Discrete Event Systems (DES) are a special type of dynamic system. The "state" of these systems change at discrete instants in time and the term "event" represents the occurrence of discontinuous change (at possibly unknown intervals). Different Discrete Event Systems models are currently used for specification, verification, synthesis as well as for analysis and evaluation of different qualitative and quantitative properties of existing physical systems. The focus of this paper is the presentation of the automata and formal language model for DES introduced by Ramadge and Wonham and its application to the domain of mobile manipulator/observer agents. We demonstrate the feasibility of the DES framework for modeling, analysis and synthesis of some visually guided behaviors of agents engaged in navigational tasks and address synchronization issues between different components of the system. The use of DES formalism allows us to synthesize complex behaviors in a systematic fashion and gua...

Reactive deliberation is a novel robot architecture that has been designed to overcome some of the problems posed by dynamic robot environments. It is argued that the problem of action selection in nontrivial domains cannot be intelligently resolved without attention to detailed planning. Experimental evidence is provided that the goals and actions of a robot must be evaluated at a rate commensurate with changes in the environment. The goal-oriented behaviours of reactive deliberation are a useful abstraction that allow sharing of scarce computational resources and effective goal-arbitration through inter-behaviour bidding. The effectiveness of reactive deliberation has been demonstrated through a tournament of one-on-one soccer games between real-world robots. Soccer is a dynamic environment; the locations of the ball and the robots are constantly changing. The results suggest that the architectural elements in reactive deliberation are sufficient for real-time intelligent control in ...

this paper, we argue that an effective way of doing this is through the integration of a rich collection of interacting techniques, organized in a principled, structured representation. These techniques include planners and parallel transition networks (Section 3.2) to aid in overall task control, and goal-based sensing, response, and (as necessary) physicsbased, kinematic or inverse kinematic behaviors to achieve environmentally-appropriate movements. Together they simplify: ffl the expression of local environmental influences without complicating their expression at the higher levels and ffl the expression of situational awareness and influences at a higher level without the added complexity of managing all potential lower level variability, while admitting: ffl the aggregation of the intentions and expectations associated with individual tasks; ffl the interaction of multiple agents, wherein agents can sense and react to the behavior and perceived intentions of other agents, as well as to the environment; ffl the distinction between expected alternatives in the way a situation may develop and seizure of opportunity or total surprise at circumstances; ffl the independent but coordinated control of a human agent's locomotion, upper body motions, visual attention, and communication systems (speech and facial expression); ffl the embedding of persistent data structures for cognitive knowledge or spatial maps as needed to remember or mark as known visible parts or characteristics of the environment;

Many real world applications require systems with both reasoning and sensing/acting capabilities. However, most often, these systems do not work properly, i.e. they fail to execute actions and rarely perceive the external world correctly. No action, even if apparently simple, is guaranteed to succeed and, therefore, no planning can be "sound" (with respect to the real world) without taking into account failure. In this paper, we present a theory of planning that provides (1) a language that allows us to express failure; (2) a declarative formal semantics for this language; (3) a logic for reasoning about (possibly failing) plans. 1 Failure Many real world applications require systems with both reasoning and sensing/acting capabilities. Reasoning allows systems to achieve high level goals. Acting and sensing allows them to work in a complex and unpredictable external environment. Most often, systems sensing and acting in the world do not work properly, i.e. they fail to exe...

. This paper explores and reasons about the interplay between symbolic and continuous representations. We first provide some historical perspective on the signal and symbol integration as viewed by the Artificial Intelligence (AI), Robotics and Computer Vision communities. The domain of autonomous robotic agents residing in the dynamically changing environments anchors well different aspects of this integration and allows us to look at the problem in its entirety. Models of reasoning, sensing and control actions of such agents determine three different dimensions for discretization of the agent-world behavioral state space. The design and modeling of robotic agents, where these three aspects have to be closely tied together, provide a good experimental platform for addressing the signal-to-symbol-to-signal transformation problem. We present some experimental results from the domain of cooperating mobile agents involved in tasks of navigation and manipulation. 1 Introduction To motivat...

This report describes RoboSoc, a system for developing RoboCup agents designed especially, but not only, for educational use. RoboSoc is designed to be as general, open, and easy to use as possible and to encourage and simplify the modification, extension and sharing of RoboCup agents, and parts of them. To do this I assumed four requirements from the user: she wants the best possible data, use as much time as possible for the decision making, rather act on incomplete information than not act at all, and she wants to manipulate the objects found in the soccer environment. RoboSoc consists of three parts: a library of basic objects and utilities used by the rest of the system, a basic system handling the interactions with the soccer server and the timing of the agent, and a framework for world modeling and decision support. The framework defines three concepts, views, predicates and skills. The views are specialized information processing units responsible for a specific part of the wo...

This paper begins by presenting AGENDA, a simulation tool developed for the simulation and design of applications involving interacting entities. This testbed consists of two different levels, the architecture level and the system-development level. The architecture level describes a methodology for designing software agents by providing several important functionalities an agent should have. On the other hand, the system-development level provides the basic knowledge-representation formalism, general inference mechanisms, and simulation tool-box supporting visualization and monitoring of agents. Following this

Robots and software agents that have to carry out tasks in the real world often have to initiate actions that involve continuous change. But activity in the real world also involves interacting with continous external processes¯those initiated by the agent and those initiated by other factors in the world. In this paper, we present an architecture called rama, for Routine Activity Management and Analysis, which combines plan execution and process monitoring. rama uses a single process based representation to describe plans, external processes, and the way they must be coordinated. As it goes about performing tasks, rama uses a reactive executor to select actions and make predictions based on stored plans and process models. We illustrate the above behaviors with examples taken from laundry world, a simulation in which rama controls a synthetic agent that does laundry. i Contents 1 Introduction 1 2 Knowing What To Do and What To Expect 1 3 Overview of RAMA's Operation 2 4 Represent...

A variety of reactive plan execution systems have been developed in recent years, each attempting to solve the problem of taking reasonable courses of action fast enough in a dynamically changing world. Comparing these competing approaches, and collecting the best features of each, has been problematic because of the diverse representations and (sometimes implicit) control structures that they have employed. To rectify this problem, we have extended the circuit semantics notion of teleo-reactive programs into richer, yet compact semantics, called structured circuit semantics (SCS), that can be used to explicitly represent the control behavior of various reactive execution systems. By transforming existing systems into SCS, we can identify underlying control assumptions and begin to identify more rigorously the strengths and limitations of these systems. Moreover, SCS provides a basis for constructing new reactive execution systems, with more understandable semantics, that can be tailor...

High-level control systems are designed to enable mobile robots to successfully perform complex missions such as office delivery and surveillance tasks. For that purpose they have to control, coordinate, and monitor different kinds of subtasks like navigation, manipulation, and perception. An important aspect of the effectiveness of high-level control systems is the ability to cope with failures that occur during the execution of such subtasks. In this paper we focus on the particular subtask of estimating the position of the robot and show how to achieve its robust integration into the high-level control system. The principle of this integration is to monitor the certainty of the position estimation and to autonomously relocalize the robot whenever the uncertainty grows too large. We present a localization approach which accurately and efficiently keeps track of the robot's position. Furthermore, it provides a measure for detecting localization failures and it is able to autonomously relocalize the robot in such situations. In addition to this, we introduce structured reactive plans, which can be interrupted by such active localization processes at any point in time and allow the robot to complete its mission afterwards. Our method has been implemented and shown to be robust in long-term experiments involving a typical office delivery scenario.