1,721,011 research outputs found
Software Integration Guidelines for PEACH
No full textThis report contains background, terminology, and practical guidelines to be followed for the development of software components to be integrated in the demonstrators of the PEACH project. Part of the technology being presented here is still work in progress; these guidelines are themselves subject to research, and will evolve while the project progresse
Towards monitorino of group interaction and social roles via overhearing
No full textWe are investigating how to provide intelligent, pervasive support of group of people within so-called "smart environments". Our current main assumption, based on literature in psychology and organizational studies, is that a group performs some complex, routine task as a structured activity, that is, by following some protocols that allow its members to coordinate and share a common understanding about the current progress towards the group's goal and the roles currently played by each member. If this is the case, a condition to provide support to a group activity by artificial agents is to share the same understanding. To this end, we have identified two initial goals: first, being able to understand if a group activity is progressing with respect to its expected evolution, by analyzing what is happening within the smart environment; second, recognizing what are the social roles of the group members, taking in mind that these are not necessarily pre-assigned and may change in time. This paper sketches a preliminary approach to these issues and a computational model for an overhearer agent. We suggest a preliminary set of rules for conversation analysis and social role recognition, and validate them against the simple case of \emph{implicit organizations}, which - being artificial - follow well-known protocol
An Approach to the Integration of Peer-to-Peer Systems with Active Entironment
No full textWe adopt a form of group communication, called "channeled multicast", for active rooms and other scenarios featuring strict real-time requirements, inherently unreliable communication, and a continuously changing set of context-aware autonomous systems. In our approach, rooted in multi-agent and team programming, coordination and cooperation are supported via ``social awareness'' and overhearing; agents form "implicit organizations" by actively participating to playing a role on a channel. We present an approach to the integration of Internet-based peer-to-peer systems into active rooms by means of agents on "proxy devices" accessing the users' "home peers"; a home peer is a remote or delocalized repository of data and services. In addition to other advantages, our approach introduces a new flavor to the concept of neighborhood in P2P systems, reduces or even avoids the need for users to physically carry P2P-equipped mobile devices with them, and easily supports virtual environments created by the composition of smart rooms at different location
A multi-sensor fall detection system: A demonstrator for ITEA
No full textThis report describes motivations, objectives and software architecture of a multi-sensor fall detection monitor built by ITC-irst and DIT. At its core, an agent integrates data coming from a wireless accelerometer carried by a person and a video processing system overlooking the room. The architecture is extensible both to support new types of sensors and to integrate external services, such as alarms directed to a remote call center and local actuators for interacting with the users (e.g., text-to-speech and speech recognition systems). This work is part of an on-going investigation into support technologies for people living alone but requiring continuous attention because of age or fragile health. The current demonstrator of the system is to be installed with other examples of commercial and experimental technologies in the Piazza Garzetti show-room of ITEA in Trento (Italy
LoudVoice v2.0: a Java implementation of Channeled Multicast for PEACH
No full textThese are the release notes for the version 2.0 of LoudVoice, a Java package implementing a channeled multicast agent communication infrastructure. Included in this version, there are some utilities especially written for the Peach projec
An Intelligent Agent Architecture for Facilitating Knowledge Sharing Communication
No full textIn this paper, we describe an agent architecture for supporting collaborative distance learning. At the core of the architecture lies an agent that analyzes knowledge sharing interactions by applying Hidden Markov Models and Multidimensional Scaling. We show how the agent was trained to assess sequences of coded online student interaction, and
dynamically recognize (1) when students are having trouble learning the new concepts they share with each other, and (2) why they are having trouble. The results of this research are intended to assist an instructor or coaching agent in facilitating situations in which groups of users collaborate to share their knowledg
Context-Aware Distributed Applications
No full textTraditionally, context-aware applications are defined as applications that react appropriately to information sensed in the environment, as opposed to applications that elaborate only information explicitly provided by users. Context is (implicitly or explicitly) thought of as a collection of features of the (physical or virtual) environment which can affect the behavior of an application. Though this notion of context is relatively unproblematic in systems with central control, it raises a number of challenging issues when applied to distributed systems, namely systems in which control is distributed over a group of heterogeneous, autonomous, interacting entities (typically, agents). Indeed, in distributed applications, we cannot assume that autonomous
entities share a context, even though each of them uses contextual information for its operations. In this paper, we discuss in detail this claim and present a notion of context which seems to be adequate for distributed applications. For the sake of illustration, we show how this notion of context can be used in a multi-agent system for
semantic-based information and knowledge management in distributed environments, such as the Internet or a big corporate Intrane
With A Little Help From A Friend: Applying Overhearing To Teamwork
No full textSoftware agents must have some degree of autonomy in order to be able to adjust to changing and sometimes unpredictable situations due to communication problems. Introducing so-called "overhearers" for monitoring team activities and helping recovery from failures seems to be a very promising approach. In this paper, we claim that having a "global" representation of the interaction protocol can be useful to monitor team communication. Moreover, using this global description of group interactions, overhearers can monitor activities without requiring a priori, precise knowledge of which and how many agents are involved. We show how, given a global description of a protocol in terms of involved roles rather than agents, it is possible for an overhearer to monitor its evolution and detect, and recover from, certain types of failures. We evaluated an implementation of the overhearer monitoring a group of agents executing a Contract-Net Protoco
Software Support for Implicit Organizations in LoudVoice
No full textThis document is the final report on the work done during 2003 on the development platform and overhearing platform, part of the tasks of the Workpackages 1 and 2 of the PEACH project. Our focus was on the support of "ambient intelligence" provided via agents communicating by means of multicast network
A Reliable Computational Model for BDI Agents
No full textBDI (Belief, Desire, Intention) is a mature and commonly adopted architecture for intelligent agents. However, the current computational model adopted by BDI has a number of problems with concurrency control, recoverability and predictability. This has hindered the construction of agents having robust and predictable behaviour.
To this end, we propose to integrate "distributed transactions", a well-established technology in distributed systems, into the computational model of multi-agent systems based on the BDI architecture. Differently from common approaches, where so-called ACID (Atomic, Consistent, Isolated, Durable) transactions are used simply to operate on external resources such as databases, in our model transactions are the foundation of the operational semantics of intentions and of collaborative tasks within team of agents. They provide a predictable, well understood behaviour in case of partial or total failure of intentions to achieve their goals or even crashes of agents. Furthermore, distributed transactions provide a simple and clear extension of the BDI semantics from the single-agent case to teams of agents.
We discuss the development of an agent system having a computational model with well-defined correctness criteria. Instead of hardwiring robustness and fault-tolerant behaviour into agent plans, well defined notions of correctness exist at the semantic level. Verification can then be undertaken at the desired level of abstraction.
Two BDI interpreter prototypes have been developed to demonstrate the feasibility of our approach. The first, TOMAS, is a Java environment that execute intentions as "nested transactions". The second is a re-implementation of TOMAS within a J2EE application server, which can be used to develop session beans (i.e., business logic); it demonstrates how the model we propose nicely fits into a state-of-the-art environment for mission critical systems in domains such as e-business and Web service
- …
