Battery-electric haulage is moving from pilot fleets into larger commercial mining operations.
By Charles Pitts
Autonomous mining technology is moving from controlled pilot projects into commercial fleet planning. EACON says its autonomous haulage platform is now deployed on more than 3,500 mining trucks globally, including more than 1,500 battery-electric units. The company also reports more than 150 million kilometres of autonomous operation.
The figures are company-reported rather than an independently audited industry census. Even so, they mark an important change in the market. Mine operators are no longer assessing autonomy only as a driver-replacement technology. They are evaluating a broader production system involving electric power, fleet dispatch, charging infrastructure, safety controls, communications and workforce redesign.
The next question is whether these systems can deliver consistent performance at commercial scale across copper, gold, iron ore, coal and other critical-mineral operations.
From autonomous pilots to commercial fleets
Early autonomous mining projects typically focused on a small number of trucks operating on controlled routes. The objective was to validate localization, obstacle detection, remote supervision and automated loading or dumping.
Commercial deployment raises a more demanding set of questions:
- Can autonomous and manually operated vehicles share the same operating zone?
- Can charging be coordinated without creating production bottlenecks?
- How does the system perform in dust, rain, low light and changing road conditions?
- What happens when communications fail?
- Can existing trucks be retrofitted without extended production interruptions?
- Does the mine have the technical workforce to maintain sensors, drive-by-wire systems and high-voltage equipment?
EACON’s ORCASTRA autonomous haulage platform is designed for both retrofit and factory-fit deployment. The company says it supports diesel, hybrid-electric and battery-electric trucks, with fleet management, autonomous driving and coordination with manually operated equipment integrated into the same system.
That mixed-fleet capability is significant. Most mines will not replace every haul truck at once. Operators are more likely to introduce autonomy in phases, with different truck models and powertrains working within defined autonomous operating zones.
Commercial autonomy must work alongside existing equipment, traffic rules and mine infrastructure.
EACON’s electric fleet changes the scale equation
EACON’s reported deployment of more than 1,500 battery-electric autonomous trucks suggests that electrification and autonomy are increasingly being planned together.
Battery-electric haulage adds constraints that diesel fleets can often absorb more easily. Operators must manage:
- Charger availability and queue times.
- Grid capacity and substation design.
- Battery state of charge across changing haul profiles.
- Regenerative braking and downhill energy recovery.
- Charging schedules during shift changes.
- Emergency response for high-voltage equipment and battery incidents.
Autonomy can help coordinate these variables. EACON says its fleet-management system can incorporate vehicle status, production tasks and charging requirements into dispatch decisions. In a commercial operation, the value is not simply that a truck can drive without a cab operator. The value comes from coordinating haulage and energy use so that trucks remain productive.
EACON has also promoted retrofit deployment as a way to lower adoption barriers. Retrofitting existing equipment can reduce the need for a complete fleet replacement, but it introduces engineering challenges. Different trucks have different braking systems, control architectures, payload characteristics and maintenance histories.
The cost comparison therefore needs to include more than the price of an autonomous kit. Mine owners must assess road upgrades, private wireless networks, control-room capacity, charging infrastructure, spare parts, training and software support.
The technology stack is becoming mine-wide
Autonomous haulage depends on a layered technology stack. Vehicle-level systems use combinations of LiDAR, radar, cameras, inertial measurement units, satellite positioning and high-precision mapping. Fleet platforms assign tasks, manage traffic and monitor equipment health.
EACON describes ORCASTRA as a system built around three elements:
- CONDUCTOR, which manages production targets, dispatch, maps and system health.
- PILOT, the truck-level autonomous driving kit.
- CREW, a collaboration and safety system for manually operated equipment entering an autonomous zone.
The architecture reflects a practical reality: autonomy is not deployed into an empty mine. Water carts, graders, excavators, light vehicles, maintenance teams and contractors continue to operate around the haul fleet.
For that reason, the most important risks often sit at the interfaces. A truck may perform reliably on a mapped route but lose productive time because of poor road maintenance, a network gap, an unplanned manual vehicle entry or a charging bottleneck.
The commercial test is therefore system uptime, not the performance of an isolated truck.
Epiroc connects underground and surface haulage
Surface haulage is not the only area moving toward broader autonomy. Epiroc is extending its Deep Automation platform across more complex underground routes.
In a 2026 announcement, Epiroc described autonomous truck haulage brought into three dimensions, using 3D LiDAR to support fleet control across multi-level ramps. The company’s official announcement says the technology is intended to improve localization, mapping and obstacle detection where GPS is unavailable or unreliable.
The underground-to-surface transition is operationally difficult. Portals and ramps can combine steep gradients, changing light, narrow roads, fixed infrastructure, pedestrians and manually operated service vehicles.
