Positioning enabling system

Estimating the position of vehicles in underground mining continues to be a major challenge as current methods are unreliable, expensive, highly inflexible and sensitive to environmental conditions.

The overall outcome of this project is the development and deployment of reliable vision-based object detection and classification methods, capable of detecting and recognising light vehicles and heavy vehicles from a significant distance, night and day, including in difficult environmental conditions.

The research project will incorporate the development of two technologies:

1. An economical and reliable camera-based positioning system for locating and tracking underground vehicles within one metre accuracy.
2. A sophisticated, multi-sensor system that delivers centimetre-accurate positioning, enabling automation of mine vehicles.

These new positioning technologies will deliver the equivalent of GPS to underground mining.

The economic benefits will be a world-first in automated positioning for underground mining without infrastructure. Socially it will dramatically improve the safety of workers as well as:

• Dramatically improve collision avoidance
• Completely remove monitoring of assets and personnel with GPS receivers
• Improve productivity and the operational control of a mine site

The Queensland Government awarded the team, lead by Robotic Vision Chief Investigator Michael Milford, $428-thousand in funding as part of its Advance Queensland Innovation Partnerships program.

Additional funding and participation by Caterpillar and Mining3, will help the team develop technologies that could ultimately enable the automation of underground mining vehicles.