The Autonomous City Explorer (ACE) project combines research from autonomous outdoor navigation and human-robot interaction. The ACE robot is capable of navigating unknown urban environments without the use of GPS data or prior map knowledge. It finds its way by interacting with pedestrians in a natural and intuitive way and building a topological representation of its surroundings. In a recent experiment the robot managed to successfully travel a 1.5 km distance from the campus of the Technische Universität München to Marienplatz, the central square of Munich.This article describes the principles and system components for navigation in urban environments, information retrieval through natural human-robot interaction, the construction of a suitable semantic representation as well as results from the field experiment.

One of the greatest challenges nowadays in robotics is the advancement of robots from industrial tools to companions and helpers of humans, operating in natural, populated environments. In this respect, the Autonomous City Explorer (ACE) project aims to combine the research fields of autonomous mobile robot navigation and human robot interaction. A robot has been created that is capable of navigating in an unknown, highly populated, urban environment, based only on information extracted through interaction with passers-by and its local perception capabilities. This paper describes the algorithms and architecture that make up the navigation subsystem of ACE. More specifically, the algorithms used for Simultaneous Localization and Mapping (SLAM), path planning in dynamic environments and behavior selection are presented, as well as the system architecture that integrates them to a complete working system. Results from an extended field experiment, where the robot navigated autonomously through the downtown city area of Munich, are analyzed and show that the robot is capable of long-term, safe navigation in real-world settings.

No abstract

In order to be truly autonomous, robots that operate in natural, populated environments must have the ability to create a model of these unpredictable dynamic environments and make use of this self-acquired uncertain knowledge to decide about their actions. A formal Bayesian framework is introduced, which enables recursive estimation of a dynamic environment model and action selection based on this estimate. Existing methods are combined to produce a working implementation of the proposed framework. A RaoBlackwellized particle filter (RBPF) is deployed to address the Simultaneous Localization And Mapping (SLAM) problem and combined with recursive conditional particle filters in order to track people in the vicinity of the robot. In this way, a complete model is provided, which is utilized for selecting the actions of the robot so that its uncertainty is kept under control and the likelihood of achieving its goals is increased. All developed algorithms have been applied to the domain of the Autonomous City Explorer robot and results from the implementation on the robotic platform are presented.

Autonomous mobile robots are deployed in a variety of application domains, resulting in scenario specific implementations. However these systems share common components responsible for perception, path planning and task execution. In order to find a formal way to identify the influence of the environmental complexity to the used methods, an approach for quantitative system interdependence analysis is introduced. The coherence between several performance indicators of different system components, as well as the influence of environmental parameters on the system, are learned and quantitatively evaluated. Performance evaluation of an autonomous robot navigating in two different urban environments is conducted and presented results demonstrate the applicability of the proposed approach.