Peter Steinhaus

A cognitive architecture for a humanoid robot: a first approach

Future life pictures humans having intelligent humanoid robotic systems taking part in their everyday life. Thus researchers strive to supply robots with an adequate artificial intelligence in order to achieve a natural and intuitive interaction between human being and robotic system. Within the German Humanoid Project we focus on learning and cooperating multimodal robotic systems. In this paper we present a first cognitive architecture for our humanoid robot: The architecture is a mixture of a hierarchical three-layered form on the one hand and a composition of behaviour-specific modules on the other hand. Perception, learning, planning of actions, motor control, and human-like communication play an important role in the robotic system and are embedded step by step in our architecture

Using augmented reality to interact with an autonomous mobile platform

To allow users without special knowledge to interact with robots, it is desirable to make interaction methods as intuitive as possible. This goal is in many cases difficult to achieve since data flow from the robot to the human is limited, especially if free locomotion of both the human and the robot are required. Therefore, new communication channels need to be created. We propose the use of an augmented reality display together with a wearable, wirelessly networked computer to achieve this goal. This system makes it possible to overlay planning, world model and sensory data provided by the robot over the wearers field of view. We discuss the system architecture, interaction methods and experimental results. We demonstrate an example application for rapid prototyping of a warehouse transport system using the augmented reality system and a mobile platform. The user can create a topological map in an unknown environment on-the fly by setting and manipulating map nodes. This is done by pointing at the floor with a special interaction device, and issuing voice commands. The map is shown to the user as an augmentation of the real world view. Additionally, the robots path planning data is visualized.

Autonomous robot navigation in human-centered environments based on 3D data fusion

Efficient navigation of mobile platforms in dynamic human-centered environments is still an open research topic. We have already proposed an architecture (MEPHISTO) for a navigation system that is able to fulfill the main requirements of efficient navigation: fast and reliable sensor processing, extensive global world modeling, and distributed path planning. Our architecture uses a distributed system of sensor processing, world modeling, and path planning units. In this arcticle, we present implemented methods in the context of data fusion algorithms for 3D world modeling and real-time path planning. We also show results of the prototypic application of the system at the museum ZKM (center for art and media) in Karlsruhe.

Developing and analyzing intuitive modes for interactive object modeling

In this paper we present two approaches for intuitive interactive modelling of special object attributes by use of specific sensoric hardware. After a brief overview over the state of the art in interactive, intuitive object modeling, we motivate the modeling task by deriving the dierent object attributes that shall be modeled from an analysis of important interactions with objects. As an example domain, we chose the setting of a service robot in a kitchen. Tasks from this domain were used to derive important basic actions from which in turn the necessary object attributes were inferred. In the main section of the paper, two of the derived attributes are presented, each with an intuitive interactive modeling method. The object attributes to be modeled a restable object positions and movement restrictions for objects. Both of the intuitive interaction methods were evaluated with a group of test persons and the results are discussed. The paper ends with conclusions on the discussed results and a preview of future work in this area, in particular of potential applications.

3D global and mobile sensor data fusion for mobile platform navigation

Efficient navigation of mobile platforms in dynamic, human centered environments is still an open research topic. We have already proposed an architecture (MEPHISTO) for a navigation system that is able to fulfill the main requirements of efficient navigation: fast and reliable sensor processing, extensive global world modeling and distributed path planning. Our architecture uses a distributed system of sensor processing, world modeling and path planning units. In this paper we present some implemented methods in the context of dynamic object detection and global and mobile sensor data fusion for 3D world modeling. Experimental results of the system in the laboratory environment are presented.

Schritthaltende 3D-Kartierung und Lokalisierung für mobile Inspektionsroboter

Um technische Anlagen erfolgreich durch autonome Roboter zu inspizieren, ist eine sichere Navigation hilfreich bzw. notwendig. Eine prazise Lokalisierung des mobilen Roboters und ein aussagekraftiges 3D-Umweltmodell erleichtern diese Aufgabe. In diesem Artikel wird ein Verfahren vorgestellt, welches mit Hilfe einer neuartigen Scanneranordnung zur Erzeugung von 3D-Punktwolken, sowie eines modifizierten EKF, diese Aufgabe losen soll. Durch eine sichere Schatzung der Lage im Raum anhand von uberlappenden Teilbildern, kann ein solides Umweltmodell erstellt, und somit eine effiziente Navigation erreicht werden.

