Javascript must be enabled for the correct page display

Grabbing Objects with the NAO Robot Using Multiple Behaviors and Interval Estimation for their Selection

Oost, C.H.G. (2012) Grabbing Objects with the NAO Robot Using Multiple Behaviors and Interval Estimation for their Selection. Bachelor's Thesis, Artificial Intelligence.

[img]
Preview
Text
AI-BA-2012-C.H.G.OOST.pdf - Published Version

Download (904kB) | Preview
[img] Text
AkkordTijnsant.pdf - Other
Restricted to Repository staff only

Download (27kB)

Abstract

Abstract. When using a robot in the real world the ability to manipulate objects is important, but the question is how to grab the objects. In this research it is shown that it is possible to do this with multiple behaviors. Each grab attempt leads to a success or failure. The best behavior is selected with interval estimation. The interval is calculated with the Adjusted Wald formula. Six behaviors are created. Also nine objects are collected. Each behavior-object combination is tested 6 times, collecting data and bootstrapping the robot on these objects. The interval was firmly reduced after 6 trials and the confidence interval moved towards the expected average. Seven of the nine objects had a behavior that could grab them in all the trials. While there is still a lot to research in successful grabbing, this research is a step in that direction.

Item Type: Thesis (Bachelor's Thesis)
Degree programme: Artificial Intelligence
Thesis type: Bachelor's Thesis
Language: English
Date Deposited: 15 Feb 2018 07:47
Last Modified: 15 Feb 2018 07:47
URI: http://fse.studenttheses.ub.rug.nl/id/eprint/9894

Actions (login required)

View Item View Item