The purpose of this paper is to provide a broad overview of the WITAS Unmanned Aerial Vehicle Project. The WITAS UAV project is an ambitious, long-term basic research project with the goal of developing technologies and functionalities necessary for the successful deployment of a fully autonomous UAV operating over diverse geographical terrain containing road and traffic networks. Theproject is multi-disciplinary in nature, requiring many different research competences, and covering a broad spectrum of basic research issues, many of which relate to current topics in artificial intelligence. A number of topics considered are knowledge representation issues, active vision systems and their integration with deliberative/reactive architectures, helicopter modeling and control, ground operator dialogue systems, actual physical platforms, and a number of simulation techniques.

We describe an approach to integrate dialogue management, machine-learning and action planning in a system for dialogue between a human and a robot. Case-based techniques are used because they permit life-long learning from experience and demand little prior knowledge and few static hand-written structures. This approach has been developed through the work on an experimental dialogue system, called CEDERIC, that is connected to an unmanned aerial vehicle (UAV). A single case base and case-based reasoning engine is used both for understanding and for planning actions by the UAV. Dialogue experiments both with experienced and novice users, where the users have solved tasks by dialogue with this system, showed very adequate success rates.

We present a discourse model integrated with a case-based reasoning dialogue system which learns from experience. The discourse model is capable of solving references, manage subdialogues and respect the current topic in a dialogue in natural language. The framework is flexible enough not to disturb the learning functions, but allows dynamic changes to a large extent. The system is tested in a traffic surveillance domain together with a simulated UAV and is found to be robust and reliable.

As long as there have been computers, one goal has been to be able to communicate with them using natural language. It has turned out to be very hard to implement a dialog system that performs as well as a human being in an unrestricted domain, hence most dialog systems today work in small, restricted domains where the permitted dialog is fully controlled by the system.

In this thesis we present two dialog systems for communicating with an autonomous agent:

The first system, the WITAS RDE, focuses on constructing a simple and failsafe dialog system including a graphical user interface with multimodality features, a dialog manager, a simulator, and development infrastructures that provides the services that are needed for the development, demonstration, and validation of the dialog system. The system has been tested during an actual flight connected to an unmanned aerial vehicle.

The second system, CEDERIC, is a successor of the dialog manager in the WITAS RDE. It is equipped with a built-in machine learning algorithm to be able to learn new phrases and dialogs over time using past experiences, hence the dialog is not necessarily fully controlled by the system. It also includes a discourse model to be able to keep track of the dialog history and topics, to resolve references and maintain subdialogs. CEDERIC has been evaluated through simulation tests and user tests with good results.

We present a robotic dialogue system built on casebased reasoning. The system is capable of solving references and manage sub-dialogues in a dialogue with an operator in natural language. The approach to handle dialogue acts and physical acts in a unison manner together with the use of plans and subplans makes the system very flexible. This flexibility is used for learning purposes where the operator teaches the system a new word and the new knowledge can directly be integrated and used in the old plans. The learning from explanation capability makes the system adaptable to the operator's use of language and the domain it is currently operating in. The implementation of a case-based planner suggested in the paper will further increase the learning and adaptation degree.

The WITAS Unmanned Aerial Vehicle Project is a long term basic research project located at Linköping University (LIU), Sweden. The project is multi-disciplinary in nature and involves cooperation with different departments at LIU, and a number of other universities in Europe, the USA, and South America. In addition to academic cooperation, the project involves collaboration with a number of private companies supplying products and expertise related to simulation tools and models, and the hardware and sensory platforms used for actual flight experimentation with the UAV. Currently, the project is in its second phase with an intended duration from 2000-2003.

This paper will begin with a brief overview of the project, but will focus primarily on the computer vision related issues associated with interpreting the operational environment which consists of traffic and road networks and vehicular patterns associated with these networks.

In 2002, Linköping University Electronic Press began development of a journal article reviewing support system (JARSS) as a tool for Editors of electronic journals. The system-s database contains submitted articles, abstracts and other secondary information, reviews, and e-mail communication for the purpose of submission, reviewing and final acceptance. The interface maintains a log of all events pertaining to the article and the associated status changes. The successive status options of an article (received, under review, conditional accept, etc.) correspond to the editorial workflow. JARRS has been used since its inception to run the Artificial Intelligence Journal (AIJ), an Elsevier publication. In essence a service is offered to take care of the technical aspects of journal publication, allowing editors more time to solicit papers of high quality.

This paper states the need for interactive teaching materials for programming languages within the area of modeling and simulation. We propose an interactive teaching material for the modeling language Modelica inspired by existing tutoring systems for Java and Scheme.

