SSL) robot soccer team, CMDragons, which performed strongly at the RoboCup'13 competition, placing second out of twenty teams after a prolonged final match ending in penalty shoot-outs. We briefly present the robots' hardware and individual skills, and then focus on our multi-robot passing, attack, and defense planning and execution in the challenging SSL adversarial multirobot environment. We introduce a pass-ahead behavior, as well as a new dynamic two-stage planner, Coerce and Attack, which explicitly considers opponent defense to hypothetical attack patterns. The Coerce stage generates a coerce attack formation to coerce the opponent robots into leaving strategic openings. The Attack stage modifies the coerce attack pattern in a fluid manner to exploit openings in the defense using pass-ahead to attempt to score. We further present our threat-based defensive multi-robot algorithm which identifies potential threats based on the opponent positioning, and plans the defense accordingly. We present the performance of CMDragons at RoboCup'13 in terms of metrics that evaluate the effectiveness of the lowlevel skills as well as the high-level defense and offense.

Integrating Business Intelligence (BI) processes in an information system requires a form of strategic scanning system for which the information is the main source of efficiency and decision support. A process of strategic scanning system network is primarily a cooperative approach to sharing knowledge that actors are "producers" of information. The dynamics of the actor's interactions allow gradual building of shared knowledge. This paper proposes Multi Agent System (MAS) architecture which facilitates the integration of a process of strategic scanning system network in the information system, to emerge relevant information from simple information while ensuring the quality and safety information. In particular, this approach is geared towards supporting system properties specially focused on cooperative multi-agent system. It gives finally an overview of implementation of a prototype of the proposed solution limited for the moment to the integration of processes most used in the majority of information systems.

5 Design of an Action Selection Mechanism for Cooperative Soccer Robots Based on Fuzzy Decision Making ... control, neural networks, genetic algorithms, decision tree, decision making,fuzzy logic and ... In this paper a new methodology for decision making in RoboCup soccer ...

A method for simulation based reinforcement learning (RL) for a multi-agent system acting in a physical environment is introduced, which is based on Multi-Agent Actor-Critic (MAAC) reinforcement learning. In the proposed method, avatar agents learn in a simulated model of the physical environment and the learned experience is then used by agents in the actual physical environment. The proposed concept is verified using a laboratory benchmark setup in which multiple agents, acting within the same environment, are required to coordinate their movement actions to prevent collisions. Three state-of-the-art algorithms for multi-agent reinforcement learning (MARL) are evaluated, with respect to their applicability for a predefined benchmark scenario. Based on simulations it is shown that the MAAC method is most applicable for implementation as it provides effective distributed learning and suits well to the concept of learning in simulated environments. Our experimental results, which compare simulated learning and task execution in a simulated environment with that of task execution in a physical environment demonstrate the feasibility of the proposed concept.

A method for simulation based reinforcement learning (RL) for a multi-agent system acting in a physical environment is introduced, which is based on Multi-Agent Actor-Critic (MAAC) reinforcement learning. In the proposed method, avatar agents learn in a simulated model of the physical environment and the learned experience is then used by agents in the actual physical environment. The proposed concept is verified using a laboratory benchmark setup in which multiple agents, acting within the same environment, are required to coordinate their movement actions to prevent collisions. Three state-of-the-art algorithms for multi-agent reinforcement learning (MARL) are evaluated, with respect to their applicability for a predefined benchmark scenario. Based on simulations it is shown that the MAAC method is most applicable for implementation as it provides effective distributed learning and suits well to the concept of learning in simulated environments. Our experimental results, which compare simulated learning and task execution in a simulated environment with that of task execution in a physical environment demonstrate the feasibility of the proposed concept.

Integrating Business Intelligence (BI) processes in an information system requires a form of strategic scanning system for which the information is the main source of efficiency and decision support. A process of strategic scanning system network is primarily a cooperative approach to sharing knowledge that actors are "producers" of information. The dynamics of the actor's interactions allow gradual building of shared knowledge. This paper proposes Multi Agent System (MAS) architecture which facilitates the integration of a process of strategic scanning system network in the information system, to emerge relevant information from simple information while ensuring the quality and safety information. In particular, this approach is geared towards supporting system properties specially focused on cooperative multi-agent system. It gives finally an overview of implementation of a prototype of the proposed solution limited for the moment to the integration of processes most used in the majority of information systems.

In LegenDary project, we have been continuing to develop our team based on Agent2D to achieve our goals and purposes. This year, we have optimized our shoot system by using Reinforcement Learning. We have improved our defense system by adding a new Marking system. In this paper, we are going to briefly describe our latest achievement to these tasks and strategies.

Abstract—In RoboCup, robots must make quick decisions under uncertainty. To this end, this paper introduces a new approach to enable humanoid soccer robots to execute kicks quickly and ensure that they move the ball down field. This paper presents a kick engine capable of kicking at a variety of distances and angles and then describes a novel kick decision method for selecting from among a large set of possible kicks. This method prunes and orders the kicks according to a metric and then chooses the first possible kick that ...