A multisensor system for observation of user actions in programming by demonstration

Good observation of a manipulation presentation performed by a human teacher is crucial to further processing steps in programming by demonstration in interactive robot programming. This paper outlines a concept of how this can be done using visual and finger measuring sensors. The input sources include: a data glove which classifies several gestures and grasps, an active stereo vision, and a fixed ceiling camera. The hardware used is presented together with the technical concepts of processing and the acquired sensor information fusion, so called elementary cognitive operators. All the sensor sources use time-efficient algorithms, since sensor data must be processed in real-time.

Aufbau und Modellierung des RoSi Scanners zur 3D-Tiefenbildakquisition

Die Akquisition von dichten 3D-Punktwolken bzw. Oberflachendaten stellt in vielen Forschungs- und Anwendungsbereichen noch ein groses Problem dar. Im Bereich autonomer mobiler Systeme stellen beispielsweise die Probleme der 3D-Kollisionsvermeidung oder des Simultaneous Localisation and Mapping (SLAM) hohe Anforderungen an die Messgeschwindigkeit, die Messwertdichten, die Verfugbarkeit aktueller Messdatensatze sowie naturlich die Genauigkeit der Einzelmessungen. Viele bereits existierende Systeme sind entweder sehr teuer, nicht mobil, langsam oder ungenau. Im Rahmen dieser Arbeit wurde aufbauend auf einem Standard Lasermesssystem (LMS) der Firma Siek ein 3D-Scanner zur Akquisition schnell und einfach triangulierbarer 3D Punktwolken entwickelt. Das Grundprinzip besteht in der rotativen Lagerung des Lasermesssystems auf einer mechanischen Achse entlang der optischen Mittelachse des Scanners. Durch Kopp lung dieser Rotationsachse an eine Standard-Motor-Getriebe-Kombination kann eine permanente Rotation der Scanebene (RoSi = Rotating Sick) und damit eine zyklische Abtastung der Umgebung erreicht werden.

A robot navigation approach based on 3D data fusion and real time path planning

Efficient navigation of mobile platforms in dynamic, human centered environments is still an open research topic. We have already proposed an architecture (MEPHISTO) for a navigation system that is able to fulfill the main requirements of efficient navigation: fast and reliable sensor processing, extensive global world modeling and distributed path planning. Our architecture uses a distributed system of sensor processing, world modeling and path planning units. In this paper, we present implemented methods in the context of data fusion algorithms for 3D (three-dimensional) world modelling, using the models for real time path planning and platform control strategies. We also show results of the prototypic application of the system at the museum ZKM (Center for Arts and Media) in Karlsruhe.

Sonderforschungsbereich 588: Humanoide Roboter – Lernende und kooperierende multimodale Roboter (Humanoid Robots – Learning and Cooperating Multimodal Robots)

Zusammenfassung Ziel des Sonderforschungsbereiches 588 “Humanoide Roboter” ist es, Konzepte, Methoden und Komponenten für einen humanoiden Roboter zu entwickeln, der seinen Arbeitsbereich mit dem Menschen teilt. Die Schwerpunkte unserer Forschungstätigkeit liegen dabei in folgenden Bereichen: Entwicklung mechatronischer Komponenten und Konstruktion eines Demonstratorsystems, Perzeption von Benutzer und Umgebung, Modellierung und Simulation von Roboter, Umgebung und Benutzer, sowie Mensch-Roboter-Kooperation und Lernen auf unterschiedlichen Verhaltensebenen. Durch das in diesem Sonderforschungsbereich entwickelte “teilanthropomorphe Robotersystem” soll der Schritt zu kooperierenden, multimodalen und lernenden Robotersystemen vollzogen werden.