GPS vs Indoor Tracking

Almost everyone reaches for GPS first. It is the tracking technology we all carry in our pockets, it works everywhere we can see the sky, and it costs nothing to use, so the instinct when someone asks "where is that pallet?" is to assume GPS will answer. Then the pallet moves indoors, the signal collapses, and the whole plan falls apart. Understanding exactly why that happens, and what takes over when it does, is one of the more useful pieces of practical knowledge in warehouse automation. This guide is the honest, practitioner's version, and it sits underneath the broader warehouse automation complete guide pillar, which frames where positioning fits alongside identification, robotics and the WMS.

The message up front: GPS and indoor positioning are not competitors, they are two halves of one continuous tracking problem. GPS owns everything with a clear view of the sky. Indoor systems own everything under a roof. The engineering skill is not choosing one over the other, it is knowing exactly where the handover happens and building an operation that tracks an asset cleanly across that boundary.

1. Why GPS stops at the door

The single most important thing to understand about GPS is that it is a line-of-sight technology that depends on receiving faint radio signals from satellites roughly twenty thousand kilometres above your head. A GPS receiver does not transmit anything. It listens for the timing signals broadcast by a constellation of satellites, measures how long each signal took to arrive, and from those tiny time differences it computes its own position. To get a reliable fix it needs a clean, direct view of at least four satellites simultaneously.

A warehouse roof ends that arrangement instantly. The concrete, steel and metal cladding of an industrial building attenuate the satellite signal so severely that a receiver inside typically cannot lock onto enough satellites to compute anything usable. Where a fix does survive, it is corrupted by multipath, the effect where the weak signal that does penetrate bounces off racking, steel structure and machinery before reaching the receiver, so the timing is wrong and the computed position drifts by tens of metres. In a facility where you need to know which aisle and which rack level an asset is on, a position that is uncertain by tens of metres is not a degraded answer, it is no answer at all.

So GPS does not gently get worse indoors, it falls off a cliff at the building line. Outside, in the yard, with sky overhead, it delivers a few metres of accuracy for free. Inside, it delivers nothing you can act on. That hard boundary at the door is the entire reason a separate category of indoor positioning technology had to be invented, and it is why any serious warehouse tracking design has to plan explicitly for the handover rather than pretend one technology covers the whole site.

2. How GPS works (and its indoor limits)

It is worth slowing down on the mechanism, because once you see how GPS computes a position you immediately understand why a roof defeats it. Each satellite continuously broadcasts its own identity, its precise orbital position, and an extremely accurate timestamp from an onboard atomic clock. The receiver picks up several of these broadcasts, and for each one it calculates the distance to that satellite from the travel time of the signal. Distance from one satellite places you somewhere on a sphere. Distance from a second narrows it to a circle, a third to a pair of points, and a fourth resolves the ambiguity and corrects the receiver's own clock error. This geometric method is called trilateration.

The diagram below shows the essential picture: GPS working cleanly in the open yard, and the same signal failing to reach a receiver once it is under the warehouse roof, where indoor technologies take over.

Everything about that mechanism assumes an unobstructed path to the sky. The signals arriving at ground level are already extraordinarily weak, on the order of the power of a car headlight seen from the far side of a continent, which is why the receiver needs a clean view to detect them at all. Put a metal-clad roof in the path and the signal is simply gone. This is not a limitation that better receivers or a firmware update will fix, it is physics. Assisted GPS and modern multi-constellation chips help at the fringes, under light tree cover or near a window, but none of them recover a usable fix in the middle of a steel warehouse. For indoor tracking you need a technology that does not rely on distant satellites, and that is exactly what the next category provides.

3. Indoor positioning technologies (BLE, UWB, RFID, Wi-Fi)

Indoor positioning solves the problem GPS cannot by flipping the geometry. Instead of listening to satellites far above, an indoor system uses a network of reference points installed inside the building, anchors, beacons, readers or access points, whose positions are known exactly. A tag on the asset then measures its relationship to those local reference points, and the position is computed from the local network. Because the reference points are close and under your control, the accuracy can be far better than GPS ever achieves outdoors. Four technologies dominate warehouse work.