Integrating Business Intelligence (BI) processes in an information system requires a form of strategic scanning system for which the information is the main source of efficiency and decision support. A process of strategic scanning system network is primarily a cooperative approach to sharing knowledge that actors are "producers" of information. The dynamics of the actor’s interactions allow gradual building of shared knowledge. This paper proposes Multi Agent System (MAS) architecture which facilitates the integration of a process of strategic scanning system network in the information system, to emerge relevant information from simple information while ensuring the quality and safety information. In particular, this approach is geared towards supporting system properties specially focused on cooperative multi-agent system. It gives finally an overview of implementation of a prototype of the proposed solution limited for the moment to the integration of processes most use...

This paper presents the system used by the team of the German University in Cairo (GUC) within the FESTO Hockey Challenge league that took place within RoboCup 2009. The goal of the FESTO Hockey Challenge is to have a competition between robotic teams where each team consists of three robots to compete in an Ice Hockey game. All robots are of the same mechanical, sensor and electronic capabilities so that the focus of the competition is to develop novel artificial intelligence techniques for robot control and coordination. The GUC team scored the 2 nd place in this competition after losing by penalty shoot outs in the final. The proposed approach for GUC team employed Hierarchical Fuzzy Logic Controllers (HFLCs) in which the low level behaviours are implemented using FLCs and the coordination between the behaviours is implemented by a high level layer. The coordination between the robotic agent team members is implemented by a situation based dynamic role allocation mechanism. The paper will describe the employed approaches and will report on the results achieved.

The new 3D environment introduces a more developed aspect of human soccer simulation which results in a better approximation of the reality. This compli-cated environment has proposed a challenging opportunity for specifying the agent's behavior. In addition to the relatively ...

The new 3D soccer simulation server, necessitated fundamental changes in the design of UI-AI team's architecture. Although this variation does not imply a deviation from the team's previous experiences in Portugal, Japan and Germany, a new hierarchy for the agent structure is inevitable. The UI-AI2007 3D soccer team is designed with a multi-layer architecture. This model, which is based on the most probable features of the simulator in Atlanta 2007 competitions, is explained in detail in the following proposal.

Dimension-reduced and decentralized learning is always viewed as
an efficient way to solve multi-agent cooperative learning in high
dimension. However, the dynamic environment brought by the
concurrent learning makes the decentralized learning hard to
converge and bad in performance. To tackle this problem, a
timesharing-tracking framework (TTF), stemming from the idea that
alternative learning in microscopic view results in concurrent
learning in macroscopic view, is proposed in this paper, in which
the joint-state best-response Q-learning (BRQ-learning) serves as
the primary algorithm to adapt to the companions' policies. With
the properly defined switching principle, TTF makes all agents
learn the best responses to others at different joint states. Thus
from the view of the whole joint-state space, agents learn the
optimal cooperative policy simultaneously. The simulation results
illustrate that the proposed algorithm can learn the optimal joint
behavior with less computation and faster speed compared with
other two classical learning algorithms.

Citation: Xin Chen, Bo Fu, Yong He, Min Wu. Timesharing-tracking framework for decentralized reinforcement learning in fully cooperative multi-agent system. IEEE/CAA Journal of Automatica Sinica, 2014, 1(2): 127-133

This paper describes the decision making process of a robot soccer agent based on the OODA Loop. In this game, an agent that is considered as a player in the field has to decide actions to perform based on its environment. The agent is able to observe the change in the environment using its sensors. Based on this sensory information, a decision for actions has to be done fast and yet accurate. Therefore, the OODA loop method is applied in the decision system to achieve a better architecture of reasoning. The method provide the robosoccer agent abilities to sense the environment, to think, and finally to act.

Abstract. This paper presents the initial research from Bahia Robotics Team. This is a new research group created to investigate the application of artificial intelligence methods in the standard problem of robotics soccer. In this work, fuzzy controllers are used to improve some abilities of the ...

Abstract We study Markov Decision Process (MDP) games with the usual±1 reinforcement signal. We consider the scenario in which the goal of the game, rather than just winning, is to maximize the number of wins in an allotted period of time (or maximize the expected reward in the same period).

This paper describes the design of humanoid robot systems of Urmia university robotic team (UURT) in KidSize competition RoboCup 2011, Istanbul, Turkey. It presents our current version of robot specifications and capabilities. The main research of the team focuses on Wireless communication between robots and controlling all three players in contact with each other. We also worked on the other areas such as mechanical structure, Image processing unit, controlling unit, Robot AI and behavior learning to have best action in the game. The project is described in two main parts: Hardware and Software. The hardware consists of the mechanical and electrical structure. The software is consists of self localization based on vision and object recognition, role engine that based on fuzzy logic, Trajectory Planning, Motion Controller and Network. Each robot is able to walk, pass, kick and dribble when it catches the ball.