How to Choose Technology for Achieving High-Precision Positioning Inside Warehouses Where GNSS Does Not Reach

Inside buildings such as warehouses and factories, GNSS (GPS) satellite signals do not reach as they do outdoors, so special measures are required for position measurement. However, there are many situations where centimeter-level high-accuracy position information is required, such as inventory management, operation of automated guided vehicles, equipment inspection, and construction measurement. This article clearly explains the main positioning technologies used in indoor environments where GNSS cannot be used and logically organizes selection points according to use cases. At the end of the article, we also introduce simple surveying with LRTK as a modern solution that enables easy high-precision positioning both outdoors and indoors.

Table of Contents

• Challenges of indoor warehouse positioning

• Main technologies used for indoor warehouse positioning - IMU (Inertial Measurement Unit) - SLAM (Simultaneous Localization and Mapping) - UWB (Ultra Wideband) - BLE beacons (Bluetooth Low Energy) - QR code guidance - Other localization technologies

• Selection points for high-precision positioning technologies - Accuracy and reliability - Real-time capability - Cost and implementation hurdles - Operability and maintenance

• Simple surveying solution for indoor/outdoor with LRTK - Use of GNSS correction positioning (RTK) - Combination with indoor localization - Field application via smartphone integration - Ease of implementation and effectiveness

• Conclusion

• FAQ

Challenges of indoor warehouse positioning

When performing high-precision positioning indoors, it is first necessary to understand the environment-specific challenges. Inside buildings like warehouses, GNSS signals from satellites overhead are mostly blocked by roofs and walls and are hardly received. Even if faint signals are received, multipath (reflections) cause large errors, so outdoor-level accuracy cannot be expected. Therefore, indoors you must estimate current position by means other than GNSS.

In addition, warehouses contain tall stacked shelves and many metal machines and materials that obstruct positioning. Radio-based methods are prone to reflections, interference, and attenuation caused by metal, while optical sensor–based methods suffer when shelves or equipment block the view. In large, flat floor areas with few distinguishing features, some sensors can easily lose track of position. Furthermore, in dynamic workplaces where forklifts and AGVs/AMRs travel, real-time high-precision positioning is required to ensure safety and enable efficient control.

For cases that require centimeter-level accuracy, selecting the appropriate technology while taking these challenges into account is important. Historically, some tasks were handled by manual tape measurements or total stations (optical surveying instruments), but there is a trend toward labor-saving and higher precision through digital automated measurement. Achieving this requires combining and utilizing various positioning technologies that function even inside warehouses. In the next chapter, we will look at representative technologies used for indoor warehouse positioning and their characteristics.

Main technologies used for indoor warehouse positioning

A variety of technologies have been developed and commercialized to realize indoor positioning. Sensor types and measurement principles vary widely, each with strengths and weaknesses. Few technologies are perfect on their own; it is common to combine multiple methods for mutual compensation. Here we summarize IMU, SLAM, UWB, BLE beacons, QR code guidance, and other localization approaches used in warehouses.

IMU (Inertial Measurement Unit)

An IMU (Inertial Measurement Unit) consists of accelerometers and gyroscopes and is a device that detects its own acceleration and angular velocity to estimate movement. In inertial navigation using an IMU, if the initial position and heading are known, subsequent accelerations and rotations can be integrated sequentially to calculate the relative travel path. In warehouses, IMUs are built into smartphones carried by people and into robots, helping detect travel distance and changes in orientation in real time.

The advantage of IMUs is that they are self-contained and do not depend on external infrastructure. Even without GPS signals, an IMU alone can track changes in position, and its high update frequency makes it well-suited for real-time control. IMUs are also small, low-cost, and low-power, so they are widely used in mobile devices and small robots.

However, the biggest challenge with IMUs is error accumulation (drift). In the calculation process that converts acceleration into position by double integration, tiny sensor errors and noise accumulate over time, causing the position estimate to gradually diverge from reality. For example, inexpensive smartphone IMUs can accrue errors of several meters after only tens of seconds of walking. Continuously tracking movement in large spaces like warehouses using only IMU is unrealistic, and maintaining centimeter-level accuracy is difficult. Therefore, IMUs alone are limited for completing high-precision positioning and are generally used in combination with other methods (such as UWB or QR codes) for periodic position correction. IMUs excel at auxiliary position estimation and motion detection and are an indispensable bridging sensor in warehouse positioning systems.