The purpose of this new teaching material, called DrModelica, is to facilitate the learning of Modelica in a modeling and simulation environment. We have developed two versions of DrModelica, one that is based on Mathematica and another that is intended for the web. With the web version of DrModelica we hope for an increased usage of Modelica.

This paper states the need for interactive teaching materials for programming languages within the area of modeling and simulation. We propose an interactive teaching material for the modeling language Modelica inspired by existing tutoring systems for Java and Scheme.

The purpose of this new teaching material, called DrModelica, is to facilitate the learning of Modelica through an environment that integrates programming, program documentation and visualization. The teaching material is intended to be used for modeling and simulation related courses at the undergraduate and graduate level.

I propose a neo-classical structure for publishing and reviewing of scientific works. This proposal has the following characteristic components:

• Electronic “preprint archives” and other similar mechanisms where research articles are made publicly available without prior formal review are considered as true and full-fledged publication of research from the point of view of priority of results.
• Large parts of the reviewing process is done publicly and in the form of published review letters and other contributions to the scientific debate, rather than through anonymous and confidential review statements which dominate today. There is a switch from anonymous “pass-fail” reviewing towards open reviewing where the identity and the comments of the reviewers are made public.
• Since open reviewing happens after publication, rather than before, there is a second step where articles are promoted to “recommended” or “certified” status through the decision of a review committee. The requirements for certification are set at least as high as for the formally published journal articles of today, so that it counts like journal publication in a CV.
• Several techniques are foreseen for facilitating the selection process of the individual reader as well as for improving communication as such between researchers.
• One should accept that there are good reasons why there may be several articles (from the same author) presenting the same result. This suggests the introduction of a formal concept of a “result” which is represented by several publications, and to allow citations to refer to results rather than to some specific publication of the result.

I refer to this system as neo-classical because it assumes that peer review is done openly and after an article has been published. It is of course only proposed as a complement which can easily co-exist with the modern system, allowing each author to choose which of the two systems he or she wishes to use for a particular article.

The newly appeared Handbook of Knowledge Representation is an impressive piece of work. Its three editors and its forty-five contributors have produced twenty-five concise, textbook-style chapters that introduce most of the major aspects of the science of knowledge representation. Reading this book is a very positive experience: it demonstrates the breadth, the depth and the coherence that our field has achieved by now.

A substantial knowledgebase is an important part of many A.I. applications as well as (arguably) in any system that is claimed to implement broad-range intelligence. Although this has been an accepted view in our field since very long, little progress has been made towards the establishment of large and sharable knowledgebases. Both basic research projects and applications projects have found it necessary to construct special-purpose knowledgebases for their respective needs. This is obviously a problem: it would save work and speed up progress if the construction of a broadly sharable and broadly useful knowledgebase could be a joint undertaking for the field. In this article I wish to discuss the possibilities and the obstacles in this respect. I shall argue that the field of Knowledge Representation needs to adopt a new and very different paradigm in order for progress to be made, so that besides working as usual on logical foundations and on algorithms, we should also devote substantial efforts to the systematic preparation of knowledgebase contents.

This article introduces and uses a representation of defeasible inheritance networks where links in the network are viewed as propositions, and where defeasible links are tagged with a quantitative indication of the proportion of exceptions, called the doubt index. This doubt index is used for restricting the length of the chains of inference. The representation also introduces the use of defeater literals that disable the chaining of subsumption links. The use of defeater literals replaces the use of negative defeasible inheritance links, expressing "most A are not B". The new representation improves the expressivity significantly. Inference in inheritance networks is defined by a combination of axioms that constrain the contents of network extensions, a heuristic restriction that also has that effect, and a nonmonotonic operation of minimizing the set of defeater literals while retaining consistency. We introduce an underlying semantics that defines the meaning of literals in a network, and prove that the axioms are sound with respect to this semantics. We also discuss the conditions for obtaining completeness. Traditional concepts, assumptions and issues in research on nonmonotonic or defeasible inheritance are reviewed in the perspective of this approach.

This note describes how the notion of nonmonotonic reasoning emerged in Artificial Intelligence from the mid-1960s to 1980. It gives particular attention to the interplay between three kinds of activities: design of high-level programming systems for AI, design of truth-maintenance systems, and the development of nonmonotonic logics. This was not merely a development from logic to implementation: in several cases there was a development from a system design to a corresponding logic. The article concludes with some reflections on the roles and relationships between logicist theory and system design in AI, and in particular in Knowledge Representation.

