We address the problem of propagating a piece of information among robots scattered in an environment. Initially, a single robot has the information. This robot searches for other robots to pass it along. When a robot...
详细信息
We address the problem of propagating a piece of information among robots scattered in an environment. Initially, a single robot has the information. This robot searches for other robots to pass it along. When a robot is discovered, it can participate in the process by searching for other robots. Since our motivation for studying this problem is to form an ad hoc network, we call it the Network Formation Problem. In this paper, we study the case where the environment is a rectangle and the robots' locations are unknown but chosen uniformly at random. We present an efficient network formation algorithm, Stripes, and show that its expected performance is within a logarithmic factor of the optimal performance. We also compare Stripes with an intuitive network formation algorithm in simulations. The feasibility of Stripes is demonstrated with a proof-of-concept implementation.
In this paper, we present the layered agent architecture INTERRAP which has been developed to cope with the basic requirements for Cooperative Intelligent Systems (CIS): agents shall behave in a situated, efficient, a...
详细信息
In this paper, we present the layered agent architecture INTERRAP which has been developed to cope with the basic requirements for Cooperative Intelligent Systems (CIS): agents shall behave in a situated, efficient, and goal-directed manner, and they shall be able to interact (ia.e. coordinate and collaborate) with other agents. Over the past few years, agent architectures combining reactive with deliberative facilities have become very trendy in DAI. However, most attempts end up at the local planning layer and fail to take into account important mechanisms for interaction and collaboration among agents. INTERRAP extends previous attempts to build layered architectures by a cooperation component which holds cooperation knowledge as well as a library of joint plans. The modules of the model and the flow of control among them are explained. The model is described and evaluated by the FORKS application, the simulation of an automated loading-dock.
The coordination of lot-sizing decisions in an inter-organizational supply chain allows the involved companies to find a global production plan with lower total costs compared to individual and non-coordinated plans. ...
详细信息
The coordination of lot-sizing decisions in an inter-organizational supply chain allows the involved companies to find a global production plan with lower total costs compared to individual and non-coordinated plans. Usually, the involved companies (referred to as agents in the following) are independent, self-interested and have private information. This contribution studies a distributed lot-sizing problem where each item may be produced by at least two agents. A collaborative planning approach based on simulated annealing is used to coordinate the individual plans of the agents. In collaborative planning mechanisms a general design challenge is the acceptance of a new incumbent solution during the negotiation process because it requires the elicitation and aggregation of the agents preferences. To deal with this challenge, we integrate the calculation of side payments based on the Shapley value into the negotiation. The computation of the Shapley-based side payments is demanding due to the NP-hardness of the characteristic function, which is represented in terms of the lot-sizing model at hand. Three different ways of computing Shapley-based side payments are presented. Our computational study shows significant differences with respect to the runtime and to a lesser extend with respect to the quality of the found solutions. Most importantly, the approaches using side payments outperform an approach without side payments in 225 out of 272 cases. (C) 2015, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved.
Exploration is a central issue for autonomous agents which must carry out navigation tasks in environments of which a description is not known a priori. In our approach the environment is described, from a symbolic po...
详细信息
Exploration is a central issue for autonomous agents which must carry out navigation tasks in environments of which a description is not known a priori. In our approach the environment is described, from a symbolic point of view, by means of a graph;clustering techniques allow for further levels of abstraction to be defined, leading to a multi-layered representation. In this work we propose an unsupervised exploration algorithm in which several agents cooperate to acquire knowledge of the environment at the different abstraction levels. All agents are equal and pursue the same local exploration strategy;nevertheless, the existence of multiple levels of abstraction in the environment representation allows for the agents' behavior to differ. Agents carry out exploration at different abstraction levels, aimed at reproducing an ideal exploration profile;each agent dynamically selects its exploration level, based on the current demand. Inter-agent communication allows for the agents to share their knowledge and to record acquaintances of the other agents. A communication protocol for organizing teams of agents is provided.
