Autonomous haulage systems have moved well beyond pilot programmes. Caterpillar’s Command for hauling now operates across more than 690 trucks globally, with a stated target of exceeding 2,000 by 2030. Komatsu’s FrontRunner fleet has hauled over 3.5 billion tonnes across nine sites. In Australia, Fortescue’s Solomon and Chichester operations run fully autonomous fleets, Roy Hill is converting 96 trucks to driverless operation, and Rio Tinto’s Pilbara mines have reported autonomous trucks operating approximately 700 hours more per year than their conventional equivalents, delivering 15% lower load-and-haul unit costs.
The productivity and safety case for AHS is well established: 15–30% productivity improvements, 40% improvement in tyre and brake life, 20% reduction in operational costs, and zero lost-time injuries attributed to autonomous haulage across hundreds of millions of kilometres driven. But these figures assume the autonomous operating zone (AOZ) is maintained to a standard that allows the system to function as designed.
In practice, dust is one of the most persistent and consequential operational problems facing AHS deployments. Unlike a human operator who can make a judgement call and drive through a dust plume at reduced speed, an autonomous truck’s safety systems respond to sensor degradation by imposing speed restrictions or triggering a full stop event. Every such event cascades through the fleet, creating knock-on stoppages, cycle time blowouts, and production losses that erode the efficiency gains AHS was deployed to deliver.
This article examines the specific mechanisms by which dust degrades AHS performance, why conventional water-based dust control creates secondary problems in autonomous environments, and how engineering-based suppression strategies can be integrated into AHS operations to reduce stoppages and maintain design-level productivity.
How Dust Degrades AHS Sensor Performance
Autonomous haul trucks navigate using a layered perception stack: high-precision GNSS for positioning, LiDAR for 3D spatial mapping and obstacle detection, radar for object tracking, and in some configurations, camera-based vision systems. Each of these sensor modalities is affected by airborne dust, but the impact on LiDAR is particularly significant.
Are environmental regulations, health and safety concerns or potential profit loss a concern right now?
LiDAR and Point Cloud Contamination
LiDAR operates by emitting laser pulses and measuring their return time to construct a three-dimensional point cloud of the surrounding environment. When an autonomous truck drives into or generates a dust plume, suspended particles scatter and reflect the laser pulses before they reach their intended targets. The result is noise injection into the point cloud: the system registers thousands of false return points in the space between the sensor and the actual ground surface, road boundaries, or obstacles.
This noise has two operational consequences. First, the perception algorithm may interpret the dust cloud itself as a physical obstacle, triggering a stop event. Second, even where the system does not classify dust as an obstacle, the degraded point cloud reduces the confidence with which the navigation system can identify the driveable surface. When confidence drops below a threshold, the truck’s safety systems impose speed restrictions or request operator intervention.
Sensor manufacturers are responding with hardware and software countermeasures. Larger optical apertures (as used in Ouster’s digital LiDAR) improve resilience to particulate obscurants. Deep-learning dust filtering algorithms—such as those developed by TAGE for Chinese open-pit operations—train neural networks on large point cloud datasets to distinguish dust returns from solid objects. Self-cleaning optical shields, like ExcelSense’s LidarShield, address the related problem of dust accumulation on the sensor housing itself. But all of these are compensatory measures. They improve the system’s tolerance for dust; they do not eliminate the underlying problem.
Radar, Camera and Communication Interference
Radar is inherently more resilient to dust than LiDAR, as the longer wavelengths used by automotive-grade radar pass through fine particulate matter more readily. However, in dense dust conditions—particularly in the immediate vicinity of loading and dumping zones—radar returns can still be affected, contributing to false proximity alerts that trigger protective slowdowns.
Camera-based systems, where fitted, suffer the most obvious degradation: reduced visibility. While cameras play a secondary role in most AHS perception stacks, they are increasingly used for object classification (distinguishing a person from a bollard, for example) and for operator review of stop events via remote operations centres. Dense dust reduces the quality of these images, complicating and delaying the remote clearance process.
