How far can cm-level accuracy in indoor positioning be achieved? Mechanisms of technologies by accuracy and use cases

Indoor positioning "accuracy" ranges widely depending on the technology used, from tens of meters to a few millimeters. To select a technology that matches the requirements on site, it is important to accurately understand what level of accuracy each technology can provide. This article explains the mechanisms of representative technologies for each accuracy band and concretely describes the situations where each excels.

Table of Contents

• Current status and challenges of indoor positioning accuracy

• Mechanisms and applications of zone-level accuracy technologies

• Mechanisms and applications of sub-meter accuracy technologies

• Mechanisms and applications of decimeter accuracy technologies

• Mechanisms and applications of cm level accuracy (half-inch accuracy) technologies

• Typical sites requiring cm level accuracy (half-inch accuracy) indoors

• Environmental factors affecting cm level accuracy (half-inch accuracy)

• Continuity with the outdoors: combination with RTK positioning via LRTK

• Indoor positioning accuracy evaluation metrics

• Relationship between positioning accuracy and safety

• Integration with digital twins

• Summary

Current Status and Challenges of Indoor Positioning Accuracy

Outdoors, GNSS can provide positioning with accuracies ranging from several meters (several ft) to several tens of centimeters (several in), but indoors GNSS signals are blocked by buildings and cannot achieve sufficient accuracy. Therefore, indoor-specific positioning technologies are required.

The accuracy of indoor positioning technologies can be broadly divided into four bands.

Zone-level accuracy (a few meters to a few tens of meters) is used for applications that need to determine which area someone is in. Proximity detection using Bluetooth Low Energy (BLE) beacons or conventional Wi-Fi falls into this accuracy range.

Sub-meter accuracy of approximately 0.5–2 m (1.6–6.6 ft) is used for finer position tracking. Wi-Fi RTT and some BLE systems aim for this level of accuracy.

Decimeter-level accuracy, approximately 10–50 cm (3.9–19.7 in), can be used for precise facility management and analysis of people's movement patterns. Entry-level UWB and high-precision Wi-Fi cover this range.

cm level accuracy (half-inch accuracy), approximately 1-10 cm (0.4-3.9 in), is an indispensable domain for high-precision work in construction, manufacturing, surveying, and similar fields. It covers high-precision UWB design, LiDAR＋SLAM, optical marker positioning, and high-precision GNSS (RTK).

For each accuracy band, we'll sequentially examine which technologies are used and what business challenges they address.

How Zone-Accuracy Technology Works and Its Applications

Zone positioning using BLE beacons is basically proximity detection: when a smartphone detects a beacon's radio signal, it judges that "it is near that beacon." It is also possible to estimate distance using signal strength (RSSI), but signal strength is strongly affected by obstacles and reflections, making it difficult to ensure stable accuracy.

The main use cases are wayfinding/navigation in shopping malls and exhibition venues, access control for entry and exit, and patient and staff location management within hospitals (zone-level). Its greatest advantages are low cost and the convenience of operating on existing smartphones.

Positioning using typical Wi‑Fi signal strength (RSSI) similarly has an accuracy of several meters to a dozen or so meters (several ft to around a dozen ft), and is suitable for zone management. Using a fingerprinting method (pre-learning the signal strength pattern at each location) improves accuracy compared with simple RSSI-based positioning.

These zone-level accuracy techniques, though low in accuracy, have very low deployment costs and can be implemented using only existing infrastructure and smartphones, making them effective as an entry point for sites adopting indoor positioning for the first time.

Mechanisms and Applications of Sub-Meter Accuracy Technologies

Wi-Fi RTT (IEEE 802.11mc) transmits and receives radio waves between a smartphone and Wi-Fi access points, measures the round-trip time (RTT: Round-Trip Time) of flight, and calculates distance. Because it uses time of flight rather than radio signal strength, it can provide more stable distance estimates than RSSI. Position is calculated by trilateration from distances to multiple access points.

Accuracy is approximately 1–2 m (3.3–6.6 ft), and it can be used to identify locations at the corridor or room level. Being able to utilize existing Wi‑Fi infrastructure is advantageous from a cost perspective.

