AR Guidance at Railway Equipment Construction Sites: Streamlining Positioning with High-Precision Positioning Technology

At construction sites where railway equipment such as signals, catenary poles, and cables are installed, accurate positioning work is essential. However, this positioning work currently faces various challenges including labor shortages, constraints from night work, and dependence on skilled technicians. This article focuses on construction support using high-precision GNSS (RTK) and AR (augmented reality) as a solution to these issues. Utilizing the latest technology that combines smartphones and compact GNSS units, we explain initiatives to improve the efficiency and accuracy of on-site positioning with concrete examples and case studies.

Challenges in Positioning Work at Railway Equipment Construction Sites

On railway equipment construction sites, high accuracy is required for the marking and positioning work that determines where signals and poles are installed. Even an error of a few centimeters (a few in) can affect train safety and equipment functionality, so careful surveying and marking are indispensable. However, traditional surveying methods have several problems.

• Human resource shortages and dependence on skilled personnel: Traditional marking is commonly done by two-person teams using total stations or tape measures, and scenes often rely on experienced technicians. However, the railway industry is also facing serious labor shortages due to an aging workforce and a lack of younger staff, making it difficult to secure the necessary personnel on site. One track maintenance company reported a 23% reduction in workers over 10 years, and cases where younger staff do not remain have been reported. At the same time, there is an imbalance in age composition, with one-quarter of site-supervising employees being 55 years or older, forcing inexperienced younger staff to lead sites. A system that depends on the skills of experienced personnel raises concerns about skill transfer and quality assurance amid labor shortages.

• Night work and time constraints: Railway equipment work is often carried out at night when train operations are suspended, and work must be completed in the limited hours between the last and first train. For example, one site progressed on a tight schedule of “21:00 material delivery, 22:00 work start, surveying and position confirmation at 03:30 the next morning, and full restoration by 05:30.” Performing accurate positioning within such a short time requires significant preparation and planning, and working in the dark reduces visibility and makes even marking difficult. Under nighttime lighting, it is hard to accurately confirm fine chalk lines or stakes, increasing the risk of mistakes and rework.

• Burden of surveying work and human error: Traditional manual surveying required time-consuming tasks such as re-setting the total station and measuring distances with tape measures. When there are many measurement points across a large site, equipment must be moved and reinstalled repeatedly, increasing the physical burden on workers. Manual positioning also inevitably involves human error. If a misreading or miswriting causes marking to shift by several centimeters from the reference, it can lead to positional deviation of the entire structure and later rework. For example, if a height reference chalk line fades and remeasurement is required, or if a drawing note for 5 m 03 cm is mistakenly written as 5 m 08 cm (5 m 03 cm (16.4 ft 1.2 in) vs. 5 m 08 cm (16.4 ft 3.1 in)), corrective work may be necessary after completion. Such construction errors and rework are difficult to recover from within the limited night work hours and may impact timetables.

As described above, positioning work for railway equipment faces a triple burden of labor shortages, time constraints, and precision requirements. Methods that rely on the intuition and experience of skilled workers have limits, and new approaches that balance efficiency and quality assurance are required.

Mechanism of Construction Support Using High-Precision GNSS (RTK) and AR

A key solution attracting attention to address these challenges is construction support technology that combines high-precision GNSS positioning (RTK) and AR (augmented reality). RTK, or Real Time Kinematic, is a technique that uses two GNSS receivers—a base station and a rover—to correct satellite positioning errors in real time and increase positioning accuracy to the centimeter level (half-inch-level). While normal GPS positioning has errors of several meters, RTK can reduce errors by orders of magnitude, realizing total-station-level accuracy over wide areas.

When centimeter-level positioning by RTK becomes possible, positioning work changes dramatically. Without the numerous stakes and guide strings previously required, one can identify a position simply by walking to the prescribed coordinates while carrying a GNSS receiver. For example, by displaying the design coordinates on a tablet or smartphone screen and finding the point where the discrepancy with the current position is zero, a worker can drive a stake on the spot and complete surveying alone. Because positioning is possible as long as satellite signals are received, even in environments that were traditionally difficult to measure—such as mountainous areas or urban districts—efficiency improves. Also, GNSS functions normally in dark conditions, so it is effective for surveying work after sunset. In one station improvement project, an RTK base station was installed on the station building roof, enabling the positioning of platform extension areas and new signal installations within the short period after the last train. Thanks to RTK, accuracy could be maintained while improving work efficiency within the limited time window.

