Investigating the feasibility of conducting human tracking and following in an indoor environment using a Microsoft Kinect and the Robot Operating System

Loading...
Thumbnail Image
Authors
Greenberg, Rebecca A.
Subjects
robot
Robot Operating System
Kinect
P3-DX
human detection
human tracking and following
Advisors
Yun, Xiaoping
Calusdian, James
Date of Issue
2017-06
Date
Jun-17
Publisher
Monterey, California: Naval Postgraduate School
Language
Abstract
Human detection, tracking, and following is one application in which computer vision can be relevant to robotics. By using a sequence of images, a human can be found and that human's movement can be followed. The Microsoft Kinect, one of the most successful color image and depth (RGB-D) sensors, is known for its human detection capabilities and has multiple software development kits available. The objective of this thesis was to determine if it was feasible to implement human tracking and following on a mobile robot in an indoor environment. Specifically, the tracking was conducted with the Microsoft Kinect and a specific software development environment, Robot Operating System (ROS) and MATLAB. ROS was utilized to run the drivers for the robot and the Microsoft Kinect, while MATLAB was utilized to run the algorithms and experiments. The skeleton tracking capabilities of the Kinect were utilized as the main tracking system. An auxiliary method was created by using histograms of depth and region properties to segment a person from a depth image. The indoor robot was able to successfully track and follow a person through the indoor environment using the raw sensor data and a combination of the two tracking methods.
Type
Thesis
Description
Series/Report No
Department
Electrical and Computer Engineering (ECE)
Organization
Identifiers
NPS Report Number
Sponsors
Funder
Format
Citation
Distribution Statement
Approved for public release; distribution is unlimited.
Rights
This publication is a work of the U.S. Government as defined in Title 17, United States Code, Section 101. Copyright protection is not available for this work in the United States.
Collections