Force- and Vision-based Detection of Contact State Transitions

  • A new and systematic basic approach to force- and vision-based robot manipulation of deformable (non-rigid) linear objects is introduced. This approach reduces the computational needs by using a simple state-oriented model of the objects. These states describe the relation between the deformable and rigid obstacles, and are derived from the object image and its features. We give an enumeration of possible contact states and discuss the main characteristics of each state. We investigate the performance of robust transitions between the contact states and derive criteria and conditions for each of the states and for two sensor systems, i.e. a vision sensor and a force/torque sensor. This results in a new and task-independent approach in regarding the handling of deformable objects and in a sensor-based implementation of manipulation primitives for industrial robots. Thus, the usage of sensor processing is an appropriate solution for our problem. Finally, we apply the concept of contact states and state transitions to the description of a typical assembly task. Experimental results show the feasibility of our approach: A robot performs several contact state transitions which can be combined for solving a more complex task.

Export metadata

  • Export Bibtex
  • Export RIS

Additional Services

Share in Twitter Search Google Scholar
Metadaten
Author:Frank Abegg, Axel Remde, Dominik Henrich
URN (permanent link):urn:nbn:de:hbz:386-kluedo-9953
Document Type:Preprint
Language of publication:English
Year of Completion:2000
Year of Publication:2000
Publishing Institute:Technische Universität Kaiserslautern
Tag:AG-RESY ; Deformable Objects ; Force-Torque ; HANDFLEX ; Manipulation ; Robotics ; Vision
Faculties / Organisational entities:Fachbereich Informatik
DDC-Cassification:004 Datenverarbeitung; Informatik
Collections:AG RESY

$Rev: 12793 $