Combined with RTK’s accurate position information, AR technology is also utilized. AR (Augmented Reality) overlays design models or target positions onto the camera view of a smartphone or tablet. In an AR system that uses high-precision coordinates, design points and models can be visualized in the real world without misalignment. Standalone AR had concerns about some positioning error, but fusing RTK precision eliminates discrepancies between digital information and the actual site, making AR displays more reliable.

Concretely, position data and models for equipment installation based on design drawings are prepared in advance. On site, a small RTK-GNSS receiver (antenna) is attached to a smartphone and a dedicated app is launched. When the smartphone acquires high-precision real-time coordinates, the app provides navigation to the target point and displays positions via AR. The screen shows arrows or guides indicating the direction to the target coordinate, and workers simply follow them to be guided to the correct location. By pointing the camera, markers or models indicating “install here” will appear over the live view, allowing intuitive identification of installation points. For example, for a signal pole foundation, an AR marker appears on the ground and driving a stake at that position completes the marking. For cable routing, virtual lines are displayed along the ground to indicate the excavation path. The fusion of GNSS-based precise positioning and AR visual guidance allows anyone to mark the prescribed location without relying on intuition or experience.

Examples of On-Site Guidance Using Smartphone + Compact GNSS Units (Signal Poles, Power Poles, Cable Routes)

Here are concrete examples of how AR on-site guidance using a smartphone and a compact RTK-GNSS unit operates at railway equipment construction sites.

• New installation of signal poles: Signal poles require precise placement for visibility from trains and clearance from other equipment. Traditionally, positions were determined by measuring offsets from the track centerline, but AR guidance can display markers at the design location on the spot. For example, pointing a smartphone at a designated spot by the track will project the foundation outline or center point of the signal pole onto the ground. Workers simply mark according to that AR display to complete design-compliant marking. Because markers appear on the screen, even in poor night visibility, measurement accuracy in the dark is improved. Virtual signal models can be displayed in AR to check the relationship with surrounding equipment and structure limits, allowing verification of interference or sightline obstructions. In one station improvement project, RTK was used to quickly mark signal installations and platform extensions during the short period after the last train, meeting strict accuracy requirements. Combining AR guidance with RTK surveying results allows all site workers to share the information and achieve more reliable construction.

• Installation of catenary poles and utility poles: Catenary poles and poles for substation equipment also need to be installed with accurate position and verticality. AR guidance can display the foundation center on the ground and also use height information for the pole. For example, if the smartphone shows the design height of the catenary pole, the built pole’s height can be verified against the design via AR (smartphone AR functions can measure height to some extent or compare with the design model). One workflow is to display the foundation center in AR for position setting, then after excavation and foundation installation, perform a point-cloud scan of the pole top to measure height. Because catenary poles are installed at height, conventional surveying checks were cumbersome, but simply pointing a smartphone can instantly confirm whether the position is correct, streamlining height checks for high work. RTK-capable compact GNSS receivers are lightweight and portable, allowing positioning even when workers are at height without interfering with operations, so they are used to verify pole positions on narrow platforms.

• Laying cable routes: For buried signal cables, communication lines, or power cables, it is important to mark the route and control depth for excavation. AR on-site guidance can display the designed route directly on the ground, allowing workers to mark as if drawing a line with a virtual spray. Workers follow the guide line on the smartphone screen and mark with chalk or spray to accurately trace complex curves and specified offsets. Curved routes that used to rely on craftsmen’s intuition can be reproduced accurately even by less experienced staff with AR guidelines. After burying cables, a smartphone’s built-in LiDAR can be used to point-cloud-scan the installation and record the positions and depths. For example, scanning the trench or conduit before backfilling and uploading to the cloud automatically generates a 3D model, enabling “x-ray” visualization of buried conduits by simply pointing a smartphone in the future. In practice, the LRTK system simplifies recordkeeping and sharing of buried pipe work with this function. In future maintenance, having past installation locations available in AR helps productivity and prevents accidents caused by mistaken excavation.