Open peer review has been proposed for a number of reasons, in particular, for increasing the transparency of the article selection process for a journal, and for obtaining a broader basis for feedback to the authors and for the acceptance decision. The review discussion may also in itself have a value for the research community. These goals rely on the existence of a lively review discussion, but several experiments with open-process peer review in recent years have encountered the problem of faltering review discussions. The present article addresses the question of how lively review discussion may be fostered by relating the experience of the journal Electronic Transactions on Artificial Intelligence (ETAI) which was an early experiment with open peer review. Factors influencing the discussion activity are identified. It is observed that it is more difficult to obtain lively discussion when the number of contributed articles increases, which implies difficulties for scaling up the open peer review model. Suggestions are made for how this difficulty may be overcome.

Linköping University has recently created a separate entity, called Linköping University Electronic Press, for the unrefereed electronic publication of research articles and other university-related materials over the Internet. The present article presents the background for why the E-Press was created and the strategies which have been chosen for its operation at least during its initial period. The article identifies three key problems in the context of this strategy:

• The purely formal problems concerning what counts as a publication;
• The persistence problem of making sure that an electronically published article does not change over time;
• The reception problems concerning how fellow researchers and the academic community regard electronically published articles.

We describe how the formal problems and the persistence problems have been addressed in the E-Press initiative. With respect to the reception problems, we argue that scientific journals and journal-like conferences presently perform four distinct functions, and that these functions can be performed better if they are "unbundled" and addressed by other means. The four functions are:

• Publication in the narrow sense - making the article publicly available;
• Scientific quality control through reviewing;
• Selection of relevant articles for the benefit of the researcher-reader;
• Promotion of the scientific results of the author.

The Electronic Press focusses on the first one of these four functions. We discuss how the other three functions can be separated and performed by other means than through a conventional journal or quality conference proceedings.

Gilles Kahn was one of the most influential figures in the development of computer science and information technology, not only in Europe but throughout the world. This volume of articles by several leading computer scientists serves as a fitting memorial to Kahn's achievements and reflects the broad range of subjects to which he contributed through his scientific research and his work at INRIA, the French National Institute for Research in Computer Science and Control. The editors also reflect upon the future of computing: how it will develop as a subject in itself and how it will affect other disciplines, from biology and medical informatics, to web and networks in general. Its breadth of coverage, topicality, originality and depth of contribution, make this book a stimulating read for all those interested in the future development of information technology.

The WITAS project addresses the design of an intelligent, autonomous UAV (Unmanned Aerial Vehicle), in our case a helicopter. Its dialogue-system subprojects address the design of a deliberative system for natural-language and graphical dialogue with that robotic UAV. This raises new issues both for dialogue and for reasoning in real time. The following topics have been particularly important for us in various stages of the work in these subprojects: - spatiotemporal reference in the dialogue, including reference to past events and to planned or expected, future events - mixed initiative in the dialogue architecture of a complex system consisting of both dialogue-related components (speech, grammar, etc) and others (simulation, event recognition, interface to robot) and more recently as well - identification of a dialogue manager that is no more complex than what is required by the application - uniform treatment of different types of events, including the robot's own actions, observed events, communication events, and dialogue-oriented deliberation events - a logic of time, action, and spatiotemporal phenomena that facilitates the above. This paper gives a brief overview of the WITAS project as a whole, and then addresses the approaches that have been used and that are presently being considered in the work on two generations of dialogue subsystems.

The Software Individuals Architecture (SIA) is a framework fordefining software systems that are capable of self-modification and of reproductionon the level of an interpretive programming language. In abstractterms, a self-modifying system is a labelled tree containing scripts at someof its nodes; these scripts are effectively programs. A computation in sucha system executes a specific script. In doing so it maintains a local computationalstate, but it also uses and updates the labelled tree. The labelledtree, the local computational state, and the command language used for thescripts are all designed in such a way as to support self-modification andreproduction in a structured and orderly fashion.We have defined a practical system of this kind both on an abstract andformal level and as an implementation using Lisp as the host language. Thisarchitecture has been used as a platform for several applications, including inparticular the speech and natural-language dialogue system for an intelligentautonomous unmanned aerial vehicle (UAV) in the WITAS project. Thearchitecture design has been revised repeatedly as a result of using it for thisapplication as well as several others.

