How does autonomous 3D mapping work?

The Artec Jet 3D scanner mounted to a drone.

Let’s start with an overview of traditional scene capture. Terrestrial laser scanning (TLS) is a well-established surveying technique that involves placing a scanner on a tripod, capturing everything visible from that position, then physically moving to another spot in the scene and repeating the process. Each scan produces a highly accurate point cloud, but stitching them together takes time – and for large, complex spaces, a number of setups can be required.

TLS remains an excellent choice where peak accuracy is the priority. But the technology has limitations when it comes to speed, flexibility, and working in environments that are difficult to access or too massive to cover efficiently from fixed positions.

Fundamentally different, 3D mapping is all about flexibility. Instead of scanning from fixed positions, mobile LiDAR mapping systems capture environments while moving through them. The key enabler is a technology called SLAM or Simultaneous Localization and Mapping – which allows the scanner to track its own position and orientation in real time, building a coherent 3D map as the operator walks, drives, or flies through a space.

LiDAR (Light Detection and Ranging): LiDAR emits rapid laser pulses and measures the time it takes for each to bounce back, calculating distance to surrounding surfaces. Millions of these measurements combine to form a detailed 3D point cloud of the environment.

SLAM (Simultaneous Localization and Mapping): SLAM algorithms allow scanners to track position while mapping at the same time. The technology continuously compares scans, matching overlapping geometric features to calculate how far your device has moved, without any need for georeferencing or satellite location data.

IMU (Inertial Measurement Unit): A motion sensor that tracks acceleration and rotation in real time. The IMU fills in the gaps between LiDAR scans, keeping positional tracking smooth and stable, especially during fast movement or moments where geometric features are sparse.

GPS/GNSS: Satellite-based positioning that applies real-world coordinates to scan data. On its own, standard GPS is accurate to a few meters, which is why it’s typically paired with correction techniques for professional applications. You may also encounter the abbreviation “GNSS.” This refers to satellite positioning technologies more generally – GPS is one specific type!

RTK (Real-Time Kinematic): A GPS enhancement that uses a fixed base station or correction network to achieve centimeter-level positioning accuracy in real time. RTK is what elevates satellite data from “roughly here” to “precisely there.”

Ground Control Points (GCPs): Pre-surveyed reference markers placed within the scanning area. Aligning your point cloud to these known positions during or after capture provides an independent accuracy check and improves overall georeferencing precision.

Georeferencing: The process of tying scan data to a real-world coordinate system: latitude, longitude, and elevation. Without georeferencing, a scan is geometrically accurate but effectively floating in space. With it, the data can be overlaid onto maps, integrated with CAD models, or compared against previous surveys.

Static terrestrial laser scanning is proven and precise, but it comes with inherent constraints that mobile SLAM scanning is purpose-built to overcome. The most obvious advantage is speed. A static scanner needs to be placed, leveled, and left to capture at each position – for a building, this could mean dozens of individual scans and a full day’s work. SLAM LiDAR scanners capture the same space in a single continuous walkthrough, often in under an hour.

That speed advantage compounds when you factor in site access. Tripod-based systems need stable, flat ground and a clear line of sight, which is fine in an open warehouse but impractical in a narrow mine shaft or cluttered industrial area. Mobile systems, whether handheld, worn as a backpack, or mounted to a drone, simply go where tripods can’t, turning underground and multi-story building capture into routine jobs.

The workflow itself is also more forgiving. Every time a static scanner is repositioned, there’s an opportunity for setup mistakes, missed overlap, or registration errors that only surface during post-processing. With mobile mapping, SLAM handles registration continuously, while a live point cloud preview lets the operator verify coverage. Gaps are caught on site, meaning there’s no need for repeat visits – which is vital whenever site access is limited.

All of this feeds into a lower project cost. Static LiDAR retains a clear edge in scenarios demanding peak single-point accuracy, but for the majority of large-scale applications, mobile SLAM mapping delivers usable results in less time.

Your level of aerial autonomy also depends on whichever mapping solution you choose to deploy. Attaching to a drone and flying manually is a cost-effective but risky approach (without additional guidance, you could crash). With combined hardware solutions like Artec Jet and Artec Twins, it’s possible to activate collision avoidance and clever SLAM algorithms that guide your payload beyond line-of-sight and help it navigate back to base.

The different levels of autonomy possible with the Artec Jet 3D scanner.

Architecture, Engineering & Construction (AEC)

Construction projects encounter problems when as-built and as-designed structures don’t match. Traditionally, verifying this means manual measurements, tape measures, and repeated site visits – a process that’s slow, error-prone, and expensive when discrepancies arise.

Artec 3D scanning makes it possible to discover and rectify errors much faster. Walk through a building mid-construction with a mobile SLAM scanner, and you have a complete, measurable 3D record of exactly what’s there, including walls, steelwork, and mechanical systems. That record can be compared directly against original designs to identify clashes.

In renovation, the same approach works in reverse. Scanning an existing building generates an accurate as-built model. It’s then possible to design around what’s actually in place rather than relying on outdated drawings. Digitizing buildings is also a great way of keeping stakeholders in the loop – and in cases where they have historic value, preserving them for future generations.

