Bourke, Christopher (2022) UAS autonomous drop zone assessment and adaptive delivery. [USQ Project]
|
Text (Project)
Bourke_C_Low_Redacted.pdf Download (3MB) | Preview |
Abstract
The aim of this dissertation was to research and implement a stereoscopic machine vision sensor system on an uncrewed aerial system. The variables affecting the quality and computational cost of the stereo algorithm were assessed and optimised to be employed on a low-cost companion computer, the Raspberry Pi 4 (4GB). The algorithms used proved computationally expensive and the results tested the limit of the companion computer’s abilities to reach the minimum operating speeds required for UAS logistic delivery missions. The outcomes demonstrated an accurate and dense disparity map on benchmark datasets but the image sizes and tuning in field testing demonstrated a limited efficacy in application.
Measured distance error slightly increased across the effective ranges where disparity was calculated. The cumulative errors from field of view and focal length, taken from the manufacturers specifications rather than directly assessed, and the baseline measurement which all effect the distance measurement after disparity is calculated.
From the implementation and assessment carried out in this paper it is evident that the largest factor affecting the calculation speed of the system is the captured image size and represents the limiting factor in employing this implementation in its current configuration.
|
Statistics for this ePrint Item |
| Item Type: | USQ Project |
|---|---|
| Item Status: | Live Archive |
| Faculty/School / Institute/Centre: | Current – Faculty of Health, Engineering and Sciences - School of Engineering (1 Jan 2022 -) |
| Supervisors: | Low, Tobias; Perera, Sirigalpatabandige Ruveen |
| Qualification: | Bachelor of Engineering (Honours) (Mechatronics) |
| Date Deposited: | 19 Jun 2023 03:32 |
| Last Modified: | 20 Jun 2023 01:09 |
| Uncontrolled Keywords: | stereoscopic machine vision sensor system; uncrewed aerial system (UAS) |
| URI: | https://sear.unisq.edu.au/id/eprint/51871 |
Actions (login required)
 |
Archive Repository Staff Only |
