Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/45025
Title: Gesture learning in social robots
Authors: Shen, Jiayu.
Keywords: DRNTU::Engineering::Electrical and electronic engineering::Control and instrumentation::Robotics
Issue Date: 2011
Abstract: Programming a robot to complete a task in 3D space has become too complicated. Therefore an alternative way is needed for the robot to learn. Learn by imitating becomes a tremendous tool for the robot to learn any task. Motion capture technology allows a person entire joints Cartesian coordinate to be captured. Thus the role of inverse kinematic comes in. Inverse kinematic is a way of finding various joint angle of a robot when the Cartesian coordinates of the robot various joints are given. The main task of the algorithm robust modular inverse kinematic implemented is to solve the inverse kinematic problem of a upper body humanoid in a modular manner and also to solve it in a way that makes the algorithm runs smoothly in a robust manner.This algorithm has catered to problems like drift and singularity by coming up with solution to solve the problems such as adding a regularization parameter to counter singularity. Various test were devised to ensure the correctness of the algorithm implemented. Generally the testing done has been successful as the desired outputs are quite similar to the acutal output.The program created provides an interface for further development of imitation learning.
URI: http://hdl.handle.net/10356/45025
Rights: Nanyang Technological University
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:EEE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
eB4025-101.pdf
  Restricted Access
1.35 MBAdobe PDFView/Open

Page view(s) 50

245
Updated on Nov 26, 2020

Download(s)

7
Updated on Nov 26, 2020

Google ScholarTM

Check

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.