This paper reviews "socially interactive robots": robots for which social human-robot interaction is important. We begin by discussing the context for socially interactive robots, emphasizing the relationship to other research fields and the di#erent forms of "social robots". We then present a taxonomy of design methods and system components used to build socially interactive robots. Finally, we describe the impact of these these robots on humans and discuss open issues. An expanded version of this paper, which contains a survey and taxonomy of current applications, is available as a technical report[61].

Developing a large belief network, like any large system, requires systems engineering to manage the design and construction process. We propose that network engineering follow a rapid prototyping approach to network construction. We describe criteria for identifying network modules and the use of `stubs' within a belief network. We propose an object oriented representation for belief networks which captures the semantic as well as representational knowledge embedded in the variables, their values and their parameters. Methods for evaluating complex networks are described. Throughout the discussion, tools which support the engineering of large belief networks are identified. 1. Introduction As belief networks become more popular and well understood as a tool for modeling uncertainty and as the computational power of belief network inference engines increases, belief networks are being applied to problems of increasing size and complexity. In the early 1990's, Pathfinder, at 109 nodes...

Modern computer systems are becoming increasingly complex ensembles of hardware and software. This complexity has many facets, such as the number and diversity of the constituent components, the range of technologies employed, and the complex interactions between the components. If these interactions are unexpected, they may lead to system failures that could result in undesirable or even catastrophic consequences. This is especially of concern in the large, increasingly sophisticated, systems that are being used to control critical services in application areas such as medicine, transport, defence, and finance. Many properties of a system become visible only upon integration (when the components are assembled) --- these are called `emergent' properties. They are particularly difficult to control and predict during the development process. The dependability of a system is one such property which we study in detail in this thesis. For systems to be dependable, their behaviour must be pr...

In this paper it is shown how specification of behavioural requirements from informal to formal can be integrated within knowledge engineering. The integration of requirements specification has addressed, in particular: the integration of requirements acquisition and specification with ontology acquisition and specification, the relations between requirements specifications and specifications of task models and problem solving methods, and the relation of requirements specification to verification.

Recent trends in software engineering, especially within the object-oriented community, reflect a clear trend toward systems engineering methods. Object oriented designs, meant for programming design, often take on the distinct appearance of system models of physical networks and devices. It is not immediately apparent how the areas of "modeling" and "programming" relate to one another, and why the convergence is taking place. To explore the convergence in depth, we discuss common concepts between models and programs, and discuss future trends in computer science which are forging a steady convergence between models and programs. The convergence is spawned by increased emphasis on object oriented design, distributed systems, complex systems, new forms of analog computation, and abstraction methodology. We close with a discussion of MOOSE, which provides a comprehensive modeling environment for both programmers and modelers.

Enterprise Modelling technology prescribes systematic use and computer aided documentation of knowledge about enterprises, their objectives, activities, concepts, and actors. The basis of this technology is the Enterprise Model, assumed to be developed by users and requirements engineers during the requirements acquisition process. The Enterprise Model consists of several submodels, that have user-oriented meta-knowledge structures for communication with human sources of information. The Enterprise Model maintains relationships between information system requirements and other relevant, sometimes intentional, knowledge, acquired about the application domain. The main objective of this technology is to obtain a better alignment between the business activities and the supporting information system. The technology presented here is also applicable in the area of business process reengineering. 1. Introduction Knowledge acquisition, known as a "bottleneck" problem in Artificial Intellig...

this paper also the process of modelling is discussed and the Enterprise Modelling approach is assessed by comparing it with a number of related approaches. 1. Introduction

One of the primary challenges for decision makers during concept exploration in engineering system design is selecting designs that are valuable throughout the operational lifetime of the system. The problem is even more difficult when designing Systems of Systems (SoS), which are dynamic, complex, higher-order systems that may be composed of both legacy and new component systems. There are several heuristics and qualitative guidelines for designing SoS in the literature, but there is a lack of practical quantitative methods for SoS concept exploration. Development of quantitative methods for SoS conceptual design will greatly improve the ability of decision makers to select SoS designs that will be value robust over time. These quantitative methods will allow decision makers to consider a larger and more complete set of alternative SoS designs than is possible with qualitative methods alone. A SoS tradespace exploration method is being developed by augmenting the existing Dynamic Multi-Attribute Tradespace Exploration method with SoS-specific considerations, such as the existence of a multi-level stakeholder value function, the incorporation of both legacy and new component systems, and the potential time-varying composition of the SoS. In this paper, a case study of an operationally responsive multi-concept surveillance system for disaster relief is used to illustrate this developing SoS tradespace exploration method. This case study was partially completed as a collaborative project between MIT and the Charles Stark Draper Laboratory, and later extended by MIT. Several surveillance concepts- satellite, aircraft, unmanned air vehicle, and sensor swarms- were considered as possible concept solutions to achieve the surveillance mission objectives. The above system concepts are compared on the same tradespace, demonstrating the ability of the proposed method to allow decision makers to quantitatively compare disparate single system concepts on a common performance

Distributed systems now encounter extreme heterogeneity in the form of diverse devices, network types etc., and also need to dynamically adapt to changing environmental conditions. Selfadaptive middleware is ideally situated to address these challenges. However, developing such software is a complex task. In this paper, we present the Gridkit self * approach to the engineering of reflective middleware; this embraces state of the art software engineering practices, and flexible dynamic adaptation mechanisms to better support system developers. Domain specific frameworks are modeled and developed to enhance configurability and reconfigurability. We evaluate this approach using case studies in the domains of service discovery and network overlays. These demonstrate the benefits of the approach in terms of aiding and simplifying the process of creating self-configuring and self-adaptive software.

Abstract. During the past decade a number of modelling frameworks for knowledge based systems have been developed. In this paper an approach to the comparison of modelling frameworks is proposed, based on the aims and purposes behind the frameworks. A purpose oriented comparison of the frameworks DESIRE, CommonKADS, PROTÉGÉ-II, MIKE, VITAL and KARL provides insight in their differences and similarities. 1