The Pilbara in Western Australia is an arid expanse of barren terrain the size of France. It’s home to the West Angelas open pit mine where millions of tons of iron ore are annually gouged up and then transported 24,000 tons at a time in mile-long ore trains bound for Cape Lambert, some 200 miles west on the Indian Ocean, for shipment even further west, mainly to the ever-hungry steel mills of China.

A win-win for robotics in the Pilbara

The Pilbara is also a huge open-air laboratory, where for the past three years robotic mining equipment has been put through its paces to see if it can stack up to the rigors and pace of high-volume ore mining. The experiment was orchestrated by the mine’s owner, Rio Tinto–the third-largest mining company in the world–which also owns fourteen other mines in the Pilbara (Pilbara region accounts for 70% of Rio’s $15.5B annual earnings (2011)).

The robot trials were part of a movement toward a future of autonomous mining in which Rio Tinto is investing $3.6 billion, the ultimate goal of which the company calls the Mine of the Future.

Starting in 2008 and concluding in February of 2012, a vanguard of nine robots–an excavator, track dozer, wheel dozer motor grader and five 930E-AT haul trucks–all supplied by Komatsu Ltd, underwent extensive field testing at West Angelas. The Pilbara robot trials were a real-world success. The most-watched machine was the autonomous, AT version of the Komatsu 930E, the best-selling off-road dump truck in the world–and now soon-to-be strategic hauling tool of choice for the Pilbara’s millions of tons of China-bound iron ore.

Rio Tinto is pushing forward with a full-scale, autonomous mine implementation, including $442 million for autonomous ore hauling trains over its rail network. Rio Tinto’s iron ore chief, Sam Walsh, predicts a positive outlook for medium to long-term iron ore demand to be driven by the Chinese, even with the current cooling of the Chinese economy.

Still the fastest growing economy in the world, Credit Suisse cut its estimate on China to 7.7 percent from 8 percent on June 14, while Deutsche Bank lowered its forecast to 7.9 percent from 8.2 percent (compare U.S. 2.3 percent for 2011). The predictions indicate the weakest growth since 1999 and compare with a 9.2 percent expansion last year (China: 11.9 percent, 1Q2010). Commodity analyst, David Lennox, Fat Prophets, sees iron ore still a strong Chinese import as the country works to meet the demands of its near-8 percent growth rate. Through it all, China remains both the world’s largest producer and largest buyer of iron ore. Good news for the Pilbara’s miners.

Rio Tinto, even with falling commodity prices and China’s declining growth rate, is targeting a 25% ramp up of all its Pilbara mine production over the next four years, to raise exports to 353 million tons per year (West Angelas produces 29.5 million tons per year), all of which will necessitate that robotics and autonomous mining must play an increasingly significant, if not a critical, role.

Komatsu, Ltd., which has a twenty-year working relationship with Rio Tinto and is the world’s second largest construction and mining equipment manufacturer (Caterpillar is the largest), is also a preeminent pioneer and the world’s leading manufacturer of robotic/autonomous mining equipment. In fact, it’s the only equipment manufacturer with autonomous vehicles actually working in real-world mining environments. Prior to the West Angelas mine, Komatsu had eleven 930E-AT autonomous haul vehicles at work in CODELCO’s copper mines in Chile.

Rio Tinto & Komatsu: perfect partners, perfect timing for robot success

In February, after a successful, three-year trial hauling waste products for nearly 900 days and over 350,000 miles, Komatzu’s five trucks were transferred to the nearby Yandicoogina mine were they will be joined by five additional driverless 930E-ATs for a full deployment. Rio Tinto has also ordered another 150 of the mega-size, off-road robots, at a cost of $6 million each, with delivery spread out over the next four years.

The 6-wheel, 200-ton behemoth, standing 24 feet high guzzles sixty gallons of gas an hour from its 1200-gallon tank (a gallon a minute!) while lumbering around at top speed of 40mph with its 320-ton payload. Each of the giant tires on the 930E-AT costs $40,000 apiece. To transform such a mega-machine into a driverless vehicle capable of operating efficiently and safely, virtually day and night in an open pit mine is an enormous undertaking and a great advance for robotic vehicles.

A robotic full monty–and then some

From start to finish, the conversion process of 930Es to driverless, field-ready 930E-ATs was carried out very intelligently between partners well matched in skill and innovative spirit. Komatsu ventured its flagship truck as the guinea pig to its best client; a client who also just happened to be ramping up pursuit of its Mine of the Future plan. Komatsu spokesman, Akitoshi Watanabe, claims that the company provides two decades of automation experience, much of which goes back even further to the mid-1980s. Risky business: considerable loss of esteem for the 930E’s maker–and maybe lost sales of the giant off-road vehicle as well–might have resulted in a failed or even slightly botched conversion to the AT model, not to mention retarding the full bore advance of Rio Tinto’s Mine of the Future. Imagine, for instance, news headlines of a bunch of 930E-ATs running amok at a mine? Not pretty.

