The idea of using robots for arduous tasks like mining probably dates from the dawn of the industrial age, and it’s long been a mainstay of science fiction. Isaac Asimov, for one, wrote of autonomous asteroid-mining robots in the 1940s.
The concept is now reality, with robotic haul trucks hard at work at several mines around the world. Many experts expect their numbers to grow substantially in the coming years, potentially changing the face of the mining industry.
Equipment manufacturers began discussing autonomous mining in the 1970s. But commercializing it took numerous technology breakthroughs, along with developments that slashed hardware costs. Among these are GPS for navigation, high-speed wireless communication networks, and so-called “perception” technologies: laser, ultrasonic, radar, and other sensors that can detect nearby stationary and moving objects.
There have also been advances in hardened electronics, processing speeds, and control algorithms. And the acceptance of innovations such as electrohydraulics — where electrical signals command powerful actuators — play a key role in “drive-by-wire” steering, braking, and work functions.
Even with all this, autonomous mining is only in its infancy and, today, it centers around massive “ultraclass” dump truck trucks that carry several hundred tons of ore. They run predictable routes in open-pit mines, loading in one place and delivering the material to another.
Fueling commercialization of these driverless trucks are the potential benefits they bring. For example, they make production more efficient by letting mine operations run consistently around the clock. There are no operators who need coffee and lunch breaks, and grow fatigued near the end of a shift. The vehicles eliminate the productivity lost during shift changes from one crew to another, when conventional trucks sit idle.
Another benefit is consistent performance. Trucks can be programmed to stop within inches of a specific location — say right beneath a shovel – time after time, making for quicker loading. Precisely controlling the truck’s speed and route improves fuel economy and tire life. And always running within design specifications should limit unplanned breakdowns and lengthen the life of mechanical components.
Such consistency leads to higher efficiency, lower operating costs, and better overall productivity. According to John Meech, professor of mining engineering at the University of British Columbia, benefits of autonomous hauling systems (AHS) include:
- Increase of 15 to 20% in output.
- Decrease of 10 to 15% in fuel consumption.
- 5 to 15% decrease in tire wear.
- Truck use (up-time) increases 10 to 20%.
- Maintenance costs drop 8%.
Eliminating breaks and other productivity improvements let the trucks work two to four more hours per day. This could significantly raise a mine’s output, or permit fewer trucks and a smaller fleet.
AHS also improves safety. Mining is dangerous. Simply getting in and out of the cab in these mammoth vehicles means climbing a 20-ft ladder — perhaps covered with mud or ice — so it’s an invitation for injury. And the work can be monotonous. Drivers who get tired or even fall asleep can cause collisions or drive off the road into mines that can easily be a 1,000-ft deep. Autonomous operations eliminate these hazards by letting trucks precisely navigate the road, spot obstacles, track other vehicles, and follow them at a safe distance.
Mine operators also say autonomous trucks help alleviate a growing shortage of qualified drivers, especially as mining activity expands worldwide and operations move into increasingly remote and rugged regions. Drivers must withstand isolation, boredom, and grueling conditions such as high altitudes, arctic climates, and arid deserts. Fewer workers are willing to tolerate the hardship and inconvenience and, for those who do sign on, turnover is frequent.
AHS may also attract high-tech workers to the industry who, in the past, would have never considered a career in mining. A new generation of employees who have grown up with powerful computers, video games, and the Internet will be needed to design, run, and maintain autonomous operations.
One downside is that without drivers, faults that crop up might initially go unnoticed. Humans intuitively sense surroundings in ways that today’s autonomous vehicles cannot. OEMs do offer onboard condition-monitoring systems that track the health of major subsystems and warn of maintenance needs or even shut down operations. But mechanics are still needed to perform visual safety checks during refueling.
Compared with traditional mines, costs for AHSequipped sites are undoubtedly higher. But a fair amount of the infrastructure required — such as GPS and communication networks — is already in place.
