Inside Unmanned Systems

FEB-MAR 2018

Inside Unmanned Systems provides actionable business intelligence to decision-makers and influencers operating within the global UAS community. Features include analysis of key technologies, policy/regulatory developments and new product design.

Issue link: https://insideunmanned.epubxp.com/i/945861

Contents of this Issue

Navigation

Page 56 of 67

ENGINEERING. PRACTICE. POLICY. 57 February/March 2018 unmanned systems inside Biologically inspired navigation strategies could soon help underground vehicles operate autonomously within mazes of mining tunnels. by Charles Q. Choi D rones that f ly through the air and self-driving cars that mo- tor down roads are the most publicized advances recently made with unmanned vehicles. However, major breakthroughs are also being made with unmanned vehicles that rumble underground. Now researchers in Australia are developing biological- ly inspired navigation strategies that could soon help giant machines op- erate autonomously within mazes of mining tunnels. Mining "is probably one of the most advanced and innovative industries in terms of using unmanned vehicles— it's been using them now for 10 or 20 years, at least," said Michael Milford, a professor of robotics and chief in- vestigator of the Australian Center for Robotic Vision at the Queensland University of Technology in Brisbane. Mines are typically quite dirty, "some of the jobs are quite dull, and it's rela- tively dangerous, and that combination of dirty, dull and dangerous is always a ripe target for introducing automa- tion," Milford said. The London-based mining corporation Rio Tinto has ex- perimented with automated trucks on the surface of mines, and researchers have conducted significant work un- derground with so-called load-haul- dump trucks "carting earth and rocks from one pile to another," Milford said. However, underground mines pres- ent a number of challenges to un- manned vehicle navigation. For in- stance, while drones can rely on GPS, underground unmanned vehicles can- not because the GPS signal doesn't penetrate through to the mine. Overcoming Navigation Challenges One approach to help unmanned ve- hicles navigate underground is to give them laser scanners, which are currently used by self-driving cars. "However, these are quite expensive and typically bulky," Milford said. Another strategy is to install wireless transmitters in mines that unmanned vehicles can use as beacons, but these infrastructure modifications are simi- larly expensive, require power cables, and the rock of a mine can interfere with the wireless signals, he said. In the past five to 10 years, there has been growing interest in using cameras as the main sensors in un- derground unmanned vehicle navi- gation, Milford said. "Camera tech- nology is rapidly developing because of the incredible pressure from con- sumer demand for new generations of smartphones—a camera you can get now is extraordinarily more capable than one you can get of the same price five years ago," he said. "Moreover, cameras are relatively low-power sensors, and are relatively compact," Milford said. "And visual data can potentially have a lot of intel- ligence applied to it, giving it the most long-term potential for enhancing au- tomation. A laser can give you the dis- tance to an obstacle, but not read the expression on a person's face or other subtle cues that a camera could pick up." Learning Your Way Around "IN THE NATURAL WORLD, ANIMALS ARE NOT WALKING WITH $10,000 LASERS STRAPPED ON THEIR HEADS TO NAVIGATE. WE LOOKED AT ANIMALS LIKE RATS TO DEVELOP MODELS OF WHAT HAPPENS IN THEIR BRAINS WHEN THEY NAVIGATE, AND NOW WE'RE DEPLOYING THEM WITH MINING SYSTEMS." Michael Milford, professor of robotics and chief investigator of the Australian Center for Robotic Vision, Queensland University of Technology in Brisbane

Articles in this issue

Archives of this issue

view archives of Inside Unmanned Systems - FEB-MAR 2018