Scene Analysis for Service Robots

  title={Scene Analysis for Service Robots},
  author={Robert Eidenberger and Thilo Grundmann and Martin Schneider and Wendelin Feiten and Michael Fiegert and Georg von Wichert and Gisbert Lawitzky},
  booktitle={Towards Service Robots for Everyday Environments},
A scene analysis module for service robots is presented which uses SIFT in a stereo setting, a systematic handling of uncertainties and an active perception component. The system is integrated and evaluated on the DESIRE two-arm mobile robot. Complex everyday scenes composed of various items from a 100-object database are analyzed successfully and efficiently. 
Probabilistic Cognition for Autonomous Systems: Abstraction, Semantics and Knowledge
The effectiveness of the proposed KSMCMC sampling technique is demonstrated in two typical tasks in the robotic domain: semantic mapping and scene analysis.
Active scene recognition for programming by demonstration using next-best-view estimates from hierarchical Implicit Shape Models
We present an approach that combines passive scene understanding with object search in order to recognize scenes in indoor environments that cannot be perceived from a single point of view. Passive
Active Scene Recognition
  • Pascal Meissner
  • Computer Science, Psychology
    Springer Tracts in Advanced Robotics
  • 2019
Detailed technical presentation of our contributions that are related to Active Scene Recognition. This includes our approaches to Object Pose Prediction and Next-Best-View estimation.
Improving object orientation estimates by considering multiple viewpoints
A histogram filter variant that uses the exploration capabilities of robots, and supports active perception through a next-best-view proposal algorithm, and a histogram-based fusion method that focuses on the orientation of the 6 degrees of freedom (DoF) pose.
Bayesian Orientation Estimation and Local Surface Informativeness for Active Object Pose Estimation
The accuracy of the orientation estimation using the proposed method is not yet comparable to state-of-the-art algorithms in the general case of unrestricted viewing directions, but the evaluation of view and surface part informativeness gives plausible and promising results for building effective view planning criteria.
Related Work


Robust 6d Pose Determination in Complex Environments for One Hundred Classes
For many robotic applications including service robotics robust object classification and 6d object pose determination are of substantial importance. This paper presents an object recognition
Probabilistic occlusion estimation in cluttered environments for active perception planning
A probabilistic framework for scene modeling and active perception planning in complex environments and concepts from computer graphics and statistics are combined for the efficient and precise estimation of future observations.
Active perception and scene modeling by planning with probabilistic 6D object poses
This paper presents an approach to probabilistic active perception planning for scene modeling in cluttered and realistic environments. When dealing with complex, multi-object scenes with arbitrary
Dynamic sensor planning with stereo for model identification on a mobile platform
  • Jeremy Ma, J. Burdick
  • Computer Science
    2010 IEEE International Conference on Robotics and Automation
  • 2010
An approach to sensor planning for simultaneous pose estimation and model identification of a moving object using a stereo camera sensor mounted on a mobile base and a two-wheeled nonholonomic robot is presented.
6D Pose Uncertainty in Robotic Perception
This work develops a robotic perception system with the aim of perceiving and modeling an unprepared kitchen scenario with many objects that requires the representation of wide spread probability distributions of the 6D pose.
Active Vision for Complete Scene Reconstruction and Exploration
This method is based on the structure from controlled motion approach that constrains camera motions to obtain an optimal estimation of the 3D structure of a geometrical primitive and develops perceptual strategies able to perform a succession of robust estimations.
Active recognition through next view planning: a survey
Stereo-based 6D object localization for grasping with humanoid robot systems
This work will present a system that makes explicit use of the stereo camera system in order to achieve maximum depth accuracy and can deal with textured objects as well as objects that can be segmented globally and are defined by their shape.
Probabilistic action planning for active scene modeling in continuous high-dimensional domains
In active perception systems for scene recognition the utility of an observation is determined by the information gain in the probability distribution over the state space, which focuses on optimizing the determination of the payoff neglecting the costs.
The robotic bar - an integrated demonstration of man-robot interaction in a service scenario
This paper describes the approach to construct a robotic assistance system and presents experience with an integrated technology demonstration and exposure of the integrated system to the public.