LiDAