Why control comes first
On every project we take on, the first stake we drive is not part of the building — it is part of the control network. Control is the coordinate skeleton that every later measurement hangs from: the topographic survey, the stake-out, the as-built, the volume calculation. If the skeleton is sound, every drawing produced over the months that follow shares one consistent reference. If it drifts even a few centimetres, that error is silently inherited by everything downstream — and it usually surfaces at the worst possible moment, during construction.
A control network is simply a set of permanent, well-described points whose coordinates we have measured to a known, documented accuracy. The art is not in occupying points — it is in choosing the right tool for each part of the site and in budgeting the error so the final coordinates land inside tolerance.
Grounded in fieldwork, not theory
- 90
- instruments in the fleet
- GNSS, total stations, levels, scanners
- 1,000+
- survey projects delivered
- 3000+
- km of roads surveyed
Two instruments, two jobs
The modern control network is almost always a hybrid. We establish absolute coordinates with GNSS-RTK — a base receiver on a known point broadcasting corrections to a rover — because satellites give us a position tied directly to the national grid without needing to see another survey mark. But satellites need open sky. The moment we move under a flyover, beside a tower, or into a plant building, we switch to a total-station traverse: a chain of line-of-sight angle and distance measurements that carries the RTK coordinates into the shadows with millimetre-level relative precision.
Choosing your positioning method
| Criterion | GNSS-RTK | Total station | GNSS static |
|---|---|---|---|
| Typical relative accuracy | ±15–25 mm | ±2–5 mm | ±3–8 mm |
| Needs open sky | Yes | No | Yes |
| Needs line of sight | No | Yes | No |
| Speed per point | Seconds | Tens of seconds | Minutes–hours |
| Best for | Open-site control & topo | Built-up / indoor traverse | Long high-accuracy baselines |
There is no single winner — a real network uses each where it is strongest.
How we design and observe a control network
- 1
Reconnaissance: study existing maps and the national grid, then walk the site to pick control locations that are stable, intervisible where a traverse is needed, and safe from construction traffic.
- 2
Establish the base: occupy a known national-grid point (or run a static GNSS session to create one) so the whole network is tied to an official datum and a documented EPSG coordinate reference system.
- 3
Propagate with RTK: fix the open-sky control points from the base, holding short baselines and good satellite geometry (low PDOP) to keep coordinates tight.
- 4
Carry a total-station traverse: into canopy- or building-shadowed zones, measure a closed traverse so the misclosure can be checked and distributed.
- 5
Adjust and check: least-squares adjust the network, verify misclosure against tolerance, and re-observe any point that fails before a single deliverable is produced.
The accuracy budget
A coordinate is never a single perfect number — it is the sum of many small uncertainties. Treating accuracy as a budget is what separates a survey that holds to tolerance from one that merely looks precise on screen. Before we promise a client centimetre-grade results, we account for each contributor and make sure their combined effect stays inside the target.
A typical ±20 mm horizontal budget
| Error source | Typical contribution | How we control it |
|---|---|---|
| GNSS multipath & geometry | 8–12 mm | Short baselines, good sky view, low PDOP, longer occupation |
| Instrument & prism centering | 2–4 mm | Forced-centering tribrachs, checked optical plummets |
| Atmospheric / tropospheric delay | 3–6 mm | Short baselines, modelled corrections |
| Geoid model (for heights) | 10–30 mm | Project-agreed geoid, local benchmark ties |
| Network adjustment residuals | 1–3 mm | Redundant observations, least-squares adjustment |
Errors combine in quadrature, not by simple addition — which is why several small sources can still fit inside a ±20 mm target. · Typical field values; actual budget is set per project. See NGS RTK guidance.
Where positional error typically comes from
Instruments are tested, not assumed
Before a project, our total stations and levels are checked against ISO 17123 field procedures, and base coordinates are re-confirmed against a known mark. An out-of-adjustment instrument or a stale base coordinate does not announce itself — it quietly biases every reading until someone re-checks. The standard exists so that 'accurate' is a documented claim, not a hope.
Control in the field
Put control to work
References
- ISO 17123 series — Field procedures for testing geodetic and surveying instruments — International Organization for Standardization (ISO)
- Guidelines for Real-Time Kinematic (RTK) GNSS surveying and geodetic control — US National Geodetic Survey (NGS/NOAA)
- International Federation of Surveyors publications on professional and cadastral standards — International Federation of Surveyors (FIG)
- EPSG registry of coordinate reference systems and map projections — EPSG Geodetic Parameter Dataset