As these examples show, smartphone + RTK-GNSS + AR on-site guidance is effective across many railway equipment construction scenarios. From above-ground signals and power to underground cables, precise positioning can be performed intuitively in all situations, leading to labor savings and improved safety.

On-Site Effects of AR Guidance (Positioning Accuracy, Work Time Reduction, Elimination of Reliance on Individuals)

What effects does introducing AR guidance technology bring to railway equipment construction sites? We summarize the benefits from three major perspectives.

• Improvement in positioning accuracy: The combination of RTK-GNSS centimeter-level positioning (half-inch-level) and AR display enables positioning according to design drawings. The few-centimeter errors that were unavoidable with tape measures and visual marking can be reduced to nearly zero by following AR markers. Digital presentation of elements that are hard for humans to grasp—such as height and angle—makes it easy to meet conditions like slopes and levels. AR guidance directly contributes to reducing human error. Workers no longer need to mentally interpret drawings and measure; everyone follows the same AR instructions, preventing misunderstandings and communication mistakes. This can prevent construction errors such as “the marking for a structure shifted by several centimeters” or “incorrect reference height causing poor pipe slope.” AR markers positioned with RTK accuracy provide reproducibility that meets the rigorous standards of the railway sector, achieving centimeter-level (half-inch-level) accuracy comparable to stationary surveying instruments.

• Reduction of work time: AR guidance dramatically improves on-site efficiency. It eliminates frequent equipment re-setup and repeated re-measurements, enabling fast positioning over wide areas. For instance, dozens of stake markings that used to take half a day can be confirmed and marked sequentially by one person with AR guidance, significantly reducing time. In surveying operations where GNSS was introduced, operators reported being able to show measurement points one after another without holding up heavy equipment operators, making machine operating time more effective. AR guidance is especially valuable for short night-time construction windows peculiar to railway work: rapid and accurate positioning within the brief period after the last train allows sufficient time for full restoration before the first train. Labor-saving from reducing tasks previously shared among multiple people also cuts waiting and setup losses for each worker, compressing total work time. The extra time gained can be allocated to safety checks and quality assurance, improving overall construction stability.

• Elimination of reliance on individuals and promotion of skill transfer: AR guidance is easy to use even without special skills. By following intuitive screen displays, site staff can perform positioning without being veteran surveyors. In one site, advanced AR and point-cloud functions were reportedly used by workers without prior training. This demonstrates that measurement and marking work that depended on years of experience can be standardized across the organization. With AR guidance, inexperienced staff can contribute as immediate assets to high-precision positioning. If complex marking that only veterans could perform becomes possible for anyone, a site with labor shortages can have “everyone act as a surveyor.” This also facilitates smoother skill transfer to younger workers: the AR system embeds know-how, allowing users to learn the latest surveying methods through use, raising the organization’s technical baseline. A system that does not rely on specific veterans reduces the risk of skill loss due to retirements or personnel transfers. Thus, AR guidance supports on-site operations not only in accuracy and efficiency but also in human resource management.

Quality Assurance and Future Inspection Use via As-Built Records and Point-Cloud Scanning

The digital benefits continue even after accurate construction is completed with AR guidance. Combining as-built records and point-cloud scanning enables quality assurance and future use for equipment inspections.

As-built records after construction are the process of proving and inspecting that completed structures and equipment were installed according to design drawings. Traditionally, main dimensions were measured with tape measures or levels and recorded on paper drawings, but point-cloud scanning allows the entire site to be digitally recorded for verification. For example, using a smartphone LiDAR to scan around a newly installed signal pole or catenary pole yields detailed 3D point-cloud data capturing position, height, and tilt. Since these point clouds are high-accuracy and coordinate-referenced, they can be overlaid with 3D models or drawing data for comparison. Color-coded overlays of point clouds and design data make it immediately clear which parts match the design and where deviations exist, streamlining as-built inspections. Tools that comply with the 3D as-built management methods recommended by the Ministry of Land, Infrastructure, Transport and Tourism are being integrated into construction management, enabling point-cloud-based quality inspections that prevent oversight and detect rework areas early, thereby further improving construction quality. For railway installations, verifying completed positions and heights with 3D data including surrounding features allows office-based checks of things like signal sightlines or clearances between catenary lines and trees. By enhancing initial construction accuracy with AR guidance and advanced inspection via point-cloud as-built data, quality assurance is doubly secured.