Use cases include medical equipment management within hospitals (tracking at the room level), people-flow analysis within large facilities, and tracking of transport vehicles such as forklifts (zone management). For workflows where accuracy within a few meters (a few ft) is sufficient, Wi-Fi RTT is the most cost-effective option.

Angle-of-arrival measurement using ultrasound (BLE Angle of Arrival: AoA) is also attracting attention as a technology capable of achieving sub-meter to decimeter accuracy. By calculating the arrival direction of BLE radio waves, positioning that does not depend on signal strength becomes possible. Installation of corresponding anchors is required, but because it is based on BLE, a low-cost technology, it has the advantage of keeping deployment costs low.

How Decimeter-Precision Technology Works and Its Applications

Entry-level UWB systems can achieve an accuracy of around 15–50 cm (5.9–19.7 in). The basic principle of UWB is time-of-flight-difference measurement using very short pulses with an extremely wide bandwidth (several GHz). Using a wide bandwidth makes it easier to identify multipath (multiple reflections of radio waves), enabling robust ranging using time-difference-of-arrival (TDoA) or two-way time-of-flight (TWR).

UWB systems with decimeter-level accuracy (3.9 in) arise when the number of anchors is small or when the geometric configuration is unbalanced. With appropriate design, an upgrade to cm level accuracy (half-inch accuracy) is also possible.

Use cases include pallet tracking in warehouses, managing worker movement at construction sites, and tracking the locations of medical equipment in healthcare facilities. Even with decimeter-level accuracy (10 cm / 3.9 in), it allows far more granular management than zone-level accuracy and can sufficiently cover many industrial applications.

Mechanism and Applications of cm level accuracy (half-inch accuracy) Technology

Mechanism of UWB-based positioning with cm level accuracy (half-inch accuracy)

To reliably achieve cm level accuracy (half-inch accuracy) (5-15 cm (2.0-5.9 in)) with UWB, several design considerations are required. It is important to evenly deploy a sufficient number of anchors (typically four or more), to perform precise time synchronization among anchors at the nanosecond level, and to adopt robust processing algorithms for multipath environments.

In positioning calculations, methods such as least squares and Kalman filters are used to integrate distance data obtained from multiple anchors. By aggregating information from multiple anchors, if some anchors suffer radio interference others can compensate, maintaining stable positioning.

With a strong track record in industrial settings, it is ideal for real-time tracking of people, objects, and robots. It is especially widely used for component position management on production lines, autonomous navigation of AGVs in warehouses, and tracking of materials and equipment on construction sites.

Mechanism of positioning with cm level accuracy (half-inch accuracy) using LiDAR＋SLAM

LiDAR emits infrared and visible-light laser pulses and measures the time until the reflected waves return (ToF: Time of Flight) to generate a three-dimensional point cloud of the surroundings. By matching this point cloud with a known three-dimensional map, self-localization with cm level accuracy (half-inch accuracy) is possible.

In SLAM algorithms, feature points are extracted from the acquired point cloud, and the position is continuously updated by tracking changes from the previous frame. When combined with an IMU, robust positioning against sudden movements is achieved.

The resolution, scan rate, and measurement range of the LiDAR used affect positioning accuracy. More precise LiDARs are more expensive, but they also improve accuracy and reliability accordingly. They are commonly mounted on AGVs and inspection robots and are used for autonomous material handling in factories, warehouse picking robots, and facility‑inspection drones.

Mechanism of positioning with cm level accuracy (half-inch accuracy) using optical markers and cameras

By affixing AR markers with known dimensions and shapes (e.g., ArUco markers) to the installation site and detecting them with a camera, you can compute the camera's position and orientation with accuracy below 1 cm (0.4 in). Camera calibration (precise measurement of intrinsic parameters) determines the accuracy.

It has low computational cost and can run on embedded systems. It requires that markers remain within the field of view, and coping with changes in illumination is a challenge. It is used for positioning inspection on manufacturing lines, for setting reference points in AR-assisted work, and as auxiliary references for 3D measurement.

High-Precision Positioning in Combination with RTK-GNSS

GNSS is generally not available indoors, but in areas near building openings (windows, doors, skylights) and semi-outdoor spaces (such as covered loading bays at factories), augmented high-precision GNSS (RTK) can sometimes be used. RTK is a technology that uses correction data from a reference station to achieve positioning with cm-level accuracy (half-inch accuracy); its use is primarily outdoors, but it is effective in special indoor environments where GNSS signals can be obtained.

