Universität Stuttgart

Permanent URI for this communityhttps://elib.uni-stuttgart.de/handle/11682/1

Browse

Filters

2014 - 20162

Settings

Start date: 2000End date: 2019Subject: search.filters.subject.004Author: search.filters.author.Zielke, Viktor

Search Results

Now showing 1 - 2 of 2
  • Thumbnail Image
    ItemOpen Access
    Instance-based learning of affordances
    (2014) Zielke, Viktor
    The discovery of possible interactions with objects is a vital part of an exploration task for robots. An important subset of these possible interactions are affordances. Affordances describe what a specific object can afford to a specific agent, based on the capabilities of the agent and the properties of the object in relation to the agent. For example, a chair affords a human to be sat-upon, if the sitting area of the chair is approximately knee-high. In this work, an instance-based learning approach is made to discover these affordances solely through different visual representations of point cloud data of an object. The point clouds are acquired with a Microsoft Kinect sensor. Different representations are tested and evaluated against a set of point cloud data of various objects found in a living room environment.
  • Thumbnail Image
    ItemOpen Access
    Design and implementation of next-best-view algorithms for automatic robotic-based (dis)assembly tasks
    (2016) Zielke, Viktor
    Robots tasked with the autonomous interaction of objects, such as assembly and disassembly tasks, in a dynamic environment require the ability to explore their environment and detect objects for interactions. State-of-the-art methods exist which can handle these tasks separately. This work describes a method for combining both tasks and therefor reduce the amount of costly operations like motion and sensing. A next-best-view system is developed which incrementally builds a map of the environment and enables the selection of view poses for an eye-in-hand robot system. The system and the performance of the selected view poses is evaluated on a robotic system. The evaluations showed that the method selected view poses which explored the environment and detected objects.