Organizational Unit:
Humanoid Robotics Laboratory

Permanent Link
https://hdl.handle.net/1853/70718
Research Organization Registry ID
Description
Previous Names
Parent Organization
Parent Organization
Includes Organization(s)
ArchiveSpace Name Record

Filters

2011 - 2012 4
4
2 2 1
4
4
Reset filters

Settings
Author: Levihn, Martin × End date: 2012 × Start date: 2011 ×

Publication Search Results

Now showing 1 - 4 of 4
  • Thumbnail Image
    Item
    Multi-Robot Multi-Object Rearrangement in Assignment Space
    (Georgia Institute of Technology, 2012-10) Levihn, Martin ; Igarashi, Takeo ; Stilman, Mike
    We present Assignment Space Planning, a new efficient robot multi-agent coordination algorithm for the PSPACE- hard problem of multi-robot multi-object push rearrangement. In both simulated and real robot experiments, we demonstrate that our method produces optimal solutions for simple problems and exhibits novel emergent behaviors for complex scenarios. Assignment Space takes advantage of the domain structure by splitting the planning up into three stages, effectively reducing the search space size and enabling the planner to produce optimized plans in seconds. Our algorithm finds solutions of comparable quality to complete configuration space search while reducing the computing time to seconds, which allows our approach to be applied in practical scenarios in real-time.
  • Thumbnail Image
    Item
    Hierarchical Decision Theoretic Planning for Navigation Among Movable Obstacles
    (Georgia Institute of Technology, 2012-06) Levihn, Martin ; Scholz, Jonathan ; Stilman, Mike
    In this paper we present the first decision theoretic planner for the problem of Navigation Among Movable Obstacles (NAMO). While efficient planners for NAMO exist, they are challenging to implement in practice due to the inherent uncertainty in both perception and control of real robots. Generalizing existing NAMO planners to nondeterministic domains is particularly difficult due to the sensitivity of MDP methods to task dimensionality. Our work addresses this challenge by combining ideas from Hierarchical Reinforcement Learning with Monte Carlo Tree Search, and results in an algorithm that can be used for fast online planning in uncertain environments. We evaluate our algorithm in simulation, and provide a theoretical argument for our results which suggest linear time complexity in the number of obstacles for typical environments.
  • Thumbnail Image
    Item
    Detecting Partially Occluded Objects via Segmentation and Validation
    (Georgia Institute of Technology, 2012) Levihn, Martin ; Dutton, Matthew ; Trevor, Alexander J. B. ; Stilman, Mike
    This paper presents a novel algorithm: Verfied Partial Object Detector (VPOD) for accurate detection of partially occluded objects such as furniture in 3D point clouds. VPOD is implemented and validated on real sensor data obtained by our robot. It extends Viewpoint Feature Histograms (VFH) which classify unoccluded objects to also classifying partially occluded objects such as furniture that might be seen in typical office environments. To achieve this result, VPOD employs two strategies. First, object models are segmented and the object database is extended to include partial models. Second, once a matching partial object is detected, the full object model is aligned back into the scene and verified for consistency with the point cloud data. Overall, our approach increases the number of objects found and substantially reduces false positives due to the verification process.
  • Thumbnail Image
    Item
    Efficient Opening Detection
    (Georgia Institute of Technology, 2011) Levihn, Martin ; Stilman, Mike
    We present an efficient and powerful algorithm for detecting openings. Openings indicate the existence of a new path for the robot. The reliable detection of new openings is especially relevant to the domain of Navigation Among Movable Obstacles in known [7] as well as unknown [2] environments. Tremendous speed-ups for algorithms in these domains can be achieved by limiting the considerations of obstacle manipulations to cases where manipulations create new openings. The presented algorithm can detect openings for obstacles of arbitrary shapes being displaced in arbitrary directions in changing environments. To the knowledge of the authors, this is the first algorithm to achieve efficient opening detection for arbitrary shaped obstacles.