High-precision point-cloud and coordinate data acquired during construction are also extremely useful for post-construction maintenance. Railway equipment undergoes displacement and deterioration over time due to train loads, so periodic inspections and monitoring are essential. Using coordinates recorded at construction as a baseline, remeasuring the same points after a period can detect tilting or settlement of signal poles or catenary poles at the millimeter level (millimeter-level (≈0.04 in)). Whereas traditional distortion measurements could only handle a few points, GNSS can measure many points quickly, making broad condition assessments easier. For example, track irregularity inspections can record key coordinates on the track in advance and compare differences in periodic measurements. Similarly, detecting equipment position changes or tilts via point-cloud comparison supports early detection of anomalies. Point-cloud data also has potential applications in infrastructure inspections: scanning an entire bridge or tunnel structure provides a baseline for tracking changes over time. Scanning a bridge pier, for instance, allows quantitative evaluation of cracks or deflections in later scans by comparing with past data. Damage records that were once kept manually or on paper can be marked and stored in point-cloud space to create a precise maintenance ledger. Thus, digital data obtained at construction becomes an asset throughout the asset lifecycle, contributing to advanced site management for future equipment renewal and inspection planning.

Sharing point-cloud and photo data via the cloud also seamlessly connects site and office operations. For example, in the LRTK system, point clouds and geotagged photos captured on site are immediately saved to the cloud for real-time office-side confirmation. This creates efficiencies such as auto-generating as-built diagrams and report bases by the time fieldwork completes. Measurements on point clouds, backfill volume calculations, and other analyses can be performed in the cloud, letting site agents and managers focus on safety while processing data remotely. Railway construction involves many stakeholder departments, but centralized real-time information sharing via the cloud aligns understanding among internal and external parties, reducing miscommunication and facilitating collaboration.

Workflow Utilizing Smartphone RTK + AR + Point Clouds Realized by LRTK

One solution that provides the technologies discussed above in a one-stop manner is a system called LRTK. LRTK is a smartphone RTK positioning system developed by a venture originating from the Tokyo Institute of Technology, consisting of a pocket-sized high-precision GNSS receiver that can be attached to a smartphone, a dedicated surveying/AR app, and a cloud service. Introducing this package enables the site smartphone to function as a high-precision surveying instrument, 3D scanner, and AR viewer, allowing seamless handling from positioning to point-cloud measurement, as-built recording, and AR usage.

Below is a case-style walkthrough of railway equipment construction using LRTK.

① Preparation (registration of design data): The project staff first upload design coordinate data and 3D models for equipment such as signal poles, catenary poles, and cable routes to the LRTK cloud. For example, the reference coordinate system is unified with the public coordinate system (JGD2011), and planned values from drawings are registered as-is. On site, smartphones will have a prepared list of target equipment positions.

② Start positioning on site: Upon arrival, workers attach the GNSS antenna to the smartphone and launch the LRTK app. The antenna is small and can be mounted on a helmet or a telescopic pole tip, so it does not get in the way. The smartphone begins receiving RTK correction information from base stations over the mobile network and immediately measures the current position with centimeter precision (half-inch-level). Base stations can be the Geodetic Control Point network or dedicated base stations set near the site in advance. Workers do not need complex equipment settings; when the smartphone displays “RTK Fix” (establishment of high-precision reception), positioning is ready.

③ AR guidance to position: The worker selects the equipment to be guided in the app—e.g., “XX signal foundation center.” The screen then displays the direction and distance to that target. By following arrow navigation or voice guidance, the remaining distance decreases until the app shows “arrived” or “here it is!” Looking through the smartphone camera, workers can see AR markers (pins or flags) indicating the installation position overlaid on the real scenery. Workers mark the spot with stakes or spray, designating the signal foundation center. Repeating this for other points completes multiple markings in a short time. One person can operate the smartphone while others mark, enabling teamwork. LRTK’s interface is simple and intuitive, allowing even those uncomfortable with machines to use it without difficulty.

