Date of Award

11-2016

Document Type

Thesis

Degree Name

Bachelor of Science (B.S.) in Mechanical Engineering and University Honors

Department

Mechanical Engineering

First Advisor

Nathalie Neve

Subjects

Drone aircraft -- Control systems -- Design and construction, Drone aircraft -- Automatic control, Optical radar -- Industrial applications

DOI

10.15760/honors.342

Abstract

Personal UAV's are experiencing an increase in both functionality and popularity. Consumer-driven markets are encouraging the UAV industry to innovate towards the average user. Despite improvements in user friendliness, cost, and technological capabilities, consumer UAV's are limited by pilot skill and a lack of sensory information. This paper presents two approaches for creating a low-cost, open-source platform capable of enabling a consumer-level UAV to avoid hazards. A single beam LIDAR system and a combined LIDAR-vision system are considered. The LIDAR variant is capable of detecting and avoiding static obstacles in simple environments. The improved LIDAR-vision system is able to detect and avoid moving hazards in complex environments such as a bike path with pedestrian foot traffic. Finally a unified system with improved hardware capabilities is proposed.

Persistent Identifier

http://archives.pdx.edu/ds/psu/18709

Share

COinS