Representative indoor sites requiring cm level accuracy (half-inch accuracy)

On indoor construction sites, confirming the positions of walls, floors, and ceilings with cm level accuracy (half-inch accuracy) is essential for quality control. By checking discrepancies against design data in real time, construction errors can be detected and corrected immediately. Integrating design BIM data with positioning data enables simultaneous management of construction progress and quality.

In factory precision parts manufacturing, cm level accuracy (half-inch accuracy) is required for the positioning of machining equipment and for managing parts transport routes. Because positional deviation directly leads to product defects, positioning accuracy greatly affects quality and yield. In real-time quality control on production lines, the accuracy of positional information is a source of competitive advantage.

In equipment inspections at large facilities, recording which device was inspected and at what location with cm level accuracy (half-inch accuracy) prevents missed inspections and recording errors. Accurate location records are especially important for equipment subject to statutory inspections. Integration with GIS (Geographic Information System) enables centralized management of equipment location information and inspection history.

In automated material handling at warehouses and distribution centers, AGVs stop in front of shelves with cm level accuracy (half-inch accuracy), ensuring the picking accuracy of robotic arms. Because shelves, robots, and people coexist in the same environment and must operate safely without collisions, positional accuracy also affects safety. As logistics automation advances, cm level positioning (half-inch accuracy) has become the foundation of automation system reliability.

Environmental factors that affect cm level accuracy (half-inch accuracy)

Even when using the same technology, the actual accuracy can vary greatly depending on the environment.

When the density of metallic structures is high, radio wave reflections and scattering increase, making non-line-of-sight (NLoS) positioning more likely. This effect is particularly pronounced with UWB, so algorithmic NLoS detection and correction are important. Considering the positioning system from the building design stage can minimize radio wave shielding that is difficult to mitigate later.

Changes in temperature and humidity affect radio wave propagation characteristics. In environments with large temperature variations (such as refrigerated warehouses), periodic updates of correction parameters are necessary. In acoustic (ultrasonic) positioning, temperature changes affect the speed of sound, so temperature correction is key to maintaining accuracy.

In environments with high electromagnetic noise (near large motors or high-frequency power supplies), sensors can experience decreased sensitivity and positioning errors caused by noise. Shielding and careful placement are necessary. Noise mitigation is also important for the power wiring of positioning equipment.

Changes to a building's layout also affect accuracy. When the placement of shelves or machinery changes, radio propagation patterns change, so recalibration and map updates become necessary. Establishing a process to share layout-change plans with the personnel responsible for the positioning system during the planning stage is important for maintaining long-term accuracy.

Continuity with the Outdoors: Combining RTK Positioning via LRTK

In sites that require cm level accuracy (half-inch accuracy) positioning indoors, continuous high-precision coordinate management with the outdoors is also a challenge.

LRTK is a device that, when attached to an iPhone, can obtain absolute coordinates outdoors with cm level accuracy (half-inch accuracy) using RTK-GNSS. By bringing high-precision reference coordinates established outdoors and linking them to indoor work location information, integrated high-precision coordinate management across indoor and outdoor environments becomes possible.

It is especially important on construction sites to carry the coordinate system established by external surveying directly into indoor work. By using LRTK to obtain outdoor reference points easily with cm level accuracy (half-inch accuracy) and applying those coordinates to indoor construction management, you can maintain a high level of consistency with the design data.

LRTK can be operated without surveying expertise because its simple mounting style allows field personnel to start using it immediately. In integration with indoor positioning systems, it also plays an important role as a source of reference points. The acquired positioning data can be centrally managed in the cloud, enabling real-time data sharing between the field and the office.