Communication systems are less directly affected by dust, but dust-induced stoppages create network congestion as multiple trucks simultaneously request remote intervention, consuming control room bandwidth and extending resolution times.
The Five Haul Road Factors Behind AHS Slow and Stop Events
Dust is not the only haul road condition that triggers autonomous truck stoppages, but it is the one most directly linked to the other four. In an autonomous operating zone, haul road maintenance issues compound through well-documented interaction effects.
| Factor | Mechanism | AHS Impact |
| Dust | Airborne particulate degrades LiDAR/radar returns and reduces camera visibility | Speed restrictions imposed; full stop events triggered when sensor confidence drops below threshold |
| Proximity | Water trucks or graders operating within the AOZ are detected as obstacles by AHS safety systems | Trucks halt until ancillary equipment clears the operating bubble; entire fleet affected on shared haul roads |
| Loss of traction | Over-watering of haul roads creates slippery surfaces that AHS traction control systems detect | Trucks reduce speed or stop to prevent wheel slip; particularly acute on ramps and inclines |
| Rutting | Degraded road surface creates depressions and irregularities detected by LiDAR as obstacles | Perception system classifies ruts as hazards; trucks stop until road is graded or manually cleared |
| Knock-on | A stopped truck creates a physical obstacle for following trucks in the fleet | Single stop event cascades through multiple trucks; production loss is multiplicative, not additive |
The critical observation for mine engineers is that three of these five factors—proximity, loss of traction and rutting—are directly caused or worsened by conventional water-based dust suppression. Water trucks must enter the AOZ to spray haul roads, triggering proximity events. Over-application of water creates the traction loss that AHS systems detect and respond to. And the wet-dry cycling that water-only approaches produce accelerates road surface degradation, creating the ruts and potholes that LiDAR interprets as obstacles.
In an autonomous environment, the standard water-cart-and-grader approach to haul road maintenance is not merely inefficient—it is actively counterproductive. Every water truck pass introduces a manned vehicle into the AOZ, generating proximity stoppages that directly reduce truck utilisation. Case study data from operations transitioning to AHS have documented 32% reductions in stoppages simply by identifying and addressing these haul road maintenance interactions.
Why Water-Only Dust Suppression Fails in Autonomous Operating Zones
Water-only dust suppression has well-documented limitations in any mining environment—rapid evaporation in arid conditions, high application frequency, enormous water consumption, and the inability to bind fine particles below PM10 effectively. But in an AOZ, these limitations interact with the AHS safety architecture to create a compounding problem.
Frequency of entry. In hot, dry conditions typical of Australian open-pit operations, water-only suppression may require re-application every 30–60 minutes to maintain effective dust control. Each water truck pass is a proximity event that stops autonomous trucks in the vicinity. On a high-traffic haul road, this can generate dozens of stop events per shift.
Inconsistent coverage. Water application is inherently variable. Even experienced operators produce uneven coverage—some sections over-watered (creating traction issues), others under-watered (allowing dust to regenerate). Autonomous trucks running the same line on every cycle are particularly sensitive to these variations, as they repeatedly traverse the same wheel paths.
Road surface degradation. The wet-dry cycling from frequent water application washes fines from the road surface, destroys the binding matrix of compacted material, and accelerates the formation of corrugations, potholes and ruts. In an AHS environment, where trucks run identical lines with minimal lateral variation, the damage concentrates in narrow wheel paths. These deformations are detected by LiDAR as obstacles, generating further stoppages and increasing grader frequency—which introduces yet another manned vehicle into the AOZ.
The net effect is a negative feedback loop: poor dust control triggers stoppages, the remedial action (more water, more grading) triggers further stoppages, and the road surface deteriorates faster, requiring even more frequent intervention. Breaking this cycle requires a fundamentally different approach to haul road dust management.
Engineering Dust Control for AHS Environments
An effective dust management strategy for autonomous operations must achieve three objectives simultaneously: suppress airborne particulate to a level that maintains sensor confidence, minimise the frequency of manned vehicle entry into the AOZ, and preserve the structural integrity of the road surface. Polymer-based suppression and stabilisation systems address all three.
