GPS Antenna for Autonomous Mining Trucks: Positioning Accuracy, Signal Reliability, and Integration

Why the antenna, not the receiver, is usually the weakest link in a pit

A GPS antenna for autonomous mining trucks has to survive conditions that no road-vehicle antenna ever sees: blast vibration, abrasive dust, a 60-degree Celsius surface temperature swing, and a pit geometry that blocks half the sky. The receiver can be the most capable RTK engine on the market and still deliver a useless position if the antenna cannot hold a clean phase measurement under those conditions. That is why antenna selection on a haul truck is an engineering decision, not a procurement afterthought.

This guide covers what actually determines positioning performance on an autonomous haul truck, where the antenna should be mounted and why, which specifications separate a mining-grade part from a general-purpose one, and how to integrate it so the system keeps working after the first rainy season.

Why Mining Autonomy Is Harder on GNSS Than Road Autonomy

An open-pit mine is an RF environment designed by accident to break satellite navigation. The truck operates at the bottom of a terraced excavation, so the visible sky is a narrow cone. Benches and high walls reflect signals that arrive a few tens of nanoseconds late, which is enough to corrupt a carrier-phase measurement. At the same time, the vehicle is surrounded by other trucks, excavators, and light vehicles whose bodies reflect yet more energy.

Add the operational realities. Haul trucks work around the clock, so the antenna is exposed to thermal cycling every day. Dust coats every horizontal surface within hours and changes the dielectric loading on a radome if moisture binds to it. And the vehicle itself is a moving source of vibration across a wide spectrum, from engine idle to the impact of a full load dropped into the tray.

A road-going autonomous car faces multipath in urban canyons. A mining truck faces multipath in a canyon that moves, with walls that change height as the pit is cut. The antenna has to deliver a phase-stable measurement in that environment for years, not for the length of a demo.

Where the Antenna Goes on a Haul Truck

Placement decides more of the outcome than most specification tables suggest. The antenna needs an unobstructed view of the sky from horizon to horizon on the side facing the pit wall, a ground plane large enough to support the radiation pattern, and a location that does not put the radiating element inside the shadow of the tray, the ROPS structure, or an air-conditioning unit.

• Roof centre beats roof edge. A centred antenna on a flat metal roof uses the roof as a ground plane and keeps the pattern symmetrical. Moving it to the edge improves the view in one direction and degrades it in the opposite direction, which is exactly the wrong trade in a pit where the truck turns constantly.
• Keep clear of the tray and exhaust. A raised tray blocks the rear hemisphere entirely when loaded. Anything mounted behind the cab on the tray side will lose satellites every time the truck tips. Heat from the exhaust also shifts the antenna's electrical behaviour if the radome is close enough to absorb radiant energy.
• Separate the two antennas for heading. Dual-antenna heading depends on a baseline. A one-metre baseline gives coarse heading; a three- to four-metre baseline across the cab roof gives a usable result but demands a rigid mounting structure, because any flex in the roof translates directly into heading error.
• Protect the cable run. The route from antenna to receiver passes close to high-current cables, inverters, and hydraulic lines. Route the coaxial cable separately, use a double-shielded assembly, and bond the shield at the receiver end only.

Signal Reliability: The Three Failure Modes That Matter

Field failures on mining autonomy usually fall into three categories. Understanding them makes the antenna specification much easier to write.

Multipath from high walls and benches

Low-elevation signals are the main carrier of multipath in a pit. An antenna with strong low-elevation gain sees more satellites but also more reflections. The resolution is not to suppress low-elevation gain blindly, but to use an element and ground plane combination that keeps the phase centre stable when the incoming signal is shallow. A choke structure or a multi-feed symmetric element does this by controlling surface currents on the ground plane.

Dust, mud, and radome degradation

A dust layer is not just cosmetic. Wet dust forms a lossy film that detunes the antenna and raises the noise floor. A radome that has been sandblasted for two years becomes rough, retains water, and changes its dielectric properties. Choosing a UV-stabilised ABS or cast-aluminium housing with a smooth, non-porous finish matters more in a pit than in a city.

Mechanical and electrical wear

Vibration loosens connectors before it breaks housings. A TNC or N-type connector that is not properly torque-set will develop intermittent contact, and intermittent contact on a GNSS feed produces exactly the kind of cycle-slip pattern that looks like a software fault. Strain relief on the cable, thread-locking on the mount, and a connector with a defined mating torque all belong in the installation procedure.

What Accuracy Does an Autonomous Haul Truck Actually Need

The answer depends on which function consumes the position. A truck travelling along a haul road at speed needs lane-level lateral accuracy and a reliable heading to stay on the designed path. A truck spotting at a shovel needs decimetre-level repeatability so the tray lands where the loading unit expects it. A truck dumping at a crusher or stockpile needs a repeatable stopping point but not absolute accuracy to the centimetre.

A fleet that mixes antenna models across trucks will show position disagreement at the shovel even when each truck is individually within tolerance. Standardising the antenna and its mounting geometry across the fleet removes that variable from the commissioning process.

Integration Requirements Beyond the Datasheet

The specification sheet stops at the connector. The installation starts there.

A 24-hour static test is the single most useful acceptance step. It reveals multipath patterns that repeat daily as the sun moves and the temperature changes, and it exposes intermittent connector faults that a five-minute check will miss. Sites that skip it usually rediscover the same problems during production, when a stopped truck costs far more than a stopped test.

Yonghao Innovation GPS and RTK Antenna Range

Yonghao Innovation Tech is a national high-tech enterprise established in 2014, based in Shenzhen, Guangdong, China. We integrate research, manufacturing, and sales, and we supply GNSS satellite antennas, positioning terminals, data communication products, and customized high-precision BeiDou solutions for surveying and monitoring, aerospace, communication time service, autonomous driving, machine control, and intelligent transportation.

For vehicle and machine applications we build both GPS antenna products and full-band RTK GNSS antenna families. The YH-660 is specified for continuous operation reference stations, deformation monitoring, autonomous driving, driver training, geodetic and marine surveying, and it tolerates permanent outdoor installation. The CM100 adds a miniaturised 3D choke with sub-millimetre phase centre accuracy and IP67 sealing in a lighter housing, which suits a truck roof where mast loading and wind vibration matter.

Mounting hardware, connector type, and cable length are the three items customers most often need changed for a haul truck. Our product customization team reviews those requests against the existing mechanical platform before any tooling commitment, and our autonomous vehicle projects page shows how the antenna and receiver combination is normally configured.

Conclusion

Selecting a GPS antenna for autonomous mining trucks comes down to matching three things: the pit geometry the truck will work in, the mechanical environment the roof will impose, and the level of phase stability the autonomy stack actually consumes. Antennas that perform well on a highway can fail in a pit not because they are poorly made, but because they were never specified for low-elevation multipath, sustained vibration, and abrasive dust at the same time. Define those three requirements first, then compare candidates on documented phase centre behaviour rather than on catalogue gain figures.