One immediate consideration when weighing costs is that manning conventional trucks, which can cost $6 million each, can be hugely expensive. According to a report in the business publication The Australian, mining executives say it costs $1 million a year or more to staff a mining truck at remote sites. For starters, drivers are expensive, with salaries as high as $120,000/yr. Four drivers are needed to keep each truck running around the clock. Add to that the costs of training, housing, meals, health care, and flying them in and out of inaccessible locations for weekly shift changes, and costs easily top the milliondollar mark.
Reducing maintenance through autonomy, say by extending tire life, also generates significant savings. Tires on the trucks can cost $40,000 apiece, and each vehicle requires six. On sizable fleets, the savings quickly add up.
Considering productivity gains, along with lower labor and operating costs, and mine operators and equipment builders alike say AHS will add millions to the bottom line.
Tokyo-based Komatsu has been operating a small fleet of autonomous haul trucks at Codelco’s Gabriela Mistral copper mine in Chile since 2008, and more recently at Rio Tinto’s “Mine of the Future” project in the Pilbara region of northwest Australia.
Komatsu’s AHS, called FrontRunner, is a comprehensive fleet-management system for mines. Rio Tinto’s fleet of driverless 930E Komatsu trucks, with payload capacities of 320 tons, are each equipped with vehicle controllers, high-precision GPS, laser and radar obstacle-detection sensors, and a wireless communications network, all controlled by a supervisory computer.
The trucks first learn travel lanes around the huge open-pit mine. The 930Es are tracked and coordinated with other vehicles, with all turns, stops, loading, and dumping sequences programmed into the computers. Any deviation from route or function, including unexpected encounters with obstacles, people, or other vehicles slows or stops the truck.
Once programmed, the trucks go to work. The supervisory computer sends data on target course and speed to the haulers while GPS determines position. When loading, the dump trucks are automatically guided to the loading spot after calculating the bucket position on a GPSfitted manned excavator or wheel loader. The supervisory computer also sends information on specific courses to dumping locations, including waste dumps, stockpiles, and crushers.
AHS improves hauler productivity, while less manpower reduces operating costs, according to Komatsu. For example, eliminating wasteful accelerations and decelerations reduces fuel consumption and prolongs tire life. And safety is enhanced because the fleet-control system prevents collisions with other dump trucks, mining equipment, and service vehicles at the site. Rio Tinto says large-scale autonomous hauling lets it move more material faster and safer, directly improving productivity, and claims results so far bear that out.
Operations began with five trucks at Rio Tinto’s West Angeles mine. In the first two years, the trucks operated around the clock and moved more than 42 million metric tons of material in approximately 145,000 cycles, traveling more than 280,000 miles — the equivalent of more than 10times around the Earth. They have since joined five new trucks at a nearby mine, where Rio Tinto expects to deploy 150 AHS-equipped haulers by 2015.
Rio Tinto originally housed the supervisory computer at the mine. But as part of its Mine of the Future, it established a control center in Perth, about 600 miles from the Pilbara mines, to remotely direct all aspects of mine operations. This will eventually include drills, excavators, dozers, graders, and even mile-long robotic ore trains.
Komatsu officials admit that one challenge for future development is making AHS more humanlike. Veteran drivers with superior driving skills get the best possible performance out of the trucks. AHS, on the other hand, needs to make a number of limitations to the driverless trucks, such as speed, mainly to ensure safety.
For example, in initial trials at West Angeles, average velocities were around 5 to 7mph (loaded) and 9 to 11mph (empty). The vehicle’s rated top speed is 40mph.
The task, say Komatsu officials, is developing a system that can match human levels of intelligence and skill while harnessing autonomy’s advantages, such as precision. The company’s goal is developing software smart enough to let driverless trucks understand how to interact with other heavy earthmoving equipment, much like humans do in the course of normal operations. If the trucks can respond flexibly to different situations without causing accidents, they should be able to perform on par with their operatordriven counterparts.
Caterpillar, Peoria, Ill., first demonstrated an autonomous truck in the mid-1990s and, more recently, has been running trials with driverless haul trucks at mines in New Mexico and Australia. In 2011, the company announced plans to supply Fortescue’s Soloman iron-ore mine in Australia with 45 autonomous 793F haul trucks by 2015.