Rio Tinto’s innovation group built alliances with the Australian Centre for Field Robotics (ACFR) at the University of Sydney early on in 2007, establishing the Centre for Mine Automation to which Rio Tinto committed an initial $21m of funding. According to the University of Sydney, its mission at the Rio Tinto Centre for Mine Automation (RTCMA) is to develop and implement the vision of a fully autonomous, remotely operated mine–both open pit and underground.

Enhancing and extending the research capabilities of the Centre for Mine Automation, Rio Tinto launched an innovation center with iGATE Patni in Pune, India. Rio Tinto head of Innovation, John McGagh, sees the center’s 300 engineers as driving innovation through to implementation at an operational level. “It will take innovative ideas that have been developed by the Rio Tinto centers and trialed in a specific mine or business unit, and modify the technology for use with other products, or in different regions.”

Komatsu, in addition to the vaunted 930E, brought along decades of R&D in what it and most other industry developers call Autonomous Haulage Systems (AHS). Together with its own homegrown AHS skillset, came additional expertise in the discipline from Komatsu’s longtime partner and now wholly-owned subsidiary, Modular Mining Systems, Inc. (Arizona), and its co-founder and industry legend, Dr. James White.

Modular’s contribution to the Komatsu AHS includes the supervisory system, operational intelligence, communications infrastructure, operational reporting, and vehicle-interaction safety technologies. Dr. White, said of the deployment, “This is a major milestone for all involved with the AHS. The system is no longer a research project, but a field-proven commercial product. This is not the end of the journey it is the beginning of an exciting new phase.”

FrontRunner is the product name Komatsu has given its AHS of high-precision GPS, cross-referencing radars, lasers, computers, video cameras and communications. Rio Tinto?s fleet of driverless, FrontRunner-equiped 930E-ATs are each equipped with vehicle controllers, GPS, an obstacle detection system and a wireless network system operated and controlled by a supervisory computer.

The mega-trucks, upon arrival on site, had to first learn their travel lanes around the huge open pit mine. Trucks are tracked and mapped, coordinated to other trucks, with all turns, stops, loading and dumping sequences programmed into the control computers. Any deviation from route or function–especially unexpected obstacles–instantly stops the behemoth in its tracks.

Once each truck is programmed it can begin working the job. Modular describes the job operational flow as follows: “Information on target course and speed is sent wirelessly from the supervisory computer to the driverless dump trucks, while the GPS is used to ascertain their position. When loading, the dump trucks are automatically guided to the loading spot after computing the position of the bucket of the GPS-fitted hydraulic excavator or wheel loader. The supervisory computer also sends information on a specific course to the dumping spot.”

At first, supervisory computer control was resident at the mine, however, Rio Tinto, with an eye toward its Mine of the Future, created a NASA-like, space-age control center in Perth?750 miles from the Pilbara mines–to remotely control all aspects of mine operations–excavators, bulldozers, wheel dozers, graders, even to the one- and two-mile-long robotic ore trains (see video of Perth control center).

The future of human drivers vs. driver-less dump trucks

So where do people, if at all, fit into this driverless Mine of the Future mix? James Petty, leader of Rio Tinto’s Mine of the Future project, put the switch to robotic trucks very simply during an interview with the Australian’s Matt Chambers: “When we looked at our growth plans, we were not in a position where we thought we could find enough people to drive trucks.”

And he’s right. The scarcity of drivers is well known in the industry. Mega-truck drivers need to withstand isolation, boredom and danger from remote, open pit mines in sparsely settled locales where many of the climates are on the edge of being minimally tolerable. Such adversity has drastically reduced their numbers over the years.

The same goes for any skilled trade in the mining industry, reports Canada’s Mining Industry Human Resource Council, which predicts there will be a shortage of 110,000 skilled workers in less than 10 years. Rio Tinto agrees, saying that even with mine automation, it will still need 150,000 workers across its global operations by 2020.

As for truck drivers, they are expensive at salaries of $120,000. Rio Tinto claims that it takes over $1 million to staff a truck for one year. In addition to driver wages (each truck is staffed by four drivers and operated around the clock); there are the additional costs of housing, feeding, training and flying them in and out of far-flung mining locations.

John Meech, professor of mining at the University of British Columbia, insists to his students, “not to justify automation based on labor savings. Safety is the first justification. Every year, three or four truck drivers are killed because they’ve fallen asleep or driven over the side. It’s a dangerous occupation.”

“Productivity can go up,” continues Meech. “You don’t need to have a lunch break. You don’t need to have coffee breaks. You can immediately gain two to three hours a day in productivity on that equipment.”

Another productivity issue is that the ore quality of older mines tends to decline over time as operations shift from the original ore body, which means that operations need to become more efficient. Hence, more truckloads of ore; hence, the kind of haulage productivity only a robot can deliver.

The ultimate question is should humans take on mining as an occupation in this decidedly dangerous and oftentimes deadly work environment? Globally, there are 6.6 mining fatalities every day (most in underground mines). In many instances robots are sent into mines to rescue humans. With autonomous mining technology here and working in the field, isn’t it time to rethink the obvious?

Keep in touch with the world of robotics with the Robotics Business Review Weekly Newsletter. Just tell us where to send it.