LiDAR

Definition: LiDAR (Light Detection and Ranging) is a remote sensing technology. It measures distances with pulsed laser beams and creates precise three-dimensional point clouds from them. The technology captures millions of measurement points per second and generates high-resolution 3D models of rooms, buildings, plants and objects.

In B2B mechanical engineering, LiDAR enables: capture of existing plants for digital twins, as-built documentation of production environments, dimensional measurement of large machines, layout planning and facility management, and spatial data for AR applications.

Accuracy: 1–10 mm, depending on the system. This is sufficient for industrial applications.

English: LiDAR (Light Detection and Ranging) German: Laserscanning, 3D-Laservermessung

How It Works

Basic principle: time of flight

LiDAR measures the travel time of laser pulses: 1. Laser emission: the scanner sends out a pulsed laser beam 2. Reflection: the beam hits an object or surface and is reflected 3. Reception: the scanner receives the reflected beam 4. Travel time measurement: the scanner measures the time between emission and reception 5. Distance calculation: distance = (speed of light × travel time) / 2

Result: each laser pulse gives one 3D point with X, Y and Z coordinates.

Scanning patterns

Scanners capture the environment systematically: - Rotating LiDAR: 360° scan, typical for terrestrial scanning - Solid-state LiDAR: fixed scanners with electronic control (faster) - Mobile LiDAR: installed on a vehicle or robot, captures data during movement

Capture rate: 50,000–2,000,000 points per second (depending on the system).

Point cloud as the result

The output is a point cloud: millions of 3D points that represent surfaces. Each point has: - Position: X, Y and Z coordinates - Intensity: reflection strength (material information) - Color: optional, via an integrated camera (RGB values)

LiDAR Systems for B2B Applications

Terrestrial LiDAR (tripod-based)

Stationary scanners on a tripod: - Range: 80–350 m - Accuracy: 1–3 mm - Application: factory floor scanning, plant documentation, building surveys - Advantages: highest accuracy, stable - Disadvantages: slow (several scan positions are necessary)

Example devices: Faro Focus, Leica RTC360, Trimble X12

Handheld LiDAR scanners

Portable scanners that scan while you walk: - Range: 30–100 m - Accuracy: 10–30 mm - Application: fast capture of large areas, facility surveys - Advantages: fast, flexible, no tripod setups - Disadvantages: lower accuracy than terrestrial scanners

Example devices: Leica BLK2GO, GeoSLAM ZEB

Mobile LiDAR (tablet/smartphone)

LiDAR in consumer devices: - Range: 5–10 m - Accuracy: 10–50 mm - Application: AR applications, room scanning, quick surveys - Advantages: the device is always with you, no extra hardware - Disadvantages: low accuracy, short range

Examples: iPad Pro LiDAR, iPhone Pro LiDAR (since 2020)

Drone LiDAR (aerial)

LiDAR installed on drones: - Application: outdoor terrain, building facades, large industrial sites - Advantages: fast capture of large areas, aerial perspective - Disadvantages: expensive, regulatory restrictions

Industrial Applications

Creating digital twins

LiDAR captures existing plants for digital twins:

Workflow: 1. LiDAR scan of the production plant or machine 2. The point cloud is the base geometry 3. Conversion to a 3D model (mesh generation) 4. Enrichment with IoT data 5. Live synchronization with sensor data

Benefit: the digital twin is based on the real, actual geometry and not on CAD data (which is often out of date).

As-built documentation

Capture of the actual state after installation: - Problem: CAD planning ≠ actual installation (changes during assembly) - Solution: a LiDAR scan after commissioning captures the as-built state - Result: exact documentation for maintenance, extensions and conversions

Application: complex production lines, process engineering, plant engineering.

Dimensional measurement

Precise measurement of large machines and components: - Quality control: compare the target geometry (CAD) with the actual geometry (LiDAR scan) - Reverse engineering: capture existing machines that have no CAD data - Tolerance check: identify dimensional deviations

Advantage over manual measurement: faster, complete, no accessibility problems.

Layout planning and facility management

Factory floor scanning for planning: - Inventory capture: the current state of the production hall - Free space analysis: where does the new machine fit? - Collision check: virtual placement of new plants in the scan - Infrastructure planning: supply lines, access routes

Workflow: LiDAR scan → import into planning software → virtual layout → collision check.

