By Sonny Rollins
Autonomous mining technology 2026 is entering a more demanding phase: proving that safety systems can be standardized across large, mixed equipment fleets.
The immediate milestone is in South Africa, where Komatsu South Africa achieved Technology Readiness Level 4 certification for its collision prevention systems. The testing, reported by International Mining, verified compliance with ISO 21815, making Komatsu the first surface mining equipment OEM to meet the standard through an independently tested collision prevention pathway.
The significance extends beyond one OEM or one country. As mines combine autonomous trucks, manned equipment, light vehicles and contractors, the industry needs safety systems that can be tested, documented and audited in a consistent way. Certification is becoming the bridge between technical demonstration and fleet-scale approval.
Why the TRL4 milestone matters
Komatsu’s collision prevention and collision warning systems were tested by the University of Pretoria at Gerotek Test Facilities and in a controlled customer mine environment under the Minerals Council South Africa’s Mine Occupational Safety and Health framework.
At TRL4, the system is not simply detecting a nearby vehicle or warning an operator. It has demonstrated the ability to intervene when an imminent collision hazard is identified by inhibiting propulsion or applying the brakes.
That distinction is central to autonomous operation. Warning systems depend on human response. Collision prevention systems must connect perception, risk assessment, machine controls and braking with sufficient reliability to reduce the likelihood or severity of an impact.
ISO 21815 provides a common framework for this process. Its requirements cover collision warning and avoidance systems on earth-moving machinery and mobile underground mining equipment, including detection zones, collision risk levels, system limitations, warnings, stopping performance and functional safety.
For operators, a standard does not remove site-specific risk. A haul road still has its own grades, visibility constraints, traffic rules, weather conditions and interactions between equipment. But a recognized standard provides a common baseline for validating whether a system performs as claimed.
That can help unlock three parts of the adoption cycle:
- Regulatory approval: documented performance gives regulators a stronger basis for assessing autonomous or semi-autonomous equipment.
- Insurance and risk underwriting: independently tested safety systems can improve the quality of evidence available to insurers and risk committees.
- Fleet expansion: operators can move from isolated pilots toward repeatable deployment across more trucks, shifts and operating areas.
The safety certification ladder
The industry’s next challenge is to turn safety certification into a practical deployment framework. A useful way to assess autonomy is the safety certification ladder:
Detect → warn → intervene → validate → scale.
- Detect: Identify people, vehicles, machinery and obstacles within the relevant risk area.
- Warn: Deliver visual, audible or haptic alerts that operators can perceive under real operating conditions.
- Intervene: Reduce speed, apply braking or inhibit propulsion when the risk reaches a defined threshold.
- Validate: Demonstrate performance through controlled tests, field trials, safety cases and operational data.
- Scale: Apply the validated system across mixed fleets, changing mine conditions and longer operating windows.
TRL4 certification places Komatsu’s system at the validation stage of this ladder. It does not mean every mine has reached full autonomy. It means the underlying collision prevention function has moved beyond a conceptual or site-specific trial and into a more structured evidence base.
That distinction will matter as procurement teams compare OEM-integrated systems with retrofit platforms. The relevant question will increasingly be: What has been certified, under which conditions, and how is that evidence maintained after deployment?
Fleet scale is arriving before full standardization
The market is already large enough for standardization to become a commercial issue.
CiDi reported more than 1,900 autonomous mining trucks shipped across nearly 40 mines by June 2026. The company said seven mines were operating fleets of more than 100 autonomous trucks, while the largest single-mine deployment exceeded 220 vehicles. Those figures are company-reported and measure cumulative shipments rather than an independently audited global installed base, but they illustrate the speed at which autonomous haulage is scaling.
Komatsu has separately commissioned more than 1,000 ultra-class autonomous haul trucks through its FrontRunner system. The company’s commercial autonomy footprint includes iron ore, copper and gold operations, with the platform designed around Komatsu equipment and fleet-management systems.
At Vale Base Metals’ Salobo copper complex in Brazil, a fleet of 19 autonomous Komatsu 930E trucks has gone live. The deployment is intended to support a roughly 7% increase in material movement as Salobo III expands its coarse particle flotation capacity. The flotation project is expected to add about 6 million tonnes per year of ore-processing capacity, creating a direct link between autonomous haulage, pit planning and plant throughput.
An autonomous haulage system must be coordinated with crushers, stockpiles and processing capacity.
SANY has also shipped its first SKT110Ei pure-electric autonomous mining trucks to South America. According to the company’s PR Newswire announcement, the project combines autonomous trucks, an intelligent dispatching system, cloud-based infrastructure and lifecycle operations and maintenance services.
The deployment is notable because it combines two adoption trends: autonomy and electrification. It also tests whether Chinese equipment manufacturers can build the local service, charging and communications infrastructure required for international mine operations.