Indoor Positioning Accuracy Evaluation Metrics

When evaluating the accuracy of indoor positioning systems, it is important to combine multiple metrics. Even systems that claim "cm level accuracy (half-inch accuracy)" based on a single metric can perform very differently in real-world environments. When selecting a system, it is recommended to verify the actual accuracy in your company's field environment through a PoC (proof of concept). It is important to ask vendors to specify measurement conditions and to present accuracy data for static, dynamic, and full-area conditions. Even if a catalog states "average error 10 cm (3.9 in)", actual field performance can differ greatly depending on measurement conditions (no obstacles; a small number of anchors; measurements taken in a stationary state). By checking measurement conditions in detail and requesting data under conditions close to your field environment, you can prevent post-deployment disappointment. In an ideal PoC, it is recommended to perform continuous measurements for at least one week and to collect accuracy data under various conditions such as day and night, during operation, and when not operating.

Mean Error is the average of the differences between the measured position and the true position. It is the most commonly used accuracy metric and is expressed like "mean ± 3 cm (±1.2 in)". However, because the mean alone can conceal the maximum error, it should be evaluated in combination with other metrics.

RMS error (Root Mean Square Error) is the square root of the mean of the squared errors, and is a metric that also takes into account the variability of the errors. Because it is sensitive to outliers (abnormally large errors), it is used to evaluate stability.

Percentile error (e.g., the 95th percentile error) indicates a guaranteed error bound such that 95% of the positioning data fall within this error. It is useful for evaluating a system's robustness and is an important criterion for selecting a highly reliable positioning system.

Availability (Availability) is the proportion of time the system is outputting positioning data correctly. An availability of 99% or higher is generally required for industrial applications. Redundant design that takes into account anchor failures, power outages, and communication faults is important for ensuring availability.

Relationship between positioning accuracy and safety

Positioning with cm level accuracy (half-inch accuracy) contributes greatly not only to operational efficiency but also to safety. In particular, in factories and warehouses where people coexist with robots and AGVs, accurate location information forms the foundation of safe operation.

When an AGV can determine its position with cm level accuracy (half-inch accuracy), the safety function that detects the presence of people on its travel path in real time and automatically stops just before a collision operates precisely. If positioning accuracy is low, a larger safety stopping margin is required, sacrificing the AGV’s operating speed and efficiency.

Also, in environments where workers' positions are managed with cm level accuracy (half-inch accuracy), entry into hazardous areas can be detected immediately and warnings issued. High-precision positioning is used as a mechanism to prevent careless approaches to work at heights and areas with a risk of falls.

In worker location management at construction sites, you can know in real time which worker is at which location, even on large sites spanning multiple floors and work zones. The ability to quickly confirm personnel and guide evacuation during emergencies (earthquakes, fires, etc.) is also an important safety value of high-precision positioning.

Integration with Digital Twins

Indoor positioning with cm level accuracy (half-inch accuracy) dramatically enhances the accuracy and usefulness of digital twins (virtual models that faithfully reproduce physical spaces digitally) in manufacturing, construction, and infrastructure management.

When accurate location data flows into the digital twin in real time, equipment operating status, inventory distribution, and workers' movement paths are visualized on the virtual model. This enables instantaneous detection of discrepancies between the real site and the virtual model, allowing early identification of deviations from the plan.

At construction sites, by checking the differences between construction data and design BIM data in real time, early detection of quality issues and reductions in rework costs are realized. In manufacturing environments, reflecting actual part positions and worker positions in the production line's digital twin in real time enhances anomaly detection and process improvement.

In such digital twin applications, acquiring high-precision absolute coordinates outdoors using LRTK functions as an important input for precisely aligning the digital twin’s coordinate system with that of the real world.

Summary

Indoor positioning with cm level accuracy (half-inch accuracy) can be achieved using multiple technologies such as UWB, LiDAR + SLAM, and optical markers. Each of these technologies has environments where it excels and environments where it does not, so choosing according to on-site requirements is important.

Selecting technology by comprehensively evaluating not only accuracy but also cost, ease of deployment, and maintainability is the quickest way to achieve stable cm level accuracy (half-inch accuracy) on-site. Also, for continuous high-precision coordinate management across indoor and outdoor environments, iPhone-mounted GNSS high-precision positioning devices such as LRTK are effective, and combining them with indoor positioning systems allows efficient digitization of the entire site.

Improving positioning accuracy is an investment that directly leads to increased work efficiency, improved quality, and enhanced safety. First, clarify the site's accuracy requirements and ensure cm level accuracy (half-inch accuracy) through appropriate technology and implementation design.