Hongseok Kim

Date of Award

3-2022

Document Type

Thesis

Degree Name

Master of Science

Department

Department of Systems Engineering and Management

First Advisor

David R. Jacques, PhD

Abstract

The aim of this research is to provide autonomous navigation of a 4 wheel vehicle using commercial, off-the-shelf depth and tracking cameras. Some sensitive operations need accuracy within a few inches of navigation ability for indoor or outdoor scenarios where GPS signals are not available. Combination of the Visual Odometry (VO), Distance-Depth (D-D), and Object Detection data from the cameras can be used for accurate navigation and object avoidance. The Intel RealSense D435i, a depth camera, generates depth measurements and the relative position vector of an object. The Intel RealSense T265, a tracking camera, generates its own coordinate system and grabs coordinate goals. Both of them can generate Simultaneous Localization and Mapping (SLAM) data. The cameras share their data to provide a more robust capability. Combining the Intel cameras with a Pixhawk autopilot, it was demonstrated that the vehicle can follow a desired path and avoid objects along that path.

AFIT Designator

AFIT-ENV-MS-22-M-216

DTIC Accession Number

AD1173775

Download

Share

COinS