Autonomous vehicles at the Solomon mine will use the company’s Command control system. Command is part of the Cat MineStar system for surface and underground mobile mining equipment. The system’s capabilities range from fleet assignment and condition monitoring to remote and autonomous control. It reportedly provides real-time machine tracking, assignment, and productivity management, giving a comprehensive overview of all operations from anywhere in the world; and uses guidance technology for high-precision loading and hauling operations. It is said to increase machine productivity and provide real-time feedback that improves efficiency, while object detection and avoidance capabilities enhance safety.
Future developments in mobile equipment
Hitachi Construction Machinery, Tokyo, of one the world’s largest mobile-equipment builders, announced last September plans to develop a comprehensive autonomous hauling system for the surface-mining industry. It will reportedly install advanced drive and traffic-control components on board vehicles. Combined with wireless communications and a fleet-management system, it will automate truck loading, hauling, and dumping.
According to company officials, the objective is to deliver an advanced yet pragmatic AHS, based on Hitachi’s dump-truck ac drive, that safely delivers higher productivity for a lower overall cost per ton of material moved. Field tests are planned for this year with expected deliveries in 2017.
In addition to haul trucks, a number of companies are researching autonomy for other types of mining equipment. Engineers at Volvo Construction Equipment, Brussels, Belgium, working with universities in Sweden and Germany, have developed a wheel loader and an excavator that do not require operators, yet can handle simple digging and loading tasks. The work is an offshoot of Volvo CE’s efforts to automate machine functions like driving and bucket filling and emptying.
Officials admit that currently, the software cannot yet cope with dynamically changing conditions. The project’s next stage will be to improve the behavior-based control network to manage complex tasks in a constantly changing environment. The current objective is that autonomous machines attain the equivalent of 70% of a skilled operator’s productivity level.
Commercialization may be decades away, but the technology could drastically reduce machine operating costs, say Volvo officials. Some aspects could be in production within a few years. For example, as vehicles gain “awareness” of their surroundings, it’s possible for them to make decisions based on that information. An excavator, for instance, can potentially sense resistance, calculate weight in the bucket, adapt the hydraulic pressure, and optimize efficiency, depending on the situation.
Another significant advantage: higher machine intelligence lets engineers introduce active safety. Volvo has set the target of reducing accidents involving its equipment to zero. Company officials say they intend to launch systems that detect nearby obstacles and people, and warn if the machine itself is in danger (for example, about to tip over). Longer term, communication networks that let machines “talk” with one another and a central controller will help avoid collisions and facilitate an efficient flow of equipment — say for loading.
Some operations are better suited than others for automation, according to Volvo. In particular, these include situations where humans are exposed to hazards, and where tasks are simple and monotonous.
For the mining industry, a longterm objective is the completely autonomous mine. It will not only include unmanned haulers, but drill rigs, excavators, shovels, loaders, and dozers. Given heightened interest, financial benefits, and growing R&D efforts, some experts predict that could happen in the next 5 to 20 years.
More automation in mobile rigs
Sandvik, based in Amsterdam, Netherlands, and Flanders Electric, Evansville, Ind., have partnered to automate surfacemining drill rigs. The Flanders Ardvarc (Advanced Rotary Drill Vector Automated Radio Control) system precisely positions and levels the rig, drills to the desired depth, retracts the bit, and resets the jacks in preparation for the next move. It also monitors the surrounding area for obstacles with a 3D-imaging system to ensure safe machine operations.
Company officials say it improves productivity by as much as 30% by reducing machine downtime. Conventional human-operated drills, in contrast, must shut down during blasting and shift changes. Ardvarc-enabled rigs have reportedly drilled more than 1 million holes.
Leica Geosystems Mining, Brisbane, Australia, and Autonomous Solutions, Mendon, Utah, have partnered on several OEM-independent autonomous mining-machine controls. The Leica J3dozer autorip, for example, is a fully autonomous ripping-control system for track and wheel dozers. Unmanned dozers can clear and prepare a predefined area and automatically stop when the job is complete.
The company’s J3truck guideline is for haul-truck convoys. It requires one human driver in the lead vehicle, with an unlimited number of unmanned vehicles following the exact path of the leader in a convoy. It reportedly is capable of high speeds, turns, quick maneuvering, and emergency stops.