Georg von Wichert

A Probabilistic Relational Model for Characterizing Situations in Dynamic Multi-Agent Systems

Artificial systems with a high degree of autonomy require reliable semantic information about the context they operate in. State interpretation, however, is a difficult task. Interpretations may depend on a history of states and there may be more than one valid interpretation. We propose a model for spatio-temporal situations using hidden Markov models based on relational state descriptions, which are extracted from the estimated state of an underlying dynamic system. Our model covers concurrent situations, scenarios with multiple agents, and situations of varying durations. To evaluate the practical usefulness of our model, we apply it to the concrete task of online traffic analysis.

Extracting semantic indoor maps from occupancy grids

Abstract The primary challenge for any autonomous system operating in realistic, rather unconstrained scenarios is to manage the complexity and uncertainty of the real world. While it is unclear how exactly humans and other higher animals master these problems, it seems evident, that abstraction plays an important role. The use of abstract concepts allows us to define the system behavior on higher levels. In this paper we focus on the semantic mapping of indoor environments. We propose a method to extract an abstracted floor plan from typical grid maps using Bayesian reasoning. The result of this procedure is a probabilistic generative model of the environment defined over abstract concepts. It is well suited for higher-level reasoning and communication purposes. We demonstrate the effectiveness of the approach using real-world data.

Can robots learn to see

Abstract Since the late 1960s autonomous robots have been the subject of worldwide research efforts. Various techniques exist that enable mobile robots to navigate robustly within their environments. Some systems are commercially available, most of them for transport and floor-cleaning applications. However, in general they are not truly autonomous, as they require human aid to build appropriate environment models, e.g. navigation maps, which they need for planning. But even high-end research robots normally need help to configure and adjust their sensors, e.g. vision systems where the user has to tune lots of parameters before the robot can ‘see’ in the given environment. Real service robots will have to to do this autonomously, as no helping scientist will be available. This paper presents a first step into this direction. It shows how a useful, self-learning vision system can be constructed, and that such a system is able to supply the robot with the information required to ‘survive’ in complex everyday environments.

MobMan - Ein mobiler Manipulator für Alltagsumgebungen

Der Einsatz von mobilen Servicerobotern beschrankt sich bisher auf Transport- und Reinigungsaufgaben, bei welcher Mobilitat und Navigation in einer 2-dimensionalen Welt die zentrale Rolle spielen. Weitere Einsatzgebiete (z.B. im Haushalt) erfordern die Fahigkeit, Gegenstande manipulieren zu konnen. Inhalt dieses Beitrages ist die Beschreibung des bei der Siemens AG hierfur entwickelten Forschungsprototypen fur Mobilitat und Manipulation (MobMan) in Alltagsumgebungen. Der Schwerpunkt liegt auf einem Ansatz zur Steuerung von Manipulationsskills in komplexen Alltagsumgebungen.

A generalizable knowledge framework for semantic indoor mapping based on Markov logic networks and data driven MCMC

Abstract In this paper, we propose a generalizable knowledge framework for data abstraction, i.e., finding a compact abstract model for input data using predefined abstract terms. Based on these abstract terms, intelligent autonomous systems, such as a robot, should be able to make inferences according to a specific knowledge base, so that they can better handle the complexity and uncertainty of the real world. We propose to realize this framework by combining Markov logic networks (MLNs) and data driven MCMC sampling, because the former are a powerful tool for modeling uncertain knowledge and the latter provides an efficient way to draw samples from unknown complex distributions. Furthermore, we show in detail how to adapt this framework to a certain task, in particular, semantic robot mapping. Based on MLNs, we formulate task-specific context knowledge as descriptive soft rules. Experiments on real world data and simulated data confirm the usefulness of our framework.

Probabilistic multi-sensor fusion based on signed distance functions

In this paper, we present an approach for the probabilistic fusion of 3D sensor measurements. Our fusion algorithm is based on truncated signed distance functions. It explicitly considers the measurement noise by modeling the surface using random variables. Furthermore, our proposed surface model provides an explicit estimation of the spatial uncertainty. The approach can be implemented on a GPU to achieve a high update performance and enable online updates of the model. The approach was evaluated in simulation and using real sensor data. In our experiments, we confirmed that it accurately estimates surfaces from noisy sensor data and that it provides a corresponding estimate of the uncertainty. We could also show that the approach is able to fuse measurements from sensors with different noise characteristics.

A Gaussian measurement model for local interest point based 6 DOF pose estimation

One of the main challenges for service robots during operation lies in the handling of unavoidable uncertainties which originate from model and sensor inaccuracies and which are characteristic for realistic application scenarios. Robustness under real world conditions can only be achieved when the dominant uncertainties are explicitly represented and purposefully managed by the robots control system. We therefore adopt a probabilistic approach in which perception is regarded as a sequential estimation process and follow a Bayesian filtering methodology. Under these assumptions probabilistic models of the robots perception systems are key. In this paper we shortly describe a model based object recognition and localization system. However, we do not not focus on the 6D pose estimation procedure itself, but on the method to quantify and compute the uncertainty associated with it. We construct a Gaussian approximation of the resulting pose error using the implicit function theorem. It is then used as a proposal density for importance sampling. Our goal is to sample from the measurement model describing 6D object localization based on local features in a Bayesian filtering context.

Applying rule-based context knowledge to build abstract semantic maps of indoor environments

In this paper, we propose a generalizable method that systematically combines data driven MCMC sampling and inference using rule-based context knowledge for data abstraction. In particular, we demonstrate the usefulness of our method in the scenario of building abstract semantic maps for indoor environments. The product of our system is a parametric abstract model of the perceived environment that not only accurately represents the geometry of the environment but also provides valuable abstract information which benefits highlevel robotic applications. Based on predefined abstract terms, such as “type” and “relation”, we define task-specific context knowledge as descriptive rules in Markov Logic Networks. The corresponding inference results are used to construct a prior distribution that aims to add reasonable constraints to the solution space of semantic maps. In addition, by applying a semantically annotated sensor model, we explicitly use context information to interpret the sensor data. Experiments on real world data show promising results and thus confirm the usefulness of our system.

Grasping on the move: A generic arm-base coordinated grasping pipeline for mobile manipulation

In robotics, grasping is often considered as an isolated topic which is independent of the base position of the operating robot. People place their robots at an empirically predefined position and assume that the desired object is perceivable to the robot and reachable for the robot arm. In case of inappropriate base positions, however, the desired object may become hidden or unreachable, so that the robot can not perform grasping at all. In this paper, we propose a generic grasping pipeline, which achieves arm-base coordinated grasping given the information about object pose and the operating environment. In addition, we propose a method to find the optimal base position for arm-base coordinated grasping. Moreover, we adopt a modular design to realize our grasping pipeline so that it stays as generic as possible. Experiments in simulation and on a real robot show promising results and thus confirm the usefulness of our approach.

Online semantic exploration of indoor maps

In this paper we propose a method to extract an abstracted floor plan from typical grid maps using Bayesian reasoning. The result of this procedure is a probabilistic generative model of the environment defined over abstract concepts. It is well suited for higher-level reasoning and communication purposes. We demonstrate the effectiveness of the approach through real-world experiments.