Date of Award
8-2020
Document Type
Thesis
Degree Name
Master of Science (MS)
Department
Aerospace, Physics, and Space Sciences
First Advisor
Markus Wilde
Second Advisor
Hector Gutierrez
Third Advisor
Brian Kish
Fourth Advisor
David Fleming
Abstract
At present, lava tubes on the Moon and Mars have been left wholly unexplored. There are many aspects of subterranean exploration that pose additional challenges over surface exploration, and one of these issues is in localization. This project seeks to expand upon the traditional communication beacon that is dropped behind a robotic operator in subterranean environments and make low cost and low power modifications to increase their functionality by making robot localization from these beacons possible. This has been tested with the use of LED beacons that could be easily added to current solutions and used for localization of the robot in areas that otherwise offer poor options for dead reckoning or odometry. This thesis presents a low-cost camera array using modern webcams and explores whether such a system can built with commercial off-the-shelf parts. The webcams are compared with 10 mm and 18 mm lenses on a Canon T6, and the system is used to calculate the global position and orientation of the array exclusively using the visual targets. Using the Robot Operating System (ROS) and OpenCV machine vision libraries, this project was able to detect the markers at ranges up to 5 m and angles of 30°. Two versions of the visual beacon system are described and tested. This research shows that while webcams still do not possess the required sensitivity and resolution for determining the color of LEDs at range, low cost DSLRs are capable of discerning not only location, but also the color of markers on such a system at range.
Recommended Citation
Capozzi, Ryan Joseph, "Beacon Aided Robotics for Martian Cave Mapping" (2020). Theses and Dissertations. 414.
https://repository.fit.edu/etd/414