• Bluetooth Low Energy (BLE): small, cheap battery-powered beacons broadcast a signal whose strength falls off with distance. Readers or tags estimate proximity from that signal strength. BLE is inexpensive, easy to deploy, and runs for years on a coin cell, which makes it the workhorse for zone-level tracking, "this asset is somewhere in receiving," rather than exact coordinates. Typical real-world accuracy is a few metres. See the deeper treatment in the BLE asset tracking guide.
• Ultra-Wideband (UWB): the precision option. UWB measures the actual time it takes a radio pulse to travel between tag and anchor, which yields distance directly and accurately, delivering positioning down to tens of centimetres. It resists the multipath problem that plagues signal-strength methods, so it is the technology of choice when you genuinely need to know which specific pallet position or which side of an aisle an asset occupies. It costs more in anchors and tags. The UWB indoor positioning guide covers this in depth.
• RFID: not a coordinate system but a presence-and-passage system. Passive RFID tags cost pennies and carry no battery; they are energised by a reader as they pass through a portal, a dock door, or a handheld scan. RFID tells you an item was read at a known reader location at a known time, which for most inventory movement is exactly the answer you need. It is the backbone of item-level identification rather than continuous positioning. The RFID in warehouse management guide details the read models.
• Wi-Fi: uses the access points a facility already has, estimating position from the signal strength of tags against known AP locations. Its appeal is that it reuses existing infrastructure, so the incremental cost is low. Its weakness is accuracy, usually several metres and sensitive to how the network was laid out, since APs were positioned for coverage, not for positioning geometry. It is a reasonable fit for coarse zone tracking when a Wi-Fi network is already dense.

The practitioner's point is that these are not ranked from worst to best, they are matched to what you need to know. If the question is "is this cage in the building," BLE or Wi-Fi answers it cheaply. If the question is "which of these forty rack positions holds the tote," only UWB gets you there. If the question is "did this carton pass the dispatch door," RFID answers it for pennies per tag. Choosing the wrong technology for the question is how indoor positioning projects overspend or underdeliver.

4. GPS versus indoor tracking

Laid side by side, the division of labour becomes obvious. GPS is unbeatable outdoors and unusable indoors; the indoor technologies are the mirror image, strong inside and irrelevant to the open yard. The table compares them across the dimensions that actually drive a design decision.

Read the table as a division of territory rather than a contest. The "environment" column is doing most of the work: nothing indoor competes with GPS outside, and GPS does not appear in the indoor rows at all because it simply is not an option there. Every real design ends up combining a row from the outdoor world with one or more rows from the indoor world, which is precisely the integration problem the rest of this guide addresses.

5. Outdoor yard versus indoor warehouse

The cleanest way to think about a warehouse site is as two zones with a doorway between them, each governed by its own physics. The outdoor yard is GPS territory. Trucks arriving and departing, trailers parked in bays, containers staged before unloading, yard tractors shuffling loads: all of these live under open sky where a cheap GPS receiver gives a few metres of accuracy for nothing. Yard management systems, telematics on the vehicles, and geofences around gates and bays are all built on this free outdoor signal, and there is rarely a good reason to reach for anything else outside.

The indoor warehouse is the mirror image. The moment a load crosses the dock door, GPS is gone and one of the indoor technologies must take over: RFID reading the pallet through the door portal, BLE tracking which zone a cage sits in, UWB pinpointing a tote to a specific rack position, or Wi-Fi giving coarse coverage off the existing network. The accuracy you can achieve indoors is often far higher than outdoors, because the reference points are close and under your control, but you pay for it in installed infrastructure that the free satellite signal never required.

The interesting and error-prone part is the dock door itself, the seam between the two worlds. This is where an asset leaves satellite coverage and enters beacon coverage, and it is where tracking systems most often lose the thread, reporting an asset as "last seen in the yard" long after it has been put away inside, or vice versa. Designing that transition deliberately, usually by making the dock door an explicit RFID or UWB read point that hands the asset from the outdoor system to the indoor one, is the difference between a tracking system that tells a continuous story and one that has a blind spot exactly where the most valuable events happen.

The honest limitation: no single technology covers a warehouse site end to end, and any vendor promising one that does is either redefining "cover" very loosely or hiding a second technology inside the box. GPS will not work indoors and indoor anchors will not track a truck on the motorway. Accept the seam, design for it, and you get a clean continuous track. Pretend it is not there and you get a system that quietly loses assets at the door.

6. Combining outdoor and indoor tracking

Because no one technology spans the whole site, real deployments are hybrid by necessity, and the design question becomes how to stitch the pieces into one continuous view of where everything is. The pattern that works is to treat position as a layered service: an outdoor layer, an indoor layer, and a handover mechanism between them, all feeding a single system of record.