A continuous autonomous haul cycle could reduce the manual handover between underground workings and surface dumping areas. But the potential benefit depends on site readiness:
- Accurate and regularly updated 3D maps.
- Reliable communications through ramps and portals.
- Clear separation between autonomous and manual traffic.
- Defined procedures for maintenance access and emergency stopping.
- Consistent road edges, berms and ramp geometry.
Epiroc’s technology demonstrates a pathway toward continuous autonomy, but a demonstration is not the same as broad commercial deployment. The key performance measures will be intervention frequency, route availability, cycle-time consistency and safe recovery from degraded conditions.
Underground-to-surface autonomy must maintain localization and safety through changing operating environments.
Hexagon adds an intervention layer
Autonomous haulage and automated safety intervention are related but separate functions. Hexagon’s Vehicle Intervention System is designed to intervene when a vehicle enters a defined high-risk condition and warnings do not produce an adequate response.
The system is positioned within the EMESRT framework as a Level 9 intervention solution. Depending on the application, it can inhibit propulsion, control speed, apply braking or bring a vehicle to a controlled stop.
That role is particularly important as mines move toward mixed fleets. A safety layer may need to protect autonomous trucks, manually operated haul trucks and other mobile equipment within the same operating environment.
Hexagon has described the system as part of a broader MineProtect safety portfolio that includes collision avoidance, operator alertness monitoring and personal protection systems. The intervention system is not a replacement for safe road design or traffic management. It is a last line of defense that must be tested against site-specific vehicle types, response times, traffic rules and regulatory requirements.
For operators, the investment question is whether safety systems are integrated into the autonomy architecture rather than installed as disconnected products.
Commercial deployment tracker
The following table separates reported commercial scale from demonstrated or concept-stage technology.
| Company or system | Reported capability | Maturity | Main implication |
|---|---|---|---|
| EACON ORCASTRA | More than 3,500 autonomous trucks | Commercial fleet deployment | Autonomy is moving into multi-site fleet planning |
| EACON battery-electric fleet | More than 1,500 autonomous electric units | Company-reported commercial scale | Charging and energy management become production variables |
| EACON autonomous operating history | More than 150 million km | Company-reported operating base | Larger datasets can support software and safety refinement |
| Epiroc Deep Automation | Underground-to-surface truck control using 3D LiDAR | Demonstrated technology | Portals and ramps can become part of one autonomous cycle |
| Hexagon Vehicle Intervention System | EMESRT Level 9 automated intervention | Deployed safety technology | Propulsion and braking can be controlled in defined hazards |
| Sandvik Sami | Cabinless battery-electric autonomous drill concept | Concept stage | Points toward wider mine-wide robotics integration |
Base, bull and bear framework
| Scenario | Costs | Safety outcome | Deployment risks |
|---|---|---|---|
| Base case | Capital spending remains significant for networks, control rooms, charging and retrofit equipment, but fleet scale improves unit economics | Lower exposure to truck-cabin hazards, with new high-voltage and remote-intervention risks | Mixed fleets, road quality, workforce transition and communications reliability slow expansion |
| Bull case | Standardized platforms, better batteries and shared infrastructure reduce cost per autonomous tonne | Integrated intervention, traffic management and remote operations produce measurable incident reduction | Regulatory acceptance and technical skills keep pace with rapid deployment |
| Bear case | Small or irregular operations cannot spread infrastructure and support costs across enough trucks | Safety benefits are limited if manual vehicles frequently enter autonomous zones | Network outages, battery incidents, cybersecurity failures or poor retrofit performance undermine confidence |
What decision-makers should measure
Fleet announcements provide evidence of market direction, but operators need site-level performance data. The most useful indicators include:
- Autonomous availability and productive hours.
- Tonnes moved per operating hour.
- Intervention and emergency-stop frequency.
- Charging queue time and energy use per tonne.
- Communications coverage, latency and failover performance.
- Sensor reliability in dust, rain and low light.
- Maintenance time for autonomous and high-voltage systems.
- Cycle-time variation across loading, hauling and dumping.
- Workforce training completion and role-transition outcomes.
Autonomous mining technology in 2026 is reaching commercial scale first in haulage, where repetitive routes and large fleets make the economics easier to justify. The next phase will depend on integration.
EACON’s reported electric fleet demonstrates the scale now being pursued. Epiroc’s underground-to-surface work addresses continuity across mine environments. Hexagon’s intervention system adds a safety layer for autonomous and manually operated equipment.
The central issue for mine owners is no longer whether a truck can drive itself. It is whether the full operating system : roads, power, communications, dispatch, maintenance, safety and people : can support reliable production.
Related reading: Skillings mining technology coverage and Autonomous mining technology: fleets, milestones and outlook.
