Hierarchical organization has become an unfashionable model of intelligent control within some communities of both natural and artificial intelligence. What has replaced it are models based on parallel distributed processes, both neural and behavior based, or dynamical systems theory, which denies modularity, let alone rigorous structure. In this paper we present experimental results demonstrating an artificial reactive hierarchybased system that outperforms fully parallel systems in a highly dynamic environment with a large number of conflicting goals. This work is conducted in Tyrrell’s (1993) Simulated Environment and can be seen as an extension of his work on comparing action selection mechanisms. We observe that the hierarchical strategy has also been well demonstrated in nature. We argue that, for complex intelligences, preserving full reactivity may not be worth the cost in terms of the complexity of action selection.

This paper describes an international effort to unify a multimodal behavior generation framework for Embodied Conversational Agents (ECAs). We propose a three stage model we call SAIBA where the stages represent intent planning, behavior planning and behavior realization. A Function Markup Language (FML), describing intent without referring to physical behavior, mediates between the first two stages and a Behavior Markup Language (BML) describing desired physical realization, mediates between the last two stages. In this paper we will focus on BML. The hope is that this abstraction and modularization will help ECA researchers pool their resources to build more sophisticated virtual humans.

Beth and Alan are sitting at a Fifth Avenue outdoors restaurant in Manhattan. Alan is telling Beth an exciting story about his vacation in Nice. Alan presents the story through gesture and speech. Then Beth's arm starts moving and her neck stiffens. We, the viewers, know that she's surprised to see an elephant in the middle of Manhattan, and that in 460 milliseconds her arm and hand motion will turn into a welldefined deictic gesture, her eyebrows will rise, and her mouth will open with surprise, at which point Alan will most certainly recognize the signs and look over at the elephant. But right now, at t-minus-460 milliseconds, Beth's gesture is barely recognizable as a communicative action, so Alan doesn't know for sure. And thus, before that all happens, in the next 460 milliseconds, Alan has to decide what to do about Beth's behavior. Should he stop telling his story? Or should he go on, in case Beth is simply adjusting her jacket? Decisions like these a

Abstract. This chapter addresses the problem of producing and maintaining progress in agent design. New architectures often hold important insights into the problems of designing intelligence. Unfortunately, these ideas can be difficult to harness, because on established projects switching between architectures and languages carries high cost. We propose a solution whereby the research community takes responsibility for re-expressing innovations as idioms or extensions of one or more standard architectures. We describe the process and provide an example — the concept of a Basic Reactive Plan. This idiom occurs in several influential agent architectures, yet in others is difficult to express. We also discuss our proposal’s relation to the the roles of architectures, methodologies and toolkits in the design of agents. 1

Abstract. This volume is intended to help advance the field of artificial neural networks along the lines of complexity present in animal brains. In particular, we are interested in examining the biological phenomena of modularity and specialized learning. These topics are already the subject of research in another area of artificial intelligence. The design of complete autonomous agents (CAA), such as mobile robots or virtual reality characters, has been dominated by modular architectures and context-driven action selection and learning. In this chapter, we help bridge the gap from neuroscience to artificial neural networks (ANN) by incorporating CAA. We do this both directly, by using CAA as a metaphor to consider requirements for ANN, and indirectly, by using CAA research to better understand and model neuroscience. We discuss the strengths and the limitations of these forms of modeling, and propose as future work extensions to CAA inspired by neuroscience.

. Endowing future robot generations with `plug-and-play' capabilities is one of the fascinating challenges of robotics research. These robots could be easily installed at home and effortlessly used immediately after switching them on. They could be commanded intuitively even by non-experts via multimodal interfaces, would dynamically adapt to ever-changing environmental conditions and never fail. Undoubtedly, current design and safety concepts, locomotion and manipulation capabilities, cooperation and communication abilities, reliability, and -- probably most importantly -- adaptability, learning capabilities and sensing skills have to be improved significantly to achieve this goal. Therefore, the first part of this paper identifies the major challenges and key technologies on our way towards truly autonomous humanoid robots. It also summarizes and comments current research trends and points out technological gaps that need to be closed by interdisciplinary research efforts. T...

One of the defining features of intelligent behaviour is the ordering of individual expressed actions into coherent, apparently rational patterns. Psychology has long assumed that hierarchical and sequential structures internal to the intelligent agent underlie this expression. Recently these assumptions have been challenged by claims that behaviour controlled by such structures is necessarily rigid, brittle, and incapable of reacting quickly and opportunistically to changes in the environment (Hendriks-Jansen 1996, Goldfield 1995, Brooks 1991a). This dissertation is intended to support the hypothesis that sequential and hierarchical structures are necessary to intelligent behaviour, and to refute the above claims of their impracticality. Three forms of supporting evidence are provided: • a demonstration in the form of experimental results in two domains that structured intelligence can lead to robust and reactive behaviour, • a review of recent research results and paradigmatic trends within artificial intelligence, and • a similar examination of related research in natural intelligence.

We present a methodology for designing and implementing interactive intelligences. The Constructionist Methodology – so called because it advocates modular building blocks and incorporation of prior work – addresses factors that we see as key to future advances in A.I., including interdisciplinary collaboration support, coordination of teams and large-scale systems integration. We test the methodology by building an interactive multi-functional system with a real-time perception-action loop. The system, whose construction relied entirely on the methodology, consists of an embodied virtual agent that can perceive both real and virtual objects in an augmented-reality room and interact with a user through coordinated gestures and speech. Wireless tracking technologies give the agent awareness of the environment and the user’s speech and communicative acts. User and agent can communicate about things in the environment, their placement and function, as well as more abstract topics such as current news, through situated multimodal dialog. The results demonstrate the Constructionist Methodology’s strength in simplifying the modeling of complex, multi-functional systems requiring architectural experimentation and exploration of unclear sub-system boundaries, undefined variables, and tangled data flow and control hierarchies.

Creative play requires a fertile but well-defined design space. This paper describes a design process for creating three-dimensional virtual reality play spaces that allow the development and exploration of social interactions and relationships. The process was developed as part of a commercial research effort to create an interactive virtual reality entertainment system that allows children to engage in creative and constructive play within an established action/adventure framework. The effort centres on designing AI characters for a constructive narrative. We claim that a behaviour-based architecture is an ideal starting point for developing agents for such a process, but that its full realization requires additional architectural structures and methodological support for the design process. Here we describe an architecture of these characters, relate architectural modularity to the realization of construction and play, and propose a three-layer design process for producing fertile and æsthetic constructive narratives. We also discuss our experience in implementing these ideals in an industrial setting.

One of the greatest obstacles to designing a mind is the complexity of integrating different process types, time frames and representational structures. This paper describes a methodology for addressing this obstacle, Behavior Oriented Designed (BOD), and explains it in the context of creating an agent capable of natural language dialogue. 1