This paper describes the general concept of Route Graphs, to be used for navigation by various agents in a variety of scenarios. We introduce the concept of an ontology and describe the modelling of general graphs as an example. This approach is then applied to define a “light-weight ” ontology of Route Graphs in an indoors environment, giving at first just a taxonomy of (sub)classes and relations between them, as well as to other (spatial) ontologies. Finally, we show how to formalise ontologies using a First Order Logic approach, and give an outline of how to develop actual data structures and algorithms for Route Graphs.

Establishing a clean relationship between a robot’s spatial model and natural language components is a non-trivial task, but is key to designing verbally controlled, navigating service robots. In this paper we examine the issues involved in the development of dialogue controlled navigating robots. In particular, we treat our robots as so-called Shared Control Systems, where robot and user cooperate to achieve a shared goal. We begin by characterising four categories of Shared Control Problems that affect verbally controlled navigating robots. Producing solutions to these problems requires a clear methodology in the linking of ’common-sense ’ representations of space used by the robots, and the language interface. To this end, we present the SharC Cognitive Control Architecture as a general purpose, agent-based dialogue control system that provides a suitable framework for relating spatial information to natural language communication. To illustrate our approach, we focus in particular on natural language understanding, and show how natural language utterances may be mapped to formally modelled spatial concepts, thus helping to overcome problems in shared control.

In this paper we apply a formal ontological framework in order to deconstruct two prominent approaches to navigation from cognitive robotics, the Spatial Semantic Hierarchy of Kuipers and the Route Graph of Krieg-Brückner, Werner and others. The ontological framework is based on our current work on ontology specification, where we are investigating Masolo et al.’s Descriptive Ontology for Linguistic and Cognitive Engineering (DOLCE) extended particularly for space and navigation by incorporating aspects of Smith et al.’s Basic Formal Ontology (BFO). Our conclusion is that ontology should necessarily play an important role in the design and modelling of cognitive robotic systems: comparability between approaches is improved, modelling gaps and weaknesses are highlighted, re-use of existing formalisations is facilitated, and extensions for interaction with other components, such as natural language systems, are directly supported.

Abstract. In this paper we report results of an experiment that investigates the effects of mobile pedestrian navigation systems on the development of route and survey knowledge acquired by the users. In the experiment directions were presented incrementally step-by-step in different modalities (i.e. audio, graphics) and through different media (PDA, clip-on display). The experiment has been carried out in the field in a Wizard-of-Oz like study. Results show that as expected all subjects had problems in building up survey knowledge of the environment. In contrast, route knowledge was learned much better. We also observed a slight gender effect showing that women had an advantage of a visual presentation condition, whereas for men the presentation mode didn’t matter. Finally, we discuss some implications on the design of pedestrian navigation systems. 1

The creation of a robot capable of navigating in unknown urban environments without the use of GPS data or prior map knowledge is envisioned in the Autonomous City Explorer (ACE) project. The robot has to retrieve direction information solely by interacting with humans. This work presents a human-robot communication system that enables the robot to ask for directions and store the retrieved route information as internal knowledge. The system incorporates theories from linguistics in a mixed-modalities communication interface. It stores acquired information into a topological route graph which is used to give feedback to the human and to navigate in unknown environments.

In shared-control systems, such as intelligent service robots, a human operator and an automated technical system are interdependently in charge of control. Effective shared control requires complex system architectures that provide safety and robustness in operation, while providing a user-friendly interface – provision of these dual requirements is a non-trivial task. In this paper, we report on our approach to addressing these issues. We present a dialogue centric cognitive control architecture, that utilises both agent-oriented programming and formal methods. Our SharC Cognitive Control Architecture is a hybrid design that distributes control of a robot amongst a community of deliberative intentional agents. Since safety is of paramount importance in shared-control systems, high-level safety issues must be addressed within such architectures. To this end, we also describe a formally modelled dialogue manager that sits at the heart of the control system. We illustrate the use of these distinct software paradigms with examples from our demonstration platform: Rolland, the Bremen autonomous wheelchair. 1

In this paper we regard the navigation aid provided by mobile navigation systems in a real environment and the effects of these mobile assistants to the development of spatial knowledge. Therefore, we report on a user study concerning the acquisition of spatial knowledge. This study sets up on a former study described by Krüger and colleagues and sheds light on problems concerning the acquisition of survey knowledge while being navigated by a mobile handheld PC.

In previous years we have been involved in several projects in which users (or visitors) had to find their way in informationrich virtual environments. 'Information-rich' means that the users do not know beforehand what is available in the environment, where to go in the environment to find the information and, moreover, users or visitors do not necessarily know exactly what they are looking for. Information-rich means also that the information may change during time. A second visit to the same environment will require different behavior of the visitor in order for him or her to obtain similar information than was available during a previous visit. In this paper we report about two projects and discuss our attempts to generalize from the different approaches and application domains to obtain a library of methods and tools to design and implement intelligent agents that inhabit virtual environments and where the agents support the navigation of the user/visitor.

In this paper we propose a spatial representation approach for a mobile robot operating in an office-like indoor environment which is intended to provide an interface between lowlevel information required for navigation and abstract information required for high-level symbolic reasoning about routes. The representation is based on a route graph [16] that links navigational decision points via edges corresponding to route segments. We describe a particular route graph representation that is derived from the generalized Voronoi diagram of the environment and enables the robot to incrementally construct the representation autonomously. Since the Voronoi-based route graph still reflects irrelevant features of the environment, our proposed representation is a hierarchical structure consisting of route graph layers representing the environment at different levels of granularity. It is shown how the more abstract layers can be derived from the original route graph by using relevance measures to assess the significance of the vertices. We provide examples of how planning, spatial reasoning, and communication can benefit from this kind of representation.

... aims to create a robot capable of navigating unknown urban environments without the use of GPS data or prior map knowledge. The robot finds its way by interacting with pedestrians and building a topological representation of its surroundings. This paper outlines the necessary ingredients for successful low-level navigation on sidewalks, information retrieval from pedestrians as well as the construction of a semantic representation of an urban environment. A system architecture for outdoor localization, traversability assessment, path planning, behavior selection and topological abstraction in urban environments is presented. The efficiency of the proposed approach is verified by a number of outdoor experiments with the ACE robot.