SLAM (Simultaneous Localization and Mapping)

SLAM (Simultaneous Localization and Mapping) is a technology that simultaneously performs localization (estimating current position) and mapping (building an environmental map) while sensing the surroundings. Representative implementations include LiDAR-based SLAM and Visual SLAM using camera images. For robots and autonomous vehicles in warehouses, SLAM allows them to operate in environments without pre-existing maps by learning the surrounding topology while determining their position.

The advantage of SLAM is that no prior infrastructure setup is required. Without installing beacons or markers in the environment, a robot can create a map and estimate its position using only its onboard sensors, making it adaptable to warehouses where layouts change frequently. The maps generated by SLAM (point clouds or feature maps) also have added value for spatial management and route planning within the warehouse.

However, SLAM has challenges. First, it imposes a large computational load. LiDAR-SLAM and Visual SLAM require high-speed processors and sufficient memory for real-time processing; dedicated computers or high-performance edge devices may be necessary. SLAM is also highly dependent on the environment: in homogeneous warehouse aisles with similar shelves, the system can lose distinguishing features and face the “loop closure” problem where it cannot determine its location. Moreover, positions obtained by SLAM are relative to the coordinate system the robot created and do not match external references (for example, the building’s global coordinates). Thus, even if SLAM produces high-quality maps, calibration is required to align those maps with real-world drawing coordinates. SLAM alone cannot easily guarantee centimeter-level absolute accuracy, and because its algorithms rely on self-correction (loop closing) to recover from accumulated errors, it may not be suitable for applications that require guaranteed sequential positional accuracy (such as some real-time control tasks).

Therefore, SLAM is useful for autonomous navigation and mapping inside warehouses, but the positioning it provides should be regarded mainly as relatively high-precision. In practical operations, SLAM is commonly combined with external references such as UWB to tie SLAM maps to a reference coordinate system or to reset position at key points to maintain accuracy.

UWB (Ultra Wideband)

UWB (Ultra Wideband) is a radio technology that uses extremely short nanosecond-level pulses to enable high-precision distance measurement. Compared with typical Wi‑Fi or Bluetooth, UWB occupies a very wide frequency band and has high temporal resolution, allowing precise measurement of Time of Flight and achieving distance measurements of less than several tens of centimeters. Leveraging this distance-measurement capability, multiple fixed stations (anchors) are deployed indoors, and by performing two-way or one-way radio communication with tags mounted on moving objects, the tag position is calculated by trilateration in a UWB positioning system.

The strengths of UWB positioning are centimeter-level accuracy and stable positioning performance. Because it calculates distance directly from signal arrival times or time-difference-of-arrival (TDoA), its theoretical accuracy is very high, and practical deployments report positioning errors of around 10–30 cm (3.9–11.8 in). UWB signals also tend to penetrate gaps in walls, floors, human bodies, and machinery more readily due to their frequency characteristics, so positioning can continue in environments with many obstacles such as warehouses. Compared with other radio methods (narrowband Wi‑Fi or Bluetooth), UWB is less susceptible to interference from other devices, making it suitable for complex radio environments in factories and warehouses.

The challenges of UWB are initial deployment cost and operational burden. To achieve high precision, multiple anchors (generally 3–4 or more) must be installed in the area and each anchor’s exact coordinates measured. Covering a large warehouse requires many anchors, which involves installation work, power and communication infrastructure, and periodic calibration and maintenance. UWB tags and anchors are dedicated devices, so if many tags are deployed, equipment costs can add up. However, UWB chip costs have been decreasing and some smartphones now include UWB (e.g., for digital keys and positioning), lowering the adoption barrier. For tracking limited important assets (forklifts or expensive measurement equipment), UWB is a strong option.