AR applications (spatial mapping)

LiDAR gives spatial data for AR:

Spatial awareness: LiDAR captures the room geometry in real time, so the AR system knows the environment: - Virtual objects stand on real floors - Occlusion: virtual objects disappear behind real walls or machines - Collision detection: virtual objects collide with real objects

AR maintenance applications: - The technician scans the machine with a LiDAR tablet - The AR system detects the machine geometry - Maintenance hints are placed exactly on the components

Plant visualization: - The customer scans the production hall with a LiDAR smartphone - The AR app places the new machine at true scale in the room - The customer checks the space requirements and the integration

Data Processing

From point cloud to 3D model

Raw LiDAR point clouds are unstructured. You must process them:

1. Registration (merge several scans): - Identify overlapping areas - Bring the scans into one common coordinate system

2. Cleaning: - Remove noise and outliers - Delete moving objects (people, vehicles) - Segment by object

3. Mesh generation: - Convert the point cloud → polygon mesh (3D model) - Algorithms: Poisson surface reconstruction, Delaunay triangulation

4. Texturing: - Apply photos to the geometry - Result: a photorealistic 3D model

Software: CloudCompare, ReCap Pro (Autodesk), Leica Cyclone, Faro Scene.

File formats

  • Point cloud: .las, .laz (compressed), .e57, .xyz
  • Mesh: .obj, .stl, .ply, .fbx
  • Project format: specific to the manufacturer (Faro, Leica)

Challenges

Reflection problems

LiDAR does not work well with: - Transparent surfaces: glass, acrylic (the laser goes through) - Highly reflective surfaces: polished stainless steel, mirrors (incorrect measurements) - Absorbent materials: deep black surfaces

Solution: combine it with photogrammetry, use special scanner settings.

Data volume

LiDAR scans produce very large amounts of data: - 1-hour scan: 10–50 GB point cloud - Large plants: hundreds of GB

Handling: powerful hardware, cloud processing, data reduction.

Costs

Professional LiDAR systems are expensive: - Terrestrial: €20,000–80,000 per scanner - Handheld: €15,000–40,000 - Service providers: €500–2,000 per day (scan service)

ROI: with frequent demand (plant engineering, facility management), the ROI becomes positive quickly.

Frequently Asked Questions (FAQ)

What is the difference between LiDAR and photogrammetry?

LiDAR: active measurement with a laser, works in all lighting conditions (also in darkness), direct distance measurement, higher accuracy (mm range), expensive.

Photogrammetry: passive image analysis, needs good lighting, calculates distances from images, lower accuracy (cm range), less expensive.

When to use which? LiDAR for precise measurement, photogrammetry for objects with a lot of texture. A combination is often best: LiDAR for the geometry, photos for the textures.

How accurate is LiDAR for industrial applications?

Terrestrial LiDAR: 1–3 mm accuracy. This is sufficient for plant documentation, quality control and as-built surveys.

Handheld LiDAR: 10–30 mm. Suitable for facility management and rough layout planning, but not for precision measurement.

Mobile LiDAR (iPad/iPhone): 10–50 mm. Sufficient for AR applications and room scanning, but not for precise measurement.

Rule of thumb: if you need an accuracy of <10 mm, use terrestrial LiDAR. For AR/VR and rough capture, mobile or handheld LiDAR is sufficient.

Can you use LiDAR scans for CAD models?

Yes, but with limitations. Workflow: LiDAR point cloud → import into CAD (Inventor, SOLIDWORKS) → manual modeling over the point cloud ("scan-to-CAD"). Automatic conversion (point cloud → parametric CAD) is possible only to a limited extent, for simple geometries.

Better for: as-built documentation, reverse engineering, collision checks. Less suitable for: new designs (parametric CAD modeling is better).

How long does a typical factory floor scan take?

Small production hall (500 m²): 2–4 hours (5–10 scan positions).

Large production hall (2,000 m²): 1–2 days (20–40 scan positions).

Post-processing (registration, cleaning, mesh): 1–3 days, depending on complexity.

Total: 3–7 days from the scan to the finished 3D model. An experienced service provider is faster.

How much does LiDAR scanning cost for a production hall?

Hire a service provider: €500–2,000 per day (incl. scanner, operator, basic post-processing). Typical: €1,500–5,000 for a medium-sized hall (incl. on-site work + post-processing).

Buy hardware: terrestrial LiDAR €20,000–80,000, handheld €15,000–40,000. This pays off at >10–20 scans per year.

Alternative: an iPad Pro with LiDAR (~€1,000) for simple AR applications and room scans.

Capture technologies: - Photogrammetry – image-based 3D reconstruction - Structured Light Scanning – projection-based capture - Time-of-Flight Cameras – depth sensing - SLAM – Simultaneous Localization and Mapping - 3D Scanning – general 3D capture

Applications: - Digital Twin – LiDAR-based digital twins - As-Built Documentation – documentation of the existing state - Reverse Engineering – reconstruction of existing parts - Facility Management – building management - AR Applications – spatial AR

Data processing: - Point Clouds – point cloud processing - 3D Model – mesh generation from scans - Mesh Processing – mesh editing - Registration – merging scans - Segmentation – object detection in scans

Systems: - Terrestrial Laser Scanning – stationary scanning - Mobile Mapping – mobile LiDAR - Aerial LiDAR – drone-based scanning - Handheld Scanners – portable systems - Smartphone LiDAR – consumer LiDAR