There is a version of a great survey that nobody even notices happened.

No lane closures. No overnight crews in reflective vests. No targets bolted to deck surfaces. No personnel standing on live structures. No disruption to the vessels passing underneath or the aircraft taxiing overhead. The infrastructure keeps operating exactly as it did before the survey team arrived and when they leave, the data is survey-grade, validated, and delivered.

Whether roadways, railways or airports, it is increasingly the standard that infrastructure owners demand, and it is driving a quiet but significant shift in how leading survey firms build their workflows. The question is no longer just Can you hit the accuracy spec? It is, Can you hit it without leaving a mark?

Civil engineering and surveying firm McKim & Creed has been methodically building toward an answer. The firm’s LiDAR group has developed a hybrid static-plus-mobile workflow that delivers survey-grade point clouds on operationally constrained infrastructure without the disruptions that traditional target-based methods require. The approach was pressure-tested on one of the more demanding structures imaginable: a four-lane, 1969-era vertical lift bridge over an active navigation channel in Wilmington, North Carolina. 

Now it’s being carried into territory that makes the bridge project look straightforward.

The Geometry of Constraint

The Cape Fear Memorial Bridge in North Carolina presented McKim & Creed’s team with a clean statement of the problem. The North Carolina DOT (NCDOT) required a survey-grade 3D model of the full structure including lift span, landings, adjacent buildings, and approaches at centimeter-level accuracy. The constraints were equally clear: no lane closures or disruptions to vessel traffic or bridge lift operations. The bridge carries four lanes of continuous traffic, has no shoulders or bike lanes, and vibrates measurably under load. There was no safe position on the deck to set and occupy traditional survey targets.

“A traditional target-based mobile LiDAR approach would have required extensive lane closures and night operations to safely set and level targets on the bridge deck,” said Matt LaLuzerne, national director of Business Development and Geospatial Services at McKim & Creed. “Given the safety risks and operational disruption, we jointly explored alternative control strategies during scoping before final field methods were established.”

Instead of looking for a workaround, the team built a workflow that made the constraints irrelevant.

Virtual Control

The solution McKim & Creed developed centers on what they call a virtual control alignment, a methodology that replaces discrete physical targets with the geometric information already present in the structure itself.

The workflow combines two complementary collection methods. High-resolution static terrestrial LiDAR was deployed from safe positions on the eastern approach in the parking areas near the tender house and adjacent facilities. These static scans were tied directly to NCDOT baseline control using GNSS and conventional leveling. Simultaneously, a mobile LiDAR system ran the full bridge and approach corridors, with the western bank tied conventionally to NCDOT baseline control.

In post-processing, the team used the structure itself as control. Large planar surfaces captured in the static point cloud—concrete barriers, curbs, parapets, and deck panels—stood in for physical targets, providing the geometric anchors needed to calibrate and adjust the mobile dataset. 

“We prioritized large, continuous, structurally rigid surfaces,” LaLuzerne said. “The goal is to maximize geometric leverage and redundancy in the least-squares adjustment.”

The same least-squares adjustment framework used in traditional control-based corridor surveys was then applied. Residual analysis, horizontal cross-section comparisons, and vertical surface-to-surface checks validated the results against NCDOT Type A Terrestrial Mobile LiDAR accuracy requirements. The methodology met spec at the centimeter level.

Scan Stability

One problem that virtual control cannot solve on its own is vibration, such as the subtle movements of a structure under live traffic loads, which can blur a point cloud if the scanner isn’t chosen and operated with that environment in mind.

The team addressed this through variable pulse repetition rates and an on-board IMU with a built-in shock notification system. Near the tender house, where structural vibration from traffic loads is most significant, the team operated at 1,200 kHz, a rate that minimizes dwell time at any given scan position and keeps vibration effects in the realm of high-frequency noise rather than coherent geometric distortion. At positions on the approach and departure slabs, where vibration risk drops but distances to bridge features increase, the team reduced the repetition rate to between 140 kHz and 600 kHz, allowing sufficient time-of-flight for accurate returns at longer range.

When vibration levels exceeded acceptable thresholds, the IMU automatically flagged it. Through overlapping scan positions, appropriate repetition rates, and cloud-to-cloud adjustments, vibration effects were kept as filterable noise rather than systematic error.

The methodology met NCDOT Type A Terrestrial Mobile LiDAR accuracy requirements at the centimeter level.