BLE beacons (Bluetooth Low Energy)

BLE (Bluetooth Low Energy) is a low-power short-range radio standard; small transmitters that broadcast location information are called BLE beacons. Beacons powered by batteries periodically emit Bluetooth signals that smartphones or dedicated receivers pick up; distance or direction to the beacon is estimated from received signal strength or angle of arrival to determine position. By installing many BLE beacons throughout a warehouse to create a radio field and having workers or carts carry smartphones, approximate current locations can be obtained.

BLE positioning’s advantages are low cost and ease of deployment. Beacon devices are inexpensive, so attaching tens to hundreds of beacons in a warehouse is not a large burden. Battery-powered beacons can operate for months to years, avoiding wiring work. Receivers can be off-the-shelf smartphones or tablets, allowing immediate start without dedicated hardware. Bluetooth’s use of the global 2.4 GHz band also makes it compatible with existing IoT sensors and mobile devices.

However, BLE beacon positioning often yields accuracy on the order of meters. Traditional methods estimate distance from received signal strength (RSSI), but signal strength varies significantly with environment; in metal-shelf-dense warehouses, reflections and blockage cause large RSSI fluctuations. Thus, simple distance calculations do not provide high accuracy, and BLE positioning has typically been used for zone detection at room level or rough area awareness. Recently, efforts to combine BLE with AoA (Angle of Arrival) aim to achieve sub-1 m accuracy, and using dedicated antenna arrays can improve accuracy. Nevertheless, BLE still has larger errors compared to UWB and is not suitable for centimeter-level needs. High-precision BLE setups become more complex and require advanced calibration analyzing phase differences from multiple beacons, raising implementation hurdles.

In summary, BLE beacon positioning can be deployed widely and inexpensively but has limited accuracy. It is useful for understanding general whereabouts of people and goods in a warehouse or for zone management visualization, but not for extremely precise autonomous driving or precision measurement. A common approach is to use BLE where high precision is unnecessary and switch to UWB or other methods when finer positioning is required.

QR code guidance

Using QR codes or AR markers (image markers) to determine position has long been used in warehouse positioning. A typical example is placing unique QR codes on the floor at regular intervals; an AGV uses a downward-facing camera to read the codes while moving and corrects its position. Each QR code is mapped to absolute coordinates within the warehouse coordinate system, and the robot resets its position to the “correct” coordinates whenever it detects a code. This prevents error accumulation from long runs and enables stable autonomous movement.

The benefits of QR code guidance are system simplicity and high accuracy. Printed codes are low-cost and can be placed relatively freely; if a camera can reliably capture them, they can be mounted on walls or ceilings as well. If matching from a read QR code ID to coordinates is accurate, position correction accuracy can be within a few centimeters depending on camera resolution. As passive markers that require no power, QR codes have fewer disturbances compared to radio signals and can serve as highly reliable position references if misreads are prevented.

The disadvantages are installation and maintenance effort. Covering a large warehouse with a sufficient number of markers requires significant labor. Floor-mounted codes can be worn or soiled by foot or vehicle traffic, necessitating periodic inspection and replacement. Between detectable codes (intermediate points), systems must rely on relative position estimation; if code spacing is too wide, drift will occur and accuracy will degrade. Maintaining high accuracy requires densely placing markers or combining with other sensors (IMU or wheel encoders) to bridge gaps.

Overall, QR code guidance provides high absolute accuracy at specific points. It is effective for guaranteeing pinpoint accuracy at “here” locations such as robot stops or shelf positions, but is auxiliary for continuous positioning. Recent applications include using AR markers for automatic shelf recognition and research on using ceiling markers read by drone-mounted cameras for self-localization; it is a simple yet versatile technology.

Other localization technologies

Beyond the above, a variety of approaches exist for warehouse positioning. Wi‑Fi fingerprinting estimates location from patterns of received signal strength and is deployed in commercial facilities and office buildings; however, accuracy tends to remain at the meter level, so its role in warehouses needing high precision is limited. Mapping small variations in geomagnetic fields for localization is another method, but warehouses’ metal structures make magnetic environments unstable and accuracy difficult to guarantee. Acoustic (ultrasonic) positioning is also under research: by measuring flight time of sound using microphones and ultrasonic speakers, centimeter accuracy is easier to achieve than with radio because sound travels more slowly, but capturing sound reliably in noisy factories and warehouses is challenging.