The range of application areas to which distributed processing has been applied effectively is limited. In order to extend this range, new models for organizing distributed systems must be developed.
The range of application areas to which distributed processing has been applied effectively is limited. In order to extend this range, new models for organizing distributed systems must be developed.
Continuous state plants place specific demands on the structure and operation of multi-agent, multi-paradigm distributed intelligent controllers. A controller for a deep shaft mine winder is proposed. Its design demon...
详细信息
Reliable and efficient distributed algorithms for power restoration are essential for self-healing electrical smart grids. Therefore, this paper presents a Multi-Agent System (MAS) for automatic restoration in power d...
详细信息
ISBN:
(纸本)9781509001910
Reliable and efficient distributed algorithms for power restoration are essential for self-healing electrical smart grids. Therefore, this paper presents a Multi-Agent System (MAS) for automatic restoration in power distribution networks. Moreover, as electrical demand fluctuates on the hourly and daily basis, an ensemble learning algorithm has been adopted for short-term forecasting of electrical energy demand. The prediction methodology is incorporated into the restoration algorithm in order to obtain a capacity-based restoration solution. Experiments carried out in two electrical networks demonstrate the importance and accuracy of the demand prediction algorithm and the feasibility of the MAS for system reconfiguration in decentralized power utilities.
This paper investigates the multi-agent cooperation problems in Web services domain. For Pareto-optimal Nash equilibrium, reinforcement learning algorithms are used to solve the coordination problem in cooperative env...
详细信息
ISBN:
(纸本)9781509026289
This paper investigates the multi-agent cooperation problems in Web services domain. For Pareto-optimal Nash equilibrium, reinforcement learning algorithms are used to solve the coordination problem in cooperative environments. Most previous works study the deterministic gain of a state. However, in practical service environments, the gain may be nondeterministic due to unstable Quality of Service (QoS). To avoid local optimal solution, we use an improved update function. The experimental results show that proposed reinforcement learning algorithm outperforms other learning methods.
In this paper, we consider each of the nine BDI principles defined by Rao and Georgeff based on Bratman's asymmetry thesis, and we verify which ones are satisfied by Rao's AgentSpeak(L), a logic programming la...
详细信息
In this paper, we consider each of the nine BDI principles defined by Rao and Georgeff based on Bratman's asymmetry thesis, and we verify which ones are satisfied by Rao's AgentSpeak(L), a logic programming language inspired by the BDI architecture for cognitive agents. In order to set the grounds for the proofs, we first introduce a rigorous way in which to define the informational, motivational, and deliberative modalities of BDI logics for AgentSpeak( L) agents, according to its structural operational semantics that we introduced in a recent paper. This computationally grounded semantics for the BDI modalities forms the basis of a framework that can be used to further investigate BDI properties of AgentSpeak( L) agents, and contributes towards establishing firm theoretical grounds for a BDI approach to agent-oriented programming.
This paper proposes Sustainable Learning Economy (SLE) based on market principles. SLE will utilize the blockchain technology in order to let learners trade their learning results in cryptocurrencies, which in turn gi...
详细信息
ISBN:
(纸本)9789897584176
This paper proposes Sustainable Learning Economy (SLE) based on market principles. SLE will utilize the blockchain technology in order to let learners trade their learning results in cryptocurrencies, which in turn gives the learners a strong motivation to acquire the knowledge independently to gain their rewards. The main challenge in SLE has been guaranteeing learning quality;however, this could be resolved using competencybased learning (CBL), an efficient learning method to prioritize acquired knowledge over knowledge acquisition. Unfortunately, due to social, corporate, and educational demands, competency models require significant time, manpower, and expertise. While Conventional Competency Management Systems (CMS) reduces costs by providing an integrated environment for CBL operations, it is not able to reduce development costs. Therefore, to reduce development costs, this paper developed a Smart CMS, which harnessed concepts of distributed artificial intelligence, the power of internet resources, and network analysis technology to automatically develop competencies tailored to the purpose, strengths, and characteristics of individual learners.
暂无评论