Technology trends in construction surveying

Construction surveying has had a louder marketing cycle than usual in recent years. Some of the trends being talked up are genuinely transforming the industry; others are gimmicks that won’t survive the next budget squeeze. Here is an honest read of what is changing and what to take seriously.

Multi-channel array GPR

Array GPR — multiple antennas firing in parallel from a single platform — has moved from speciality to mainstream over the last decade. A single pass of an array system covers what previously required many passes of a single-channel one. The result is much higher coverage rates, denser datasets, and the ability to survey site-scale areas in commercially realistic time.

For utility mapping at PAS 128 levels, array GPR is now the default. For highway and pavement work, vehicle-mounted arrays have reshaped how subsurface investigation is done. The cost-per-square-metre has fallen significantly, and the data quality has gone up.

This is a real trend, with real impact. Treat it seriously.

Mobile and SLAM-based LiDAR

Traditional terrestrial LiDAR is tripod-based: set up, scan, move, repeat. Mobile LiDAR — handheld and backpack-mounted scanners using simultaneous localisation and mapping (SLAM) — moves through a space and captures continuously. For appropriate applications (interior as-built, repeat captures, areas with constrained access), it is faster than tripod-based capture by a wide margin.

Accuracy is the trade-off. SLAM-based mobile LiDAR is typically less accurate than registered tripod-based capture by an order of magnitude — centimetres rather than millimetres. For applications where centimetre accuracy is sufficient, it is a transformative tool. For engineering-grade capture, tripod-based LiDAR remains the right choice.

The honest read: another real trend, but one to apply with judgement about accuracy.

Drone-LiDAR integration

A drone carrying a LiDAR payload is increasingly affordable. The combination — aerial reach and laser-measured accuracy — opens up applications that neither method alone handles well: tall structures with fine geometry, complex industrial sites, mixed indoor-outdoor capture.

For most general-purpose drone surveying, photogrammetry is still the right choice — cheaper, lighter, and sufficient for the accuracy. For specialised applications (façades, complex structures, high-vegetation sites where photogrammetry struggles), drone-LiDAR is now a viable option.

A trend to watch but not yet to default to.

ML-assisted interpretation

Machine learning is being marketed heavily across surveying. The honest assessment varies by application:

• Point cloud classification. ML-assisted classification of LiDAR clouds (separating ground from vegetation, walls from floors, etc.) is genuinely useful and increasingly built into mainstream processing software.
• GPR feature picking. ML tools that flag candidate reflectors in GPR data speed up the slowest part of GPR processing. They do not replace qualified interpretation, and any tool that claims to do so should be treated with scepticism.
• Defect detection in imagery. ML-assisted defect detection on drone imagery is improving but is still very dependent on training data. For high-stakes inspection, human review of the detection output is essential.
• Automated BIM modelling from point clouds. This has been promised for years and remains partial. ML can accelerate certain modelling tasks but does not yet produce engineering-grade BIM models without manual cleanup.

The pattern: ML is most useful as an assistant to qualified people, less useful as a replacement for them. Tools that fit the first description tend to deliver value; those that fit the second tend to disappoint.

Continuous monitoring

Permanently installed or repeat-deployed sensing — tilt sensors, strain gauges, repeat LiDAR scans — is becoming common on critical infrastructure assets. Continuous monitoring captures behaviour over time rather than at a single point in time. For assets where movement, deflection, or progressive damage matters, it is a more powerful diagnostic than any single-point survey.

For UK construction, the most visible application is structural monitoring of bridges, retaining structures, and large excavations. The economics work where the cost of failure is high and the asset’s behaviour cannot be characterised in a single inspection.

A real trend with strong long-term traction.

Cloud-based collaboration

Survey deliverables increasingly live in cloud-based platforms — point clouds streamed from a server, BIM models accessed through a browser, drone imagery hosted on a project portal. The shift reduces the friction of moving large datasets between disciplines and supports better collaboration on complex projects.

For most projects, this is now the default. A surveyor still delivering data only as flat files on a memory stick is increasingly behind the curve.

What’s overhyped

A few trends being marketed harder than the evidence supports:

• “Self-driving” survey systems that promise zero operator skill. Survey is interpretation; interpretation requires qualified people. Self-driving promises tend to over-claim.
• Real-time BIM updates from drone imagery. Possible at low fidelity; not yet practical at engineering accuracy.
• Universal automated interpretation of GPR or NDT data. Useful as an assistant to qualified surveyors; not yet a replacement.

Treat any vendor claim along these lines with the scepticism it deserves.

What to take seriously now

If you are commissioning UK construction surveying work today, the trends most worth incorporating are:

1. Array GPR for site-scale subsurface work.
2. Combined LiDAR-and-photogrammetry capture for complex jobs.
3. ML-assisted but human-supervised interpretation.
4. Cloud-based deliverables and collaboration.
5. Continuous monitoring where the brief calls for it.

These are mature enough to deliver value now, and they distinguish modern survey work from the older alternatives. The trends not on this list are interesting to watch but not yet to base procurement decisions on.

The basic discipline — clear briefs, qualified people, calibrated equipment, defensible deliverables — has not changed. The technology around that discipline has, and the right surveyor today is one who has integrated the genuine improvements without being distracted by the marketing.