NAO greift - Explorieren und Greifen unbekannter Objekte auf einem performanzlimitierten humanoiden Roboter mit Hilfe vorberechneter Tabellen
File | Description | Size | Format | |
---|---|---|---|---|
00105055-1.pdf | 27.69 MB | Adobe PDF | View/Open |
Other Titles: | NAO is Grasping - Exploring and Grasping Unknown Objects on a Performance-limited Humanoid Robot Using Precomputed Tables | Authors: | Müller, Judith | Supervisor: | Frese, Udo | 1. Expert: | Frese, Udo | Experts: | Gräser, Axel | Abstract: | Why can't I buy a flower watering robot? No unusual question considering that there are already affordable vacuum cleaner and lawn mower robots available. Thus the next questions are, what requirements must be met by a flower watering robot and does it take a high-priced research robot to do it? As the activity of flower watering contains a variety of difficult tasks this work is concentrated on the intermediate step of autonomously exploring, detecting and grasping an unknown object such as a watering can. In particular, an overall system is introduced that enables this functionality for the performance-limited humanoid robot NAO. Furthermore, the presented system operates online and thus without external computation at runtime. In order to increase the system performance several precomputed tables containing the workspaces of different body parts are used. Moreover, the precomputed workspace constitutes as a central element as it is not only used for motion planning but also for the viewpoint planning. Furthermore, the workspace provides a connection between the very limited degrees of freedom and the need of mobility in reality. |
Keywords: | exploration; grasping; humanoid robot; NAO; autonomously; motion planning; viewpoint planning; kinematics; laser range finder; depth information; unknown objects | Issue Date: | 9-Feb-2016 | Type: | Dissertation | Secondary publication: | no | URN: | urn:nbn:de:gbv:46-00105055-15 | Institution: | Universität Bremen | Faculty: | Fachbereich 03: Mathematik/Informatik (FB 03) |
Appears in Collections: | Dissertationen |
Page view(s)
394
checked on Nov 22, 2024
Download(s)
135
checked on Nov 22, 2024
Google ScholarTM
Check
Items in Media are protected by copyright, with all rights reserved, unless otherwise indicated.