A typical layered architecture for a warehouse and yard looks like this:

The layer organisations consistently underinvest in is the integration layer, the part that reconciles an outdoor GPS coordinate and an indoor beacon zone into one coherent location for the same asset. A GPS fix speaks in latitude and longitude; an indoor system speaks in zones or local coordinates; RFID speaks in read events. Something has to translate all three into the vocabulary the warehouse management system understands so that "location" means one thing across the whole site. This is fundamentally an integration challenge, the same kind of operational-technology-to-enterprise-IT bridging that shows up across telematics, sensors and building systems, and it is where I have seen far more warehouse tracking projects stumble than on any radio or accuracy problem. The tags worked and the anchors worked; nobody built the layer that made the two halves agree on where things were. For how positioning data lands in the system that runs the floor, see what is a WMS.

7. Choosing the right approach

The mistake I see most often is starting from the technology instead of the question. Someone reads about UWB delivering centimetre accuracy and decides the whole warehouse needs it, then discovers the anchor and tag bill for a large facility is enormous and most of that precision is tracking things that only ever needed to be located to the nearest zone. Start instead from what you actually need to know about each class of asset, and let that pick the technology.

• Is the asset ever outdoors? If it moves in the yard or in transit, it needs GPS for that phase, full stop. Nothing indoor helps there and GPS is effectively free.
• How precisely do you need the indoor position? If "which zone" is enough, BLE or Wi-Fi is the economical answer. If you genuinely need "which exact position," UWB is the only technology that reliably delivers it, and you should be sure the value justifies the cost.
• Do you need continuous position or just passage? If you only need to know an item entered or left an area, RFID portals answer that for pennies per tag and you do not need a positioning system at all.
• What infrastructure already exists? A dense, well-placed Wi-Fi network can give coarse indoor tracking at almost no incremental hardware cost, which can be the right first step before committing to dedicated anchors.
• How many assets and how valuable are they? Cheap, high-volume items suit cheap passive RFID; a small number of high-value or safety-critical assets can justify UWB tags and anchors.

Run that questioning across the asset classes on a real site and the answer is almost never one technology. It is GPS in the yard, RFID at the doors and for item-level inventory, BLE for zone-level tracking of reusable containers and equipment, and UWB reserved for the specific high-value or high-precision cases that genuinely need it. That concentration is not indecision, it is the correct outcome: each technology applied exactly where its strengths pay and its costs are justified, stitched together by an integration layer into one continuous location picture. The warehouse automation complete guide places this positioning stack alongside the identification, robotics and WMS decisions it has to line up with.

8. References

The technical claims in this guide rest on the published standards and documentation behind each technology. Useful primary and reference sources for going deeper:

• GPS.gov, the official U.S. government information site on the Global Positioning System, on how GPS works, accuracy, and the limits of satellite signal reception indoors.
• The Bluetooth SIG specifications and direction-finding documentation for Bluetooth Low Energy, covering beacon operation and signal-strength-based proximity.
• The FiRa Consortium and IEEE 802.15.4 / 802.15.4z material on Ultra-Wideband ranging and time-of-flight positioning.
• GS1 standards documentation on RFID (EPC UHF Gen2) for item-level identification, portal reads and supply-chain use.
• The IEEE 802.11 family documentation and vendor location-services references for Wi-Fi-based indoor positioning.

Alongside those standards, the companion guides on this site go one level deeper into each indoor technology: UWB indoor positioning, BLE asset tracking and RFID in warehouse management.

Final thoughts

GPS and indoor positioning are not rivals to be judged against each other, they are two halves of a single tracking problem separated by a roof. GPS owns the open yard and the road, where it works for free with a few metres of accuracy and nothing else comes close. Indoor technologies own everything under the roof, where GPS is physically dead and BLE, UWB, RFID and Wi-Fi each answer a different question at a different price. The engineering that matters is not picking a winner, it is knowing exactly where the handover happens and building an operation that carries an asset cleanly across the dock door from one world into the other.

If you are designing warehouse or yard tracking, the most useful discipline is to start from the question, not the catalogue. Decide what you need to know about each asset, indoors and out, let that choose the technology for each phase, and then invest properly in the integration layer that makes the outdoor and indoor pictures agree. Do that and you get a continuous, trustworthy view of where everything is across the whole site. Skip it, reach for one technology and hope it stretches to cover both worlds, and you get a tracking system with a blind spot at exactly the door where the most important events happen.

Related reading: Warehouse automation complete guide, UWB indoor positioning, BLE asset tracking, RFID in warehouse management, What is a WMS.