Development of Comparative Test Protocols for the Assessment of Autonomous Tractor Performance

Torrance, Logan (2020) Development of Comparative Test Protocols for the Assessment of Autonomous Tractor Performance. [USQ Project]

[img]
Preview
Text (Project)
Torrance_L_Baillie_Redacted.pdf

Download (22MB) | Preview

Abstract

As autonomous and semi-autonomous tractors (ASAT) become more prevalent and affordable within the agricultural industry, various standards that outline both the safety and design principles for ASATs have been developed. Current features on late model tractors inform most of the major components that would be required for an autonomous tractor – with these technologies ultimately providing a pathway from basic automation to full autonomy.

While such standards ensure certain levels of safety and/or performance are achieved, there are currently no universally accepted documents or testing protocols that assess the field-readiness or the level of performance/maturity of (semi-)autonomous tractors. Therefore, this project aims to develop ASAT testing protocols and to assess the performance of current tractor technology, relative to the suggested requirements.

Building upon existing research and standards of the mining, transport and agricultural sectors, a list of expected operations that an ASAT should be expected to perform was compiled, prior to developing test procedures to exploit certain operations and/or protocols. A John Deere 6120R case study was then implemented, to assess the appropriateness of test procedures and the performance of a market-ready ASAT.

The project presents recommendations for the introduction of universally accepted, independent testing procedures to ensure ASATs meet accepted levels of performance and field-readiness, pertaining to awareness and perception, automated tractor guidance, headland management and operational safety. The project also assessed the maturity and performance of existing tractor technologies, relevant for autonomous and semi-autonomous farming operations. Implementing this scoring method, the 6120R case study performed well across a number of elements, obtaining an overall mark of 8.3/10. The tractor benefitted from advanced headland management and operational safety protocols, while lacking in-depth perception and awareness practices – ultimately limiting its driverless capabilities.

While procedures were outlined for obstacle detection and avoidance systems, these protocols could not be tested due to limitations of available machinery. Therefore, further work should involve assessing the practical implementation of perception systems, prior to presenting these recommended tests to tractor manufacturers for feedback and refinement – thereby accelerating acceptance and uptake of these tests.


Statistics for USQ ePrint 43026
Statistics for this ePrint Item
Item Type: USQ Project
Item Status: Live Archive
Additional Information: Bachelor of Mechatronic Engineering (Honours)
Faculty/School / Institute/Centre: Current - Faculty of Health, Engineering and Sciences - School of Mechanical and Electrical Engineering (1 Jul 2013 -)
Faculty/School / Institute/Centre: Current - Faculty of Health, Engineering and Sciences - School of Mechanical and Electrical Engineering (1 Jul 2013 -)
Supervisors: Baillie, Craig; Brett, Peter; Baillie, Justine
Date Deposited: 11 Aug 2021 06:18
Last Modified: 11 Aug 2021 06:43
URI: http://eprints.usq.edu.au/id/eprint/43026

Actions (login required)

View Item Archive Repository Staff Only