As the above shows, indoor positioning uses various physical phenomena—light, radio waves, magnetism, sound—each with different characteristics, accuracy, and implementation cost. Compatibility with existing infrastructure is also important. For example, if a warehouse already has many Wi‑Fi access points, Wi‑Fi positioning may be reasonable; if introducing a new high-precision system, choosing future-oriented UWB may be preferable. Ultimately, selecting and combining multiple technologies according to environment and needs is the key to achieving high-precision and efficient positioning.

Selection points for high-precision positioning technologies

As described above, indoor positioning has various methods, each with pros and cons. When introducing a high-precision positioning system in your warehouse or project, what viewpoints should you use to choose a technology? Here we explain four main comparison axes to consider: “accuracy and reliability,” “real-time capability,” “cost and implementation hurdles,” and “operability and maintenance.”

Accuracy and reliability

The most important point is the required positioning accuracy level. Generally, the stricter the desired accuracy, the fewer the choices. For centimeter-level accuracy, options are currently limited to technologies such as UWB, high-performance SLAM, and special visible-light markers. For example, if a forklift’s automated insertion into shelves must be within ±5 cm (±2.0 in), a high-precision RTLS like UWB or LiDAR-based precision positioning is indispensable. Conversely, if around ±0.5 m (±1.6 ft) is acceptable operationally, lower-cost technologies like BLE beacons or Wi‑Fi positioning may suffice.

Alongside accuracy, consider reliability. Even if positioning accuracy is temporarily high, it is unsuitable for real operations if it quickly degrades or drops out with environmental changes. Evaluate whether the system can stably maintain the target accuracy and how robust it is to disturbances. For instance, UWB tends to deliver consistent high accuracy in factory settings, while BLE RSSI positioning fluctuates with changes in human or object placement. IMU and SLAM accumulate errors over time, so without periodic corrections or closed-loop recalibration, their reliability declines. Assess whether the system can maintain required accuracy over time and whether it includes self-monitoring to guarantee performance.

Real-time capability

The required real-time property varies with how position information is used. For real-time control of autonomous mobile robots or collision-avoidance systems, high update frequency (e.g., 10 Hz or more) and low latency to obtain the latest position are desirable. IMUs provide information at sub-millisecond intervals and contribute to smooth motion, and UWB systems typically compute positions in tens of ms to around 100 ms, making them suitable for tracking dynamic targets. BLE and Wi‑Fi RSSI-based positioning can have longer scan intervals and filtering delays of several seconds, making them unsuitable for immediate control of moving objects.

On the other hand, tasks like equipment inspection or as-built measurement—where a person walks and measures points—do not require very high update rates; delayed results are acceptable if accuracy is guaranteed. In such cases, operations that prioritize precision over real-time performance, like mapping with SLAM followed by offline processing, are acceptable. When selecting a system, clarify whether your use case needs immediate position data or whether high-precision data obtained later is sufficient. High-precision indoor positioning often involves large data processing, so whether you need on-the-spot results or can rely on offline analysis will change the suitable approach.

Cost and implementation hurdles

As a practical constraint, budget and implementation hurdles must be considered. No matter how capable a system is, it must fit the budget, and systems requiring large-scale installation or specialized skills may be difficult to handle on-site. Evaluate hardware and software initial costs as well as installation and setup labor and required personnel skills.

For example, installing UWB or BLE anchors/beacons may involve fixing dozens of devices to ceilings or walls. Consider not only the device cost but also high-place work, power provisioning, and relocation costs when layouts change. QR codes can be started on a low budget but affixing them manually can be labor-intensive, and careful design and management of codes are needed. High-performance LiDAR and GPU-equipped PCs for SLAM are expensive, and robots equipped with such hardware are not cheap. Conversely, smartphone-based solutions (e.g., LiDAR-equipped phones plus apps for surveying) can be deployed relatively cheaply by repurposing devices.

Also consider skills and learning costs. Is the system easy for existing workers to operate, or does it require special training? Systems with many configuration options may be hard to use in sites without IT experts. Smartphone apps and intuitive AR interfaces are easier for on-site workers to accept and help with adoption.