No lane closures. No night work. No disruption to vessel traffic or bridge lift operations. Centimeter-level accuracy across the full structure. No fingerprints. Survey delivered.

The methodology that emerged from Cape Fear is now standard practice at McKim & Creed, incorporated into internal checklists and processing guidance that any team can implement without having been on the project.

“Anything that’s visible in the static scan that’s also visible in a mobile scan with really good control, this process can be done,” LaLuzerne said. “It’s driven by safety requirements, speed, and accuracy.” The firm sees it as transferable to high-traffic corridors, constrained urban environments, and any infrastructure where placing physical control is impractical or unsafe. 

What that looks like in its next iteration may be the most compelling demonstration yet of where the profession is heading.

Two Millimeters at Midnight

McKim & Creed recently kicked off a project that pushes that commitment into genuinely uncharted territory: high-resolution pavement inspection on an active commercial airport runway at night. The challenge isn’t control strategy this time—it’s imagery, illumination, and a workflow that nobody has proven out before.

Current pavement inspection practice on runways requires daytime shutdowns and specialized inspection crews walking or driving the surface under controlled conditions. McKim & Creed’s approach works differently: collection happens during brief scheduled nighttime windows, keeping airport operations and revenue generation largely intact—and delivering higher-resolution data in the process.

The team is using a dual-scanner mobile mapping system that pairs LiDAR with downward-facing cameras calibrated to the point cloud, capable of two-millimeter pixel resolution sufficient for pavement crack detection. The system’s dedicated pavement camera mount is optimized for close-to-nadir downward imaging of road surfaces—in daylight.

“We have to collect at night, so we need to illuminate the surface so the cameras can see,” LaLuzerne explained. “Our challenge is then calibrating cameras to the LiDAR and developing the workflows around that.”

To illuminate the surface, the team mounted light bars at varying heights and angles, adjusted camera gain settings, and dialed in vehicle speed to optimize image quality during nighttime conditions. Multiple tests were conducted on similar surfaces and conditions as part of project preparations.

The data has direct utility for airport rehabilitation planning. Two-millimeter resolution is sufficient to detect and characterize cracking, spalling, and faulting across the full runway surface, giving design teams what they need to differentiate repair from replacement on a section-by-section basis — decisions that can mean substantial cost differences in construction. 

“We believe this is a game-changer for the airport rehabilitation industry,” LaLuzerne said, “providing high-resolution imagery and topographic data for desktop review and analysis, saving significant effort in the field for the design team, all while maintaining normal operations and revenue generation for the airports.”

As far as LaLuzerne is aware, this specific combination—high-resolution ortho imagery from mobile LiDAR on an active runway, collected at night—hasn’t been done before. Testing and calibration began in spring 2026, with primary data collection beginning in June.

In the end, the measure of both projects is the same: survey-grade data, delivered without requiring operations to pause. The bridge didn’t close. The runway keeps running. Nobody noticed the survey team was ever there.

The Virtual Value

The value of any novel methodology ultimately comes down to a simple question: how well did it actually work, and how do you know?

On the Cape Fear Memorial Bridge, McKim & Creed validated the virtual control approach through three parallel checks: least-squares adjustment residual analysis, horizontal cross-section comparisons between static and mobile datasets, and vertical surface-to-surface differencing across the overlap zone.

The least-squares adjustment told the first part of the story. Residuals in the unadjusted mobile solution, the raw point cloud before the plane-based control was applied, were measurably larger than in the final calibrated result. The adjustment pulled those residuals down significantly, and critically, the pattern was random and well-distributed rather than systematic. 

Consistent directional offsets or position-correlated residuals would have flagged either a poorly conditioned plane network or an unresolved calibration error. Neither appeared.

Horizontal consistency was checked by running cross-sections through linear features—parapets, curb lines, barrier faces—that appeared in both the static and mobile datasets independently. The sections showed tight lateral agreement, confirming that the plane-based control was constraining the mobile dataset in the horizontal plane as effectively as physical targets would have.

The vertical check was the most direct test that geometric redundancy between two independently controlled datasets could substitute for carrying elevations across the river by conventional means. Surface-to-surface differencing between the final static and mobile point clouds across the eastern approach showed vertical agreement consistently at the centimeter level, meeting NCDOT Type A Terrestrial Mobile LiDAR accuracy requirements.

Taken together, the checks provided what a single-method approach couldn’t: independent confirmation from multiple directions that the adjustment had worked, and that the results were reliable enough to deliver as survey-grade data.