Adoption tracker
| Company or operation | System or fleet | Reported scale or status | What it indicates |
|---|---|---|---|
| Komatsu | FrontRunner autonomous haulage | 1,000+ ultra-class trucks commissioned | OEM-integrated autonomy has reached global commercial scale |
| Komatsu South Africa | Collision prevention and warning systems | TRL4; ISO 21815 compliance reported | Safety certification is becoming a deployment requirement |
| Caterpillar | MineStar Command for hauling | Nearly 700 autonomous trucks reported globally | Autonomy is expanding across mines, quarries and mixed operating environments |
| CiDi | MetaMine autonomous haulage | 1,900+ trucks across nearly 40 mines | Chinese integrators are scaling high-density fleets |
| SANY | SKT110Ei battery-electric autonomous trucks | First fleet shipped to South America | Electric autonomy is moving into new regions |
| Vale Base Metals | Komatsu 930E fleet at Salobo | 19 trucks; supports planned 7% material-movement increase | Autonomy is being tied directly to pit-to-plant debottlenecking |
| Rio Tinto | AutoHaul autonomous heavy-haul rail | Up to roughly 50 autonomous trains on a 1,700-km network | Automation can extend beyond trucks into long-distance logistics |
| BHP | Autonomous mine-site haulage | Multi-operation autonomous truck deployments | Large operators are embedding autonomy into broader operating models |
The figures are not directly comparable. Some represent commissioned trucks, others cumulative shipments, contracted units or an entire rail network. The table is therefore best read as an adoption map rather than a single global fleet total.
Mixed fleets are the real integration test
Most mines will not convert every mobile asset at once. Autonomous trucks will operate alongside manned haul trucks, graders, water carts, loaders, light vehicles and maintenance crews.
That creates a more complex safety case than a fully autonomous, tightly controlled fleet. The system must recognize different vehicle classes, account for human driving behavior and remain functional when equipment from multiple manufacturers uses different control architectures.
This is where ISO 21815-style requirements become important. Operators need evidence that collision systems can:
- detect intended objects in defined risk areas;
- classify the level of collision risk;
- provide warnings that are perceptible in the cab or control room;
- calculate stopping performance using machine speed, braking capability and system delays;
- identify system limitations and notify users when the system is unavailable;
- preserve other machine safety functions after integration.
The technical challenge is matched by a data challenge. Mine owners, regulators and insurers will increasingly demand audit-grade records covering intervention events, near misses, system availability, sensor faults, manual takeovers, route changes and maintenance history.
A dashboard showing fewer incidents is not enough. Operators will need to demonstrate how incidents were defined, how exposure changed, whether reporting rules remained consistent and whether performance held across shifts and weather conditions.
Control-room integration is becoming as important as the autonomous vehicle itself.
ESG benefits require measurable baselines
Autonomy is often presented as an ESG technology, but the claims need operational evidence.
A well-designed autonomous fleet can reduce exposure to collision hazards, fatigue, dust and vibration. More consistent haul cycles may also reduce unnecessary idling, improve payload discipline and support lower fuel consumption per tonne. When combined with battery-electric trucks, trolley assist or optimized dispatch, the emissions opportunity can become more significant.
Those benefits should be measured against a baseline rather than assumed. Useful indicators include:
| ESG or operating measure | Data required |
|---|---|
| Fuel intensity | Litres consumed per tonne moved before and after deployment |
| Emissions | Scope 1 emissions linked to haulage activity and power source |
| Safety exposure | Vehicle-kilometres, personnel exposure hours and high-potential events |
| System reliability | Autonomous operating hours, intervention rates and unplanned downtime |
| Maintenance | Tire, brake and component life across comparable routes |
| Workforce transition | Training hours, redeployment, new technical roles and safety qualifications |
Vale’s Salobo deployment shows how autonomy can be connected to a wider production and efficiency program rather than treated as a standalone technology project. The commercial test will be whether higher material movement and better consistency are achieved without creating new bottlenecks in charging, maintenance, crushing, water management or tailings handling.
Risks move into software, networks and workforce design
Autonomy does not eliminate operational risk; it changes where risk appears.
Cybersecurity is one concern. A compromised dispatch system, positioning signal or communications network could affect multiple machines at once. Mines will need network segmentation, access controls, redundant communications and tested recovery procedures.
Integration is another. Retrofitting a truck can be faster than replacing a fleet, but differences in braking systems, payload behavior, steering controls and maintenance condition can make mixed-fleet certification more difficult.
Workforce transition will also determine the pace of adoption. Driver roles may decline, while demand grows for remote supervisors, control-room technicians, sensor specialists, cybersecurity professionals, reliability engineers and safety-assurance teams. The transition will require structured retraining rather than assuming that existing roles will automatically convert into new ones.
The direction of travel is clear. Fleet scale is growing, electric autonomous equipment is entering new markets, and safety certification is moving toward the center of investment decisions.
The next competitive advantage will not belong solely to the manufacturer with the largest autonomous truck count. It will belong to the operator that can connect certified collision prevention, mixed-fleet dispatch, energy management, workforce planning and audit-grade performance data into one operating system.
For more reporting, see Skillings’ coverage of mining technology, mining operations and ESG compliance.