Operability and maintenance

Finally, ease of long-term operation and maintenance is important. If many battery-powered devices are involved, periodic battery replacement and inspection are required; sensors installed in the environment affect accuracy if they fail or move. For example, tracking and replacing batteries for hundreds of BLE beacons can be a significant task; UWB anchors benefit from periodic re-measurement and calibration of anchor-to-anchor distances. If warehouse layouts change, marker or anchor configurations must be changed and reconfigured.

On the software side, map data and coordinate system updates, firmware updates, and tuning positioning algorithms are ongoing tasks. Complex systems generally require more maintenance and may need a dedicated manager or outsourced maintenance service.

To reduce maintenance burden, choose simple, automated mechanisms where possible. Systems with few installed devices and no battery changes are ideal; if many devices are unavoidable, use management software to monitor battery levels and anomalies centrally. Cloud-connected systems that enable remote software updates and data management are also convenient.

In summary, do not underestimate the resources needed to keep a positioning system running. Consider not only initial cost but also annual maintenance labor and additional expenses, and choose technologies that match the site’s operational capabilities for long-term success.

Simple surveying solution for indoor/outdoor with LRTK

Finally, as a modern solution that achieves centimeter-level positioning both outdoors and indoors using the technologies discussed above, we introduce simple surveying with LRTK. LRTK is a positioning solution that integrates an ultra-compact RTK‑GNSS receiver that can be used with smart devices, a dedicated app, and cloud services. It reconfigures RTK surveying—which traditionally required expensive equipment and specialized knowledge—into a form that anyone can use easily, enabling one-person centimeter-level positioning from vast outdoor sites to indoor warehouses.

Use of GNSS correction positioning (RTK)

LRTK’s main feature is its full use of GNSS correction positioning, RTK (Real-Time Kinematic). A small antenna receiver attached to a smartphone or tablet receives signals from multiple satellites, and by applying reference station data or cloud correction information received via the Internet in real time, it reduces standalone positioning errors of several meters to a few centimeters. This enables outdoor position measurements that meet construction site requirements (approximately 2–3 cm (0.8–1.2 in)). Unlike conventional surveying instruments, you do not need to secure line-of-sight or set up repeatedly; simply carrying the antenna and walking to a point allows coordinate acquisition, dramatically improving efficiency for surveying and layout tasks.

RTK is fundamentally powerful in open outdoor conditions, but LRTK is designed to work with nationwide satellite reference station networks and augmentation systems (such as Japan’s QZSS sub-meter/centimeter augmentation services), making high-precision GNSS positioning stable even in mountainous or urban areas. You can also set up your own mobile base station for local RTK if needed, offering flexibility to operate without dependence on communications.

Combination with indoor localization

What makes LRTK unique is its support for positioning even in GNSS-denied indoor environments. Specifically, by using high-precision reference coordinates obtained outdoors as a starting point and combining them with smartphone internal IMU and camera/LiDAR–based localization technologies, you can continue tracking position inside buildings. For example, before entering a warehouse, obtain a reference point coordinate near the entrance using LRTK; once inside, use the phone’s AR features and IMU data to track relative movement. QR code markers or known points encountered along the way can be referenced for corrections as needed to maintain positioning indoors. In this way, LRTK seamlessly connects absolute accuracy from GNSS with relative position estimation via sensor fusion, enabling centimeter-level positioning even in environments where satellites do not reach.

This approach allows surveying work that crosses outdoor and indoor areas to produce results in a consistent coordinate system. For example, positions of indoor columns or equipment can be measured and mapped directly relative to an external origin or reference axis obtained outdoors, eliminating the need to survey indoors and outdoors separately and then merge data. LRTK enables a single workflow where previously two separate systems had to be used and integrated later.

Field application via smartphone integration

LRTK is designed to integrate with smartphones and tablets, offering excellent usability on site. In the dedicated app’s intuitive interface, your current position is displayed on a map, and direction and distance to target points are shown in real time using AR. This allows even non-experts to accurately stake out points or place equipment at specified coordinates. Measured data are automatically saved and shared to the cloud, removing the need for manual note-taking and allowing immediate drawing creation or information sharing in the office. Using familiar smartphones lets untrained workers operate the positioning system almost like a game, smoothing adoption.

