Salvatore Lopes

A Norm-Governed Holonic Multi-agent System Metamodel

Modeling and designing systems that require a high level of coordination, control and automation is a very difficult task. The problem is the lack of design processes able to cover all the features these systems present. This paper presents an extension of the ASPECS metamodel for supporting organizational and normative principles and it allows to define models not only from an holonic agent viewpoint but also from a normative organization perspective. Moreover, our work emphasizes and makes it explicit the norms that regulate the structural, behavioral and finally adaptive aspect of an organizational system. The extended metamodel was experimented creating a Virtual Enterprise model for the optimization of distributions inside the logistic districts. This organizational model is implemented using JaCaMo.

A Notation for Modeling Jason-Like BDI Agents

The design and development of a large Multi Agent System (MAS) is a complex and difficult activity where a proper modeling notation may offer a significant contribution to the formulation of the best solution. The support provided by a specific CASE tool can significantly contribute to make the chosen approach technically valid and it is also a fundamental element of a feasible development strategy. The present work reports a UML profile and the related graphical notation for describing a MAS based on the Jason meta model. Moreover a specific CASE tool has been developed for supporting MASs design and automatic code generation. The proposed notation is shown in details using a classical example from the Jason tutorial (domestic robot).

Common and domain-specific metamodel elements for problem description in simulation problems

It is well known that the multi-agent system paradigm is well suited for modelling and developing simulations of complex systems belonging to several application domains. Simulation study aims at developing simulation models useful for representing, studying and analyzing entities and their behavior in a system according to specific purposes. With our work we are trying to understand what are the right elements to be considered and included in the description of a simulation problem. In order to root our resulting metamodel in the state of the art of multi-agent simulations we started from the study of twelve papers dealing with four different application domains: Crowd Dynamics, Traffic and Transportation, Electricity Power Engineering and Supply Chain and Logistic. From this study we obtained a metamodel that may be used by an analyst as a guideline and concept repository for facing a new system design. The metamodel is the result of a well defined approach that is described together with the obtained results consisting in one core metamodel containing elements that are common to all the four application domains and some domain extension contents. These latter contain the elements that are specific of each of the studied domains and are not present in the others.

Requirement analysis abstractions for AmI system design

Current trends in the AIs evolution are going towards enriching environments with intelligence in order to support humans in their everyday life. AmI systems are plunged in the real world and humans expect to interact with them in a way that is similar to the one they have with other humans. In this kind of systems, where eliciting requirements involves several documents and stakeholders (mainly users that will be the first consumers of the system), the requirement analysis phase can be affected by incomplete, ambiguous and imprecise information. Hence, the need to find a fruitful way for knowledge management and its representation at design time. In this paper we propose a set of abstractions to be used during the early requirements analysis of AmI systems development. The result is a simple and at the same time powerful set of concepts and guidelines for providing environment knowledge representation for AmI systems.

Towards a design process for modeling MAS organizations

The design of MAS organizations is a complex activity where a proper methodological approach may offer a significant advantage in enabling the conception of the best solution. Moreover, the aid provided by a supporting tool significantly contributes to make the approach technically sound and it is a fundamental ingredient of a feasible strategy to the development of large MASs. In this paper, we introduce a portion of methodological approach devoted to design MAS organizations and a preliminary version of a specific case tool, named MoT (Moise+ Tool), for supporting activities from design production to automatic code generation. MoT provides four kinds of diagrams based on a definite graphical notation for representing organizational elements. Our process is applied to a classical write paper simulator example. Results include portion of the automatically generated code according to Moise+ specifications.

Temporal Segmentation of MPEG Video Sequences

The video segmentation is a fundamental tool for the video semantic content evaluation. In multimedia application, videos are often in MPEG-1 format. In this paper, an algorithm for the automatic shot segmentation of MPEG-1 sequences is presented. The adopted method is based on heuristic considerations concerning the characteristics of MPEG-1 video streams. In particular, the pattern structure and the I-, B- and P-frame sizes are taken in account. The proposed algorithm has been applied to MPEG-1 sequences and some results are reported.

A Goal-Oriented Approach for Self-Configuring Mashup of Cloud Applications

This paper presents a general approach for automatic composing mashups of applications distributed over the cloud. The approach implies to wrap existing services in smart and autonomic entities, namely cloud capabilities. These are able to interact and coordinate themselves in order to establish different ways to orchestrate their contained services. The main enabler of this technology is based on an explicit distinction between users goals and the way to address them. A couple of language has been adopted to describe respectively the mashup logic in terms of goals and the available functionalities in terms of capabilities. A running example has been developed for extending B2B business processes of a fashion enterprise.

A New Parallel Differential Method for Optical Flow Estimation

Optical flow estimation is a recurrent problem in several disciplines and assumes a primary importance in a number of applicative fields such as medical imaging [12], computer vision [6], productive process control [4], etc. In this paper, a differential method for optical flow evaluation is being presented. It employs a new error formulation that ensures a more than satisfactory image reconstruction in those points which are free of motion discontinuity. A dynamic scheme of brightness-sample processing has been used to regularise the motion field. A technique based on the concurrent processing of sequences with multiple pairs of images has also been developed for improving detection and resolution of mobile objects on the scene, if they exist. This approach permits to detect motions ranging from a fraction of a pixel to a few pixels per frame. Good results, even on noisy sequences and without the need of a filtering pre-processing stage, can be achieved. The intrinsic method structure can be exploited for favourable implementation on multi-processor systems with a scalable degree of parallelism. Several sequences, some with noise and presenting various types of motions, have been used for evaluating the performances and the effectiveness of the method.

Agent-based simulation study for improving logistic warehouse performance

Logistic warehouses are critical nodes in a supply chain and improving their performance is a crucial issue when trying to avoid unproductive bottlenecks. Warehouse optimization involves several problems, some of which must be considered at the design stage and others during real-time operations. In this study, we performed an agent-based simulation to analyze the behavior of automatic logistic warehouses under the influence of specific factors, thereby obtaining indicators to supporting decision making during warehouse performance improvement. This study focused mainly on automatic warehouses where goods are moved by automatic guided vehicles.

Scene detection for MPEG video sequences

The indexing process required by content-based video databases is correlated to visual characteristics of their content. A preliminary step consists in the partitioning of the video into a sequence of short dynamic scenes, generally characterized by a set of homogeneous features. Each scene may be therefore characterized by the features of one of more representative frames, i.e. still images. In this paper, a method for scene detection of MPEG-1 and MPEG-2 video sequences is reported. The method does not need to decode the streams, because it is based on the analysis of their external characteristics, such as the frame pattern and the sized of I-, P- and B-frames. Changes in above characteristics are used to detect the frames representing each scene, using heuristics and statistical considerations. Since the analysis is based on very simple computation, the algorithm performs very fast. Its computational cost is linear dependent on the number of frames. On the other hand, the method is not well suited for video clip of short length, for which a statistical analysis is not significant. For its low computation cost and accuracy, the method is a suitable tool for preliminary segmentation step of long video sequences.