Mining (underground & overground)

Underground mines represent some of the most challenging conditions for any measurement technology. There’s often no natural light, extreme temperatures, and environments where sending people in can be dangerous. Without a GPS mapping signal, it’s also impossible to geolocate people, objects, and assets – or fly GPS-dependent drones for inspection.

These are exactly the conditions SLAM-based LiDAR scanning is designed for. Mounted to a drone, Artec Jet can be flown autonomously into shafts and stopes that are too unstable or inaccessible for crews, mapping the geometry as it goes and returning with a dense 3D point cloud. For accessible areas, handheld or backpack-mounting lets you walk through tunnels and processing plants, capturing all the data needed in a single continuous pass.

Autonomous mapping devices like Artec Jet can be deployed to map underground.

This approach isn’t restricted to shafts either. Digitize a stockpile, and you get an accurate volume calculation, so there’s no estimation required. Scan a tunnel profile to compare it with excavation designs. Capture the same heading repeatedly to track ground movement at a scale of just a few millimeters for early warning of structural changes. Site monitoring is such a broad field – but it’s one where Artec 3D scanning is a proven solution with headroom for growth.

Public safety

When responding to a public safety incident, it’s critical to accurately record the scene before it’s changed or swept away. This could be for response planning or investigation, but the important thing is not to disturb any evidence. Artec Jet can be carried through such environments to capture spatial evidence like the position of vehicles, debris, or structural damage.

After a building collapse or flood, 3D scanning is also ideal for damage assessment. Generating an accurate spatial model of the affected area is important for planning safe routes, assessing structural integrity remotely, and coordinating recovery efforts. When data is georeferenced, it can be overlaid onto maps as well – and shared with everyone working on the same incident.

Defense

Military and security operations often need to map unfamiliar environments quickly, accurately, and without putting personnel at unnecessary risk. GPS signals may be jammed or unavailable, lighting conditions may be poor or nonexistent, and the spaces involved, like tunnels, bunkers, and damaged buildings, are often too confined or hazardous for manual surveying.

SLAM-based mobile scanning addresses all of these constraints directly. Because the technology tracks its own position using the geometry of the environment rather than satellite signals, it works in GPS-denied settings without any external infrastructure. LiDAR’s active laser sensing means complete darkness has zero impact on data quality. And because the scanner can be mounted to a drone or ground robot, it can enter and map spaces autonomously, gathering detailed 3D intelligence without requiring anyone to physically go inside.

Resulting data has immediate tactical and planning value. A 3D model of a building or tunnel network gives a precise spatial understanding of the environment. Over time, repeat scans of the same site reveal changes that may not be visible in satellite imagery or on foot. Scanning provides the same as-built documentation monitoring benefits as seen in civilian construction, applied to forward operating environments where speed and accuracy matter most.

Factory planning

Manufacturing facilities are dense, layered environments with machines bolted to floors, piping running overhead, electrical conduits threaded through walls, and equipment arranged in layouts that may not have been updated on paper. When it comes to planning a new production line, relocating equipment, or retrofitting a facility, the first challenge is simply knowing what’s already there and where it all sits.

A warehouse captured in full color within minutes using Artec Jet.

3D scanning solves this by capturing the entire facility as it stands today. Walk through with a mobile SLAM scanner, and you produce a complete, dimensionally accurate model of the space. That model becomes the foundation for planning. Engineers can design new layouts, route piping, and check clearances digitally before any physical work begins.

For large plants, scanners can cover different scales. A mobile LiDAR scanner maps the overall building structure and floor layout, while a handheld structured-light scanner captures individual machines and parts at sub-millimeter detail. The end result: fewer surprises during installation, relocation, or modification – as well as a digital record of the facility for future planning.

Geospatial mapping

Surveying and mapping large outdoor areas have traditionally been a slow, labor-intensive process. Conventional methods involve collecting measurements point by point, which produces accurate results but takes time, and the output is typically sparse, so there’s enough data to define key features but not enough to capture full surfaces in detail.

Mobile LiDAR scanning changes the economics of geospatial data collection. A scanner mounted to a backpack, vehicle, or drone captures millions of points per second as the operator moves through the environment, producing a dense 3D point cloud in the process.

For road, railway, and pipeline mapping, a vehicle-mounted scanner captures kilometers of linear assets at driving speed, with RTK positioning providing real-world coordinates accurate to a few centimeters. For site surveys, a backpack or handheld setup covers open ground and transitions seamlessly into buildings or structures where GPS drops out, with SLAM maintaining positional accuracy throughout. Drone-mounted scanning adds an aerial perspective for stockpile volumes, earthworks, and terrain that’s difficult to traverse on foot.

Long story short, you’ve got a solution for tackling all kinds of terrain. Autonomous 3D mapping is perfect for capturing such large environments. Improvements to platforms like Artec Twins are only making it easier to process, visualize, and enhance the accuracy of resulting datasets.