Another smartphone advantage is multipurpose use of the device. One smartphone plus an LRTK receiver can handle surveying, photo capture, cloud communication, drawing viewing, and other field DX tasks. What used to require separate surveying instruments, cameras, drawing files, and communication devices can be consolidated into a single smartphone with LRTK, improving efficiency.

Ease of implementation and effectiveness

LRTK’s low implementation hurdles despite its high precision are also attractive. Simply attaching the receiver to a mobile device requires no special installed equipment or construction, and initial setup can be completed in minutes by following guided steps, allowing immediate use on site. The hardware is compact and battery-powered for easy portability across multiple sites.

On price, LRTK is generally more accessible than traditional total stations or 3D scanners (specific prices are omitted here). Small contractors and facilities management teams can often afford it and can start with a single unit for pilot deployment before broader rollout. Because cloud services are included, buyers receive continuous software updates and support, minimizing maintenance burden.

Overall, LRTK’s simple surveying is a groundbreaking solution that lets anyone achieve centimeter-level positioning even where GNSS does not reach. It reduces labor in surveying tasks at understaffed sites and promotes digital transformation of construction management through digitized positioning data. If you need fast and accurate position information indoors or outdoors, consider adopting LRTK.

Conclusion

To achieve centimeter-level positioning in warehouses without relying on GNSS, it is essential to combine multiple technologies appropriately. Understand the characteristics of IMU, SLAM, UWB, BLE, and QR code methods, and design a system that matches your requirements for accuracy, real-time capability, cost, and operational constraints. Fortunately, sensor performance has improved and costs have declined in recent years, making high-precision indoor positioning increasingly feasible. Conduct the comparative review described in this article and adopt the optimal technical solution to dramatically improve productivity and safety in warehouse management and construction sites. In particular, new positioning tools like LRTK that work even where GNSS does not reach have the potential to overturn conventional wisdom. Actively utilize the latest technologies to solve positioning challenges and drive on-site digital transformation.

FAQ

Q: Why can’t GPS (GNSS) be used inside warehouses? A: Satellite signals are blocked by building roofs and walls and therefore hardly reach indoors. Even if faint signals are received, radio waves reflect off floors and walls inside warehouses, causing large errors in position calculation. As a result, outdoor-level positioning cannot be achieved. Therefore, inside warehouses you must measure position using alternatives to GPS, such as UWB, beacons, or cameras.

Q: Can centimeter-level accuracy really be achieved indoors? A: It is achievable. For example, UWB positioning systems can stably provide accuracy around 10 cm (3.9 in), and marker-based methods like QR codes can correct position within a few centimeters when read. Advanced sensor fusion combining IMU and vision (camera/LiDAR) can maintain relatively high precision even without GNSS. Solutions like LRTK, which extend RTK-level accuracy from outdoors to indoors, have also emerged; with appropriate methods, centimeter-level positioning indoors is feasible.

Q: How should I choose between UWB and BLE beacons? A: Use UWB where high precision is the top priority—for example, for position control of forklifts, robots, or safety management, where the extra cost is justified. If you need to track many targets or prioritize ease and low cost over precision, BLE beacons are an attractive option. BLE is often sufficient for general whereabouts of people and cargo. Note that BLE with AoA can improve accuracy to just under 1 m (less than 3.3 ft), so balance your budget against accuracy requirements.

Q: What is needed to introduce LRTK? A: The basic set includes an LRTK receiver, a compatible smartphone or tablet, and an Internet connection. Attach the receiver to the device and launch the dedicated app; after a few minutes of initial setup, you can start positioning. Correction data are automatically obtained via the cloud, so you do not need to operate your own base station (though you can set up a local reference station if needed). No special surveying instruments or complex configuration are required—common IT devices and network access are enough.

Q: Can someone without surveying experience use LRTK? A: Yes. LRTK is designed to be usable by non-experts, with intuitive app operation and AR guidance. In practice, workers without conventional staking experience have successfully used LRTK for tasks like stake placement. Although brief instruction is helpful, it is much easier to learn than traditional surveying instruments, and support systems are in place for assistance. The fact that anyone can use it is a major LRTK advantage.