An autonomous heavy haul railway system capable of carrying driverless trains up to 2.5 kilometres long took home WA’s top engineering award for 2020. 

AutoHaul, dubbed the “world’s largest robot” by operator Rio Tinto, became the world’s first fully autonomous long-distance heavy-haul rail network when it went into full operation in June 2019. 

The vast network comprises some 1700km kilometres of track, with each train delivering tens of thousands of tonnes of iron ore from 16 Rio Tinto mines to ports in Dampier and Cape Lambert. 

In September it was announced as one of six West Australian winners of Engineers Australia’s Engineering Excellence Award and also the sole WA finalist for the Sir William Hudson Award, awarded annually to the best engineering project in the country. 


Get the latest stories from delivered to your inbox.

| Privacy policy

While the establishment of AutoHaul involved collaboration with partners from Japan and the US, its success is also testimony to the skills of WA engineers who were pivotal to a project that was a decade in the making. 

Lido Costa, Rio Tinto’s Principal Engineer for AutoHaul, previously worked on the high-speed Madrid-Seville train in Spain and was also chief engineer for the Perth-Mandurah line. 

He told the Rio Tinto website that AutoHaul was the “by far” the most exciting project he had worked on and delivered both time savings and safety benefits by eliminating the need for trains to stop for end-of-shift driver changeovers. 

A typical journey on a traditional train would require three such stops, adding more than an hour to journeys. 

“The time-saving benefit is enormous because the train network is a core part of the mining operation. If we can prevent those stoppages, we can keep the network ticking over, allowing more ore to be transported to the ports and shipped off more efficiently,” Lido said. 

See also  Turning waste into taste in the Pilbara

“The other major benefit is safety. We are removing the need to transport drivers 1.5 million kilometres each year to and from trains as they change their shift. This high-risk activity is something that driverless trains will largely reduce.” 

While a controller at Rio’s operations centre in Perth sets the initial route for AutoHaul trains, computers take over once they are up and running, making decisions about speeds, avoiding contact with other trains via collision detection systems, and ensuring there is nothing obstructing level crossings. 

AutoHaul also detects faults with trains and will stop them if necessary. 

The average return distance for AutoHaul trains is about 800 kilometres, a journey which – factoring in loading and dumping – takes about 40 hours.