Rasoul Mojtahedzadeh, Abdelbaki Bouguerra, Erik Schaffernicht and Achim J. Lilienthal
Probabilistic Relational Scene Representation and Decision Making Under Incomplete Information for Robotic Manipulation Tasks
In Proc. of the IEEE International Conference on Robotics and Automation (ICRA), HongKong, China., 2014
Abstract:

In this paper, we propose an approach for robotic manipulation systems to autonomously reason about their environments under incomplete information. The target application is to automate the task of unloading shipping containers. We generate a probabilistic representation to capture possible support relations between objects in partially known static configurations. We employ \textit{support vector machines} (SVM) to estimate the probability of the support relation between pairs of detected objects using features extracted from their geometrical properties and 3D sampled points of the scene. The set of probabilistic support relations is then used for reasoning about optimally selecting an object to be unloaded first. The proposed approach has been extensively tested and verified on data sets generated by simulation and real world configurations.

Keywords: Relational Scene Representation, Decision Making, Manipulation, Object Selection
Download:
Paper: PDF (1.8MB)
@ARTICLE{Mojtahedzadeh:ICRA:2014,
  AUTHOR = {Mojtahedzadeh, Rasoul and Bouguerra, Abdelbaki and Schaffernicht, Erik and Lilienthal, Achim J.},
  TITLE = {Probabilistic Relational Scene Representation and Decision Making Under Incomplete Information for Robotic Manipulation Tasks},
  BOOKTITLE = {In Proc. of the IEEE International Conference on Robotics and Automation (ICRA)},
  YEAR = {2014},
  PAGES = {5685-5690}
}