④ Construction and on-the-fly checks: After marking, excavation and installation proceed. LRTK is used for checks at key stages. For example, when installing anchor bolts for a signal foundation, the worker can remeasure the foundation center with the smartphone to confirm there is no displacement. AR can display design models (such as the anchor plate shape) to verify rebar positions and formwork dimensions on the spot. Instead of relying on drawings and tape measures, the AR view constantly shows the “correct answer,” enabling immediate correction when a discrepancy is noticed.

⑤ Point-cloud scanning for as-built records: After construction, LRTK is used to digitally record as-built conditions. Using the smartphone LiDAR or camera, workers capture a 360° scan around a signal foundation or exposed trough piping to generate a point cloud. The LRTK app starts and stops scanning with a single button; the acquired point-cloud is automatically scaled and merged and then uploaded to the cloud. RTK-provided absolute coordinates are attached, so point-cloud data are tied to real geodetic coordinates (latitude/longitude/height). For railway site work, coordinates are unified in the local plane coordinate system, and all data are managed in public coordinates. On the cloud, a 3D mesh model is automatically generated from the point cloud and can be viewed in a web browser by stakeholders. From point-clouds of buried cables, cross-sections can be extracted to measure diameters or compute backfill volumes immediately.

⑥ As-built inspection and report generation: Compare point-cloud data uploaded to the cloud with the design data to perform as-built inspections. Selecting a design coordinate on the LRTK cloud automatically searches the corresponding point in the point cloud and computes as-built dimensions. For a signal foundation, the system might list “difference from design = +2 cm east-west, -1 cm north-south, height difference +0.5 cm.” If all values are within tolerances, records and photographs are linked in the cloud and saved as electronic as-built deliverables. Photos taken by site workers are automatically tagged with measured coordinates, so information like “XX signal foundation center (X=..., Y=..., H=...)” is associated with each photo. These outputs can be formatted to comply with internal electronic delivery standards, greatly reducing the effort to compile reports.

⑦ Handover for maintenance: The data acquired and organized during construction are handed over for future maintenance operations. Maintenance staff at railway companies can reference LRTK cloud as-built data to understand the precise position, height, and structure of newly installed equipment. For example, during a later tilt inspection of a signal pole, comparing the current measurements with the point-cloud model from construction lets inspectors quantify the tilt. For buried cables, future excavators can use LRTK on site to “see through” and confirm buried objects in AR, preventing accidental damage. The ability to use data consistently from construction through maintenance is a major advantage of LRTK, improving safety and reliability across the equipment lifecycle and supporting more efficient inspections and renewal planning.

This workflow is one example, and various DX (digital transformation) cases using LRTK are already emerging in the field. Examples include using an iPhone + RTK receiver on a tripod to rapidly establish public-coordinate reference points in civil works, and attaching high-precision coordinates to smartphone photos for time-series comparisons in infrastructure inspections. In the railway sector, LRTK has been used to quickly position tracks and structures and tag construction photos with accurate coordinates. The compact, easy-to-handle LRTK unit is unobtrusive for night or high-elevation work and can be used quickly when needed, significantly improving on-site surveying efficiency.

Closing

The introduction of RTK + AR guidance at railway equipment construction sites is a groundbreaking advance that bridges people and technology. By supporting surveying and marking work that once depended on seasoned technicians with digital technology, a balance of accuracy and efficiency is becoming achievable. Overcoming site constraints such as labor shortages and night work to create an environment where anyone can perform millimeter-level positioning is essential to maintaining safe and resilient railway infrastructure. From a practical standpoint, careful validation is needed at the start, but once its effects are experienced, many say “we can’t go back to the old way.” Improved surveying accuracy, reduced work time, fewer human errors, and promoted skill development—the benefits of high-precision positioning × AR open new possibilities for the future of railway sites.

It is expected that more railway companies and construction sites will adopt such digital technologies, advancing “railway equipment × DX.” In safety-first environments supporting train operations, mastering modern technologies as practical solutions is one answer to worsening labor shortages and increasingly stringent safety standards. The wave of smart construction using RTK and AR is steadily spreading to improve efficiency and sophistication in railway equipment construction. Consider embracing this trend and challenging your site’s DX initiatives.