Further development of a human tracking and motion interaction capability for a robotic avatar

Abstract In this world of globalization, there is a higher demand for people to work at different places or meeting with customers, clients, suppliers, or supervisor from different countries around the world. Hence, conference incorporated with telepresence function become a popular research topi...

Full description

Saved in:
Bibliographic Details
Main Author: Lee, Yong Liang
Other Authors: Seet Gim Lee, Gerald
Format: Final Year Project
Language:English
Published: 2014
Subjects:
Online Access:http://hdl.handle.net/10356/60957
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:Abstract In this world of globalization, there is a higher demand for people to work at different places or meeting with customers, clients, suppliers, or supervisor from different countries around the world. Hence, conference incorporated with telepresence function become a popular research topic recently. It aims to provide people from all walks of lives an opportunity to communicate and interact with other people at different locations effectively without leaving their existing place. With this motivation and interest behind, Robotics Research Centre started a research on the development of Telepresence Robot as known as “Robotics Avatar”. “Robotics Avatar” is a mobile robot, which is able to interact with humans and their environment through holonomic motion, vision, audio, head, and arm gesturing. At the same time, it will create a telepresence effect to people under the remote control of the user. In this project, research and development work will be focused on the module of holonomic motion of the robot. In the previous research project, two human following capabilities, which are Back Following Mode and Side-by-Side Following Mode, had been implemented on the mobile robot. In pursuit of accomplishing all the human following capabilities, Front Guiding Mode was under study and implemented in this project. Various functions like sound guidance, graphical user interface, human selection, waypoint selection, waypoint navigation, and obstacles avoidance were executed in the Front Guiding Mode. Besides, an algorithm had been developed to control the speed of the robot with respect to the human-robot distance. Furthermore, in the pursuance of executing three human following capabilities on a mobile robot effectively, three fixed Kinect sensors were changed to a rotating Kinect sensor, actuated with a dynamixel servo motor. Hence, in this project, a research on rotating Kinect sensor was done and a new human tracking capability was implemented onto a mobile robot with rotating Kinect sensor. In addition, an experiment was designed and conducted to analyse the robustness of the human tracking capability on rotating Kinect sensor. Experiment results indicate that the human tracking capability was excellent but the response of the robot to human motion needed to be improved.