Polymer Dust Suppression: Extending the Re-Application Window
Polymer-based dust suppressants work by binding fine particles into a cohesive surface matrix that resists disturbance from tyre contact and wind. Unlike water, which evaporates and leaves the road surface unprotected within an hour in typical Australian conditions, polymer treatments extend effective suppression to days or weeks depending on traffic volume and environmental conditions.
The operational consequence for AHS is significant. Extending the re-application interval from 30–60 minutes to several days reduces the number of water truck entries into the AOZ by an order of magnitude. Fewer entries mean fewer proximity events, fewer traction events from over-watering, and fewer grader cycles needed to repair water-induced road damage. The polymer also retains the fine fraction within the road surface rather than allowing it to become airborne, which directly reduces the PM10 loading that degrades LiDAR performance.
Haul Road Stabilisation: Eliminating the Rut and Obstacle Problem
Autonomous trucks run the same optimised line on every pass. In a conventional mining environment, this is a benefit—it allows road designers to optimise geometry for the exact wheel path. But it also means that any road surface weakness in the wheel path is subjected to concentrated, repetitive loading without the lateral variation that distributes wear across a wider surface in manned operations.
Soil stabilisation products improve the structural integrity of the road by increasing the shear strength, bearing capacity and cohesion of the compacted surface material. A stabilised haul road resists rutting, corrugation and pothole formation under the repetitive, concentrated loading patterns characteristic of AHS traffic. The result is a surface that LiDAR consistently reads as a smooth, driveable plane—eliminating the false obstacle detections that generate stop events from road deformation.
The dual application of polymer dust suppression and road stabilisation creates a compounding benefit in an AHS context: the polymer prevents the loss of fines that leads to surface degradation, and the stabiliser prevents the structural deformation that LiDAR interprets as obstacles. Together, they maintain a consistent, sensor-friendly road surface with minimal intervention.
Automated Dosing and Integration with Mine Control Systems
The value of polymer suppression in an AHS environment is amplified when delivery is automated. SMART dosing units meter suppressant concentrate into the water supply at a controlled ratio—typically at the standpipe, eliminating the need for manual mixing or dosing adjustments in the field. Solar-powered units are standard for remote standpipe locations where grid power is unavailable.
For AHS-integrated operations, the SMART dosing system’s telemetry layer provides real-time reporting on dosing rates, suppressant consumption and application events. This data can be fed into the mine’s integrated remote operations centre (IROC), allowing the autonomous fleet’s dispatch system to account for scheduled dust management activities in its route planning. The objective is to coordinate suppression with fleet movements rather than react to dust after it has already degraded sensor performance.
This represents a shift from reactive dust management (wait for dust, send a water truck) to predictive dust management (apply long-duration suppression before the fleet generates problematic dust levels). For sites transitioning to or operating AHS, this shift is not optional—it is a prerequisite for achieving the cycle time and utilisation targets that justify the autonomous investment.
Design Considerations for AHS Haul Road Dust Management
For mine engineers and planners designing or optimising autonomous haul road networks, the following technical considerations should inform the dust control strategy.
Narrower roads intensify the problem. One of the recognised benefits of AHS is the potential to reduce haul road widths—from the conventional 3.5× truck width to as little as 2× truck width with pull-outs. Narrower roads reduce strip ratios and waste stripping costs. However, narrower roads also mean that dust generated by one truck is concentrated in a smaller cross-section, creating denser plumes in the path of following vehicles. The dust control strategy must account for the higher effective dust density per unit road width.
Intersections and dump zones are critical. Stop-start activity at intersections, dump pads and loading zones generates the highest instantaneous dust concentrations on site. These are also the locations where AHS safety systems are most active (proximity detection during loading/dumping, intersection management logic). Engineering dust control at these pinch points—through stabilised surfaces, targeted suppression and, where appropriate, wind barriers—disproportionately reduces total fleet stoppages.
