Mine machine guidance using radar

Hargrave, Chad Owen (2014). Mine machine guidance using radar PhD Thesis, School of Information Technol and Elec Engineering, The University of Queensland. doi:10.14264/uql.2014.281

Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads
s3311488_phd_submission.pdf Thesis full text application/pdf 11.55MB 13

Author Hargrave, Chad Owen
Thesis Title Mine machine guidance using radar
School, Centre or Institute School of Information Technol and Elec Engineering
Institution The University of Queensland
DOI 10.14264/uql.2014.281
Publication date 2014
Thesis type PhD Thesis
Supervisor Vaughan Clarkson
Amin Abbosh
Nicholas Shuley
Marek Bialkowski
Total pages 288
Total colour pages 123
Total black and white pages 165
Language eng
Subjects 100505 Microwave and Millimetrewave Theory and Technology
090609 Signal Processing
090602 Control Systems, Robotics and Automation
Formatted abstract
This thesis examines the application of radar sensor technology to the problem of mine machine guidance. Microwave radar, with its good performance characteristics in dusty, dark and rugged environments, is well suited to the mining industry and in particular to the underground longwall coal mining domain. A specific aspect of longwall machine guidance was selected for examination, with two key parameters to be measured by the radar-based sensor: longwall creep and longwall retreat. The former is a radar-ranging task, the latter a waypoint navigation activity with a particular emphasis on the need for reliable target recognition. The two corre-sponding goals of this research programme were: first, the development of a practical creep sensor that would integrate into the longwall automation system; and second, an examination of the feasibility of using radar target recognition for waypoint detection and discrimination.

A practical microwave radar creep sensor (customised from the core radar-ranging components of a commercial radar level sensor) was developed and trialled in two separate underground mine-sites: the system achieved a better measurement performance than the required mean absolute ranging error of less than 50 mm. This novel sensor was developed to the point where it is due for installation in an operational coal mine in 2014. A pioneering wireless Ethernet communications system capable of integrating the creep sensor output with an existing inertial navigation unit was installed and optimised for operation across the face of a coal mine longwall; this was, to the author’s knowledge, a world-first achievement at the time of the network installation. The high speed radar ranging sensor developed for the creep-retreat measurement system was also installed as a collision avoidance sensor to provide protection for the boom of a bulk shiploader, working as a complementary technology in conjunction with scanning laser technology.

Accurate and robust longwall retreat measurement, a requirement for an autonomous longwall, can be achieved by means of waypoint tracking utilising the ubiquitous bolt-plates that are installed at regular intervals along the rib-walls of underground mine roadway tunnels. A novel application of resonance-based target recognition to a navigation problem, including a detailed study of the resonant target signature of the bolt-plate structure, confirmed that discrimination is possible between this target and the typical sources of clutter commonly found in the under-ground. In order to achieve a practical target recognition capability in the harsh mine environ-ment, two novel theoretical developments were achieved: first, a new method for the rapid estimation of the optimal window for processing the late-time resonance response of a broad-band excited radar target, and second, a novel radar target discrimination method based on histogram analysis.
Keyword Radar
Target identification
Singularity expansion method (SEM)
Complex natural resonances (CNRs)
Mining automation
Longwall coal mining
Machine guidance
Collision avoidance

Citation counts: Google Scholar Search Google Scholar
Created: Tue, 26 Aug 2014, 09:12:26 EST by Chad Hargrave on behalf of Scholarly Communication and Digitisation Service