Lutz Frommberger

Qualitative spatial representation and reasoning in the SparQ-toolbox

A multitude of calculi for qualitative spatial reasoning (QSR) have been proposed during the last two decades. The number of practical applications that make use of QSR techniques is, however, comparatively small. One reason for this may be seen in the difficulty for people from outside the field to incorporate the required reasoning techniques into their software. Sometimes, proposed calculi are only partially specified and implementations are rarely available. With the SparQ toolbox presented in this text, we seek to improve this situation by making common calculi and standard reasoning techniques accessible in a way that allows for easy integration into applications. We hope to turn this into a community effort and encourage researchers to incorporate their calculi into SparQ. This text is intended to present SparQ to potential users and contributors and to provide an overview on its features and utilization.

Mobile4D: crowdsourced disaster alerting and reporting

Small and large-scale disasters are a major factor for poverty. When information is sent out at an early stage and directly to people affected, impact on environment, people, livestock, crop, and belongings can be minimized. We present Mobile4D, an integrated mobile crowdsourcing-based disaster alerting and reporting system tested in Lao PDR. With Mobile4D it is possible to gather information from affected people, to establish direct communication channels between affected people and administrative units, and to rapidly distribute information to regions and people struck by disasters.

LEARNING TO BEHAVE IN SPACE: A QUALITATIVE SPATIAL REPRESENTATION FOR ROBOT NAVIGATION WITH REINFORCEMENT LEARNING*

The representation of the surrounding world plays an important role in robot navigation, especially when reinforcement learning is applied. This work uses a qualitative abstraction mechanism to create a representation of space consisting of the circular order of detected landmarks and the relative position of walls towards the agents moving direction. The use of this representation does not only empower the agent to learn a certain goal-directed navigation strategy faster compared to metrical representations, but also facilitates reusing structural knowledge of the world at different locations within the same environment. Acquired policies are also applicable in scenarios with different metrics and corridor angles. Furthermore, gained structural knowledge can be separated, leading to a generally sensible navigation behavior that can be transferred to environments lacking landmark information and/or totally unknown environments.

Generating adaptive route instructions using hierarchical reinforcement learning

We present a learning approach for efficiently inducing adaptive behaviour of route instructions. For such a purpose we propose a two-stage approach to learn a hierarchy of wayfinding strategies using hierarchical reinforcement learning. Whilst the first stage learns low-level behaviour, the second stage focuses on learning high-level behaviour. In our proposed approach, only the latter is to be applied at runtime in user-machine interactions. Our experiments are based on an indoor navigation scenario for a building that is complex to navigate. We compared our approach with flat reinforcement learning and a fully-learnt hierarchical approach. Our experimental results show that our proposed approach learns significantly faster than the baseline approaches. In addition, the learnt behaviour shows to adapt to the type of user and structure of the spatial environment. This approach is attractive to automatic route giving since it combines fast learning with adaptive behaviour.

Machine learning for interactive systems and robots: a brief introduction

Research on interactive systems and robots, i.e. interactive machines that perceive, act and communicate, has applied a multitude of different machine learning frameworks in recent years, many of which are based on a form of reinforcement learning (RL). In this paper, we will provide a brief introduction to the application of machine learning techniques in interactive learning systems. We identify several dimensions along which interactive learning systems can be analyzed. We argue that while many applications of interactive machines seem different at first sight, sufficient commonalities exist in terms of the challenges faced. By identifying these commonalities between (learning) approaches, and by taking interdisciplinary approaches towards the challenges, we anticipate more effective design and development of sophisticated machines that perceive, act and communicate in complex, dynamic and uncertain environments.

Micro-mapping with smartphones for monitoring agricultural development

Monitoring and evaluating progress and impact of development projects is a critical aspect. In this paper, we show how we can use a smartphone based system to intuitively retrieve the exact geometry of smaller objects, making it suitable to assess agricultural entities like fields or ponds with their exact extent and location. The systems simple and intuitive workflow can be used by laymen and thus allows for crowdsourcing geo-data on a local level.

Structural knowledge transfer by spatial abstraction for reinforcement learning agents

In this article we investigate the role of abstraction principles for knowledge transfer in agent control learning tasks. We analyze abstraction from a formal point of view and characterize three distinct facets: aspectualization, coarsening, and conceptual classification. The taxonomy we develop allows us to interrelate existing approaches to abstraction, leading to a code of practice for designing knowledge representations that support knowledge transfer. We detail how aspectualization can be utilized to achieve knowledge transfer in reinforcement learning. We propose the use of so-called structure space aspectualizable knowledge representations that explicate structural properties of the state space and present a posteriori structure space aspectualization (APSST) as a method to extract generally sensible behavior from a learned policy. This new policy can be used for knowledge transfer to support learning new tasks in different environments. Finally, we present a case study that demonstrates transfer of generally sensible navigation skills from simple simulation to a real-world robotic platform.

Lowering the barrier: how the what-you-see-is-what-you-map paradigm enables people to contribute volunteered geographic information

In developing countries, Volunteered Geographic Information (VGI) can be a valuable source of information, as often no detailed geo-data is available. Furthermore, it enables to aggregate local knowledge that can only be provided by local stakeholders. However, due to the complexity of tools and workflows of VGI applications, the participation within VGI projects is typically limited to technically skilled persons. We propose to focus on task-specific interfaces to integrate inexperienced user groups in the data collection process. For the example of micro-mapping we show how What-You-See-Is-What-You-Map (WYSIWYM) interfaces can lower the technological barriers compared to generic user interfaces and allow the contribution of precise geometric geographic data even for technologically uneducated persons. We report on two user studies in two different settings: the first study compares the general quantitative and qualitative aspects of user-generated data of three different geographic data collection tools, and the second study investigates the robustness of these results with a focus on the technological barrier with technologically untrained farmers in rural Laos. We are able to show that WYSIWYM interfaces foster the contribution of data of promising quality and at the same time significantly lower the barrier of usage. Thus, WYSIWYM is well suited to integrate contributors with limited technological knowledge into VGI processes and enables crowdsourcing geographic information at a local level.

Representing and Selecting Landmarks in Autonomous Learning of Robot Navigation

Navigation based on detected landmarks is an important facet of robot navigation. This work investigates into a qualitative representation of landmarks for an autonomous learning task where a robot learns a goal directed navigation strategy with reinforcement learning. We discuss how to build a suitable landmark-based representation. In particular, we focus on selection of landmarks to regard when experiencing a multitude of landmarks, because representing all of them would blow up the state space inappropriately. Thus, we examine strategies for this selection. Furthermore, we introduce a background knowledge based structure-aware landmark selection mechanism to limit landmark observation to the cases where it is really needed.

Situation awareness in crowdsensing for disease surveillance in crisis situations

Crowdsensing can provide real time and detailed information about rapidly evolving crisis situations to facilitate rapid response and effective resource allocation. But while challenges such as heterogeneity of data content and quality, asynchronicity, and volume call for robust data integration and interpretation capabilities, situation awareness in crowdsensing for crisis management remains a largely unexplored area of research. In this paper we extend the mobile4D smartphone-based disaster reporting and alerting system with a situation awareness data interpretation and integration layer and demonstrate its application to the problem of tracking cholera outbreaks. The communication workflow in mobile4D-SA supports interaction between crowdsensed information, system predictions, and multifaceted communication between authorities and affected people on the ground.