The present book is a festschrift in honor of Luigia Carlucci Aiello. The 18 articles included are written by former students, friends, and international colleagues, who have cooperated with Luigia Carlucci Aiello, scientifically or in AI boards or committees. The contributions by reputed researchers span a wide range of AI topics and reflect the breadth and depth of Aiello's own work

The concept of a 'publication' no longer applies only to printed works, information technology has extended its application to several other types of works. This article describes a facility called the Common Knowledge Library that publishes modules of formally structured information representing facts and knowledge of various kinds. Publications of this new type have some characteristics in common with databases, and others in common with software modules, however, they also share some important characteristics with traditional publications. A framework for citation of previous work is important in order to provide an incentive for contributors of such modules. Peer review - the traditional method of quality assurance for scientific articles - must also be applied, although in a modified form, for fact and knowledge modules. The construction of the Common Knowledge Library is a cumulative process, new contributions are obtained by interpreting the contents of existing knowledge sources on the Internet, and the existing contents of the Library are an important resource for that interpretation process.

When invoking a function or a procedure in an ordinary programming language, it is normally assumed that the arguments may be given as composite expressions, and that they are not restricted to atomic constants or variable symbols. However, although active web pages in HTML-based web servers can be viewed as a kind of procedures, they do not enjoy the same flexibility.  The present paper reports on a software package that extends the embedded web server in the ACL (Allegro Common Lisp) system and that provides it with the kind of functional flavor just described. In passing, the software also adds a number of other convenience measures to the LHTML (Lisp-encoded HTML) of the ACL server.

This paper addresses the two-way relation between the architecture for cognitive robots on one hand, and a logic of action and change that is adapted to the needs of such robots on the other hand. The relation goes both ways: the logic is used within the architecture, but we also propose that an abstract model of the cognitive robot architecture shall be used for defining the semantics of the logic. For this purpose, we describe a novel architecture called the Double Helix Architecture which, unlike earlier proposals, emphasizes a precise account of the metric discrete timeline and the computational processes that take place along that timeline. The computational model of the Double Helix Architecture corresponds to the semantics of the logic being used, namely the author's Cognitive Robotics Logic which is based on the 'Features and Fluents' theory.

One reason Qualitative Spatial Reasoning (QSR) is becoming increasingly important to Artificial Intelligence (AI) is the need for a smooth ‘human-like’ communication between autonomous agents and people. The selected, yet general, task motivating the work presented here is the scenario of an object configuration that has to be described by an observer on the ground using only relational object positions. The description provided should enable a second agent to create a map-like picture of the described configuration in order to recognize the configuration on a representation from the survey perspective, for instance on a geographic map or in the landscape itself while observing it from an aerial vehicle. Either agent might be an autonomous system or a person. Therefore, the particular focus of this work lies on the necessity to develop description and reconstruction methods that are cognitively easy to apply for a person.

This thesis presents the representation scheme QuaDRO (Qualitative Description and Reconstruction of Object configurations). Its main contributions are a specification and qualitative classification of information available from different local viewpoints into nine qualitative equivalence classes. This classification allows the preservation of information needed for reconstruction into a global frame of reference. The reconstruction takes place in an underlying qualitative grid with adjustable granularity. A novel approach for representing objects of eight different orientations by two different frames of reference is used. A substantial contribution to alleviate the reconstruction process is that new objects can be inserted anywhere within the reconstruction without the need for backtracking or rereconstructing. In addition, an approach to reconstruct configurations from underspecified descriptions using conceptual neighbourhood-based reasoning and coarse object relations is presented.

In this paper we describe a qualitative approach for natural language communication about vehicle traffic. It is an intuitive and simple model that can be used as the basis for defining more detailed position descriptions and transitions. It can also function as a framework for relating different aggregation levels. We apply a diagrammatic abstraction of traffic that mirrors the different possible interpretations of it and with this the different mental abstractions that humans might make. The abstractions are kept in parallel and according to the communicative context it will be switched to the corresponding interpretation.

We present a process for reconstructing object configurations described by a set of spatial constraints of the form (A northeast B) into a two-dimensional grid. The reconstruction process is cognitively easy for a person to fulfill and guides the user to avoid typical mistakes. For underspecified object configuration descriptions we suggest a strategy to handle coarse object relationships by representing a coarse object in a way that all disjunctive basic relationships that the coarse relationship consists of are represented within one reconstruction.

We provide a technique for describing, reconstructing and updating an object constellation of moving objects. The relations between the constituent objects, in particular axis-parallel and diagonal relations, are verbally expressed using the double cross method for qualitatively characterizing relations between pairs of objects.  The same underlying representation is used to reconstruct the constellation from the given description.

We provide a technique to reconstruct an object configuration that has been described on site by only using intrinsic and relative frames of reference into an absolute frame of reference, as seen from the survey perspective.