Autonomous trucks amplify road wear patterns. The precision of GPS-guided wheel paths means that tyre loading is concentrated in narrow bands rather than distributed across the full road width. Road design and maintenance programmes must account for this concentrated loading. Polymer stabilisation of the wear zone, combined with periodic reshaping (at intervals far longer than with water-only approaches), maintains the surface profile that the LiDAR system expects to see.
Real-time dust monitoring enables closed-loop control. Mounting particulate sensors at key points along the haul road—intersections, ramps, dump approaches—provides the data needed to trigger suppression activity before dust levels reach the threshold that triggers AHS speed restrictions. When integrated with the fleet management system, this creates a closed-loop control system: dust level rises, suppression is activated, dust level falls, fleet resumes normal speed.
Quantifying the Operational Impact
The productivity cost of dust-related AHS stoppages is significant and measurable. Each stop event typically takes 3–5 minutes to resolve (sensor re-acquisition, remote clearance, ramp-back to operating speed), during which the stopped truck and all trucks behind it in the queue are unproductive. On a 20-truck fleet running 30 cycles per shift, even a modest frequency of dust-related events can consume hundreds of productive hours per month.
The compounding economics are straightforward. AHS deployments are capital-intensive, with per-truck conversion costs and supporting infrastructure running into tens of millions of dollars per fleet. The entire investment case is built on utilisation rates and cycle time consistency. Every percentage point of utilisation lost to preventable dust-related stoppages represents a direct reduction in return on that investment.
Conversely, sites that implement effective long-duration dust suppression and road stabilisation report measurable improvements in AHS performance: reduced cycle time variability, fewer unplanned stoppages, lower ancillary vehicle interaction rates, extended road maintenance intervals, and improved tyre and component life. These outcomes directly support the operational efficiency and cost reduction targets that AHS is deployed to achieve.
The Integration Imperative
Autonomous haulage represents the most significant operational technology investment most mine sites will make in the coming decade. The global AHS market is projected to grow at over 8% CAGR through 2033, with Australia remaining one of the highest-adoption markets in the world. Caterpillar’s ambition to triple its autonomous fleet by 2030, combined with the expansion of retrofit and OEM-agnostic solutions, means that AHS will progressively move from tier-one iron ore operations to mid-tier mines, quarries and diversified commodity producers.
As that adoption broadens, the dust control challenge will intensify—particularly at operations in arid and semi-arid regions where water-only approaches are already marginal, and at smaller sites where the economics of AHS are more sensitive to utilisation losses.
The message for mining engineers and supervisors responsible for AHS performance is direct: dust control is not a peripheral maintenance activity in an autonomous environment. It is a core enabler of system performance. The sensor stack that keeps your fleet moving safely was not designed to operate through a permanent dust haze, and the water truck your site relies on to manage that dust is itself one of the leading causes of fleet stoppages.
Transitioning to an engineering-based approach—polymer suppression for airborne dust, stabilisation for road integrity, automated dosing for minimal AOZ intervention—is the practical path to unlocking the full productivity potential of your autonomous fleet.
Sources and References
Caterpillar Investor Day 2025: Autonomous haulage targets. International Mining, November 2025
Komatsu: FrontRunner Autonomous Haulage System
AMC Consultants: Implementing Autonomous Haulage in Open Cut Mining
FutureBridge: Autonomous Haulage Systems – The Future of Mining Operations
Mining Doc: Case of Autonomous Haulage Stoppage Reduction
Ouster: How TAGE Uses Digital LiDAR to Solve Dust and Puddles in Mines
MDPI Sustainability: Algorithm for Point Cloud Dust Filtering of LiDAR for Autonomous Vehicles in Mining Area (2024)
MDPI Drones: Enhancing Autonomous Truck Navigation in Underground Mines: 3D Object Detection Systems (2025)
ExcelSense Technologies: Self-Cleaning LiDAR Innovation for Mining Automation (2024)
GlobalData: Development of Autonomous Trucks in Global Mining Sector, 2025
Dustac: Proven Strategies for Road Dust Control on Mine Sites (2025)
Komatsu: Understanding the New ISO 23875 Air Quality Standards for Mining
