Autonomous haulage trucks have been part of Australian mining for over a decade now. The early pilots at Rio Tinto’s Pilbara operations are old news. What’s interesting in 2026 is what’s happening at the edges — the operational challenges that don’t get covered in glossy industry press releases.

I’ve been speaking with operations supervisors and maintenance crews across three Pilbara sites this quarter. Here’s what they’re seeing.

The fleet has plateaued, not grown

Rio Tinto’s autonomous truck count is roughly stable. They’re not retrofitting more trucks at the pace they were three years ago. That’s not because the technology stopped working — it’s because the easiest applications have been done.

The remaining manual fleet works in conditions that are still difficult for autonomous systems: complex bench layouts, mixed traffic with light vehicles, unpredictable ground conditions in newer pits. Solving those is more expensive per truck than the early conversions were.

Maintenance is the underestimated cost

The autonomous trucks don’t have drivers, but they need more sensor maintenance, more calibration time, and more software updates than the fleet they replaced. The labour cost has shifted, not vanished.

A typical autonomous truck spends about 12% more time in maintenance than its manual equivalent. Most of that is sensor cleaning, LIDAR alignment, and software validation after updates. None of it is dramatic, but it adds up across a fleet of 80+ vehicles.

The supervisors I’ve spoken with describe this as a managed cost rather than a problem. They’ve staffed up the maintenance crew accordingly. But the fleet payback period is longer than the original business case projected.

Reliability metrics are where they should be

The trucks themselves are reliable. Outage rates from technology failure are below 2% on most measured days, which is well within tolerance for production planning. When trucks do go down, the cause is more often a communication issue (fleet management network, GPS drift, mesh network handoff) than a vehicle fault.

This is a quiet success that doesn’t get press because it’s not new. The trucks haul ore. They do it 24/7. They don’t get tired. The basic value proposition is delivered.

The labour conversation has changed

Five years ago, the discussion was about whether autonomous trucks would replace drivers. That conversation is mostly resolved — yes, they did, on the routes where it made economic sense. The remaining question is what those workers transitioned into.

Some moved into operations roles in the integrated remote operations centres in Perth. Some moved to maintenance crews. Some left mining entirely. The transition was less dramatic than the early commentary suggested but more dislocating than the company press releases admitted.

What’s coming next

Three things are on the medium-term horizon:

Autonomous water carts and graders. These are harder than haulage trucks because they need to interact with the surface they’re working on, not just travel along defined routes. Several manufacturers have working pilots, but production deployment is still 18-24 months out at scale.

Battery-electric autonomous trucks. The combination is interesting because the duty cycle of autonomous operation suits battery management, but the charging infrastructure across a Pilbara mine site is a major capital project. Expect to see paired pilots emerging in the next year.

Mixed-fleet integration. The harder operational challenge is managing autonomous and manual trucks in the same pit during transition periods. Software for safe interaction has gotten better but it’s still where most operational incidents originate.

What the smaller miners are doing

The major miners have driven the technology forward. The mid-tier operators have mostly waited. That’s starting to change as the cost per retrofit truck has come down and as proven case studies make capital approval easier.

The next wave of autonomous haulage adoption is going to come from operators with 15-30 truck fleets where the economics didn’t work before. Several Western Australian iron ore producers are now in serious planning conversations.

Whether they get the same results as the early adopters depends on whether they account for the maintenance overhead and the labour transition that the bigger operators learned the hard way. The technology works. The change management around it is what determines whether the project delivers.