Robotic Grasping of Large Objects for Collaborative Manipulation
Loading...
Journal Title
Journal ISSN
Volume Title
Sähkötekniikan korkeakoulu |
Master's thesis
Unless otherwise stated, all rights belong to the author. You may download, display and print this publication for Your own personal use. Commercial use is prohibited.
Author
Date
2017-10-23
Department
Major/Subject
Space Robotics and Automation
Mcode
ELEC3047
Degree programme
Erasmus Mundus Space Master
Language
en
Pages
61
Series
Abstract
In near future, robots are envisioned to work alongside humans in professional and domestic environments without significant restructuring of workspace. Robotic systems in such setups must be adept at observation, analysis and rational decision making. To coexist in an environment, humans and robots will need to interact and cooperate for multiple tasks. A fundamental such task is the manipulation of large objects in work environments which requires cooperation between multiple manipulating agents for load sharing. Collaborative manipulation has been studied in the literature with the focus on multi-agent planning and control strategies. However, for a collaborative manipulation task, grasp planning also plays a pivotal role in cooperation and task completion. In this work, a novel approach is proposed for collaborative grasping and manipulation of large unknown objects. The manipulation task was defined as a sequence of poses and expected external wrench acting on the target object. In a two-agent manipulation task, the proposed approach selects a grasp for the second agent after observing the grasp location of the first agent. The solution is computed in a way that it minimizes the grasp wrenches by load sharing between both agents. To verify the proposed methodology, an online system for human-robot manipulation of unknown objects was developed. The system utilized depth information from a fixed Kinect sensor for perception and decision making for a human-robot collaborative lift-up. Experiments with multiple objects substantiated that the proposed method results in an optimal load sharing despite limited information and partial observability.Description
Supervisor
Kyrki, VilleThesis advisor
Muthusamy, RajkumarKeywords
grasp planning, multi-agent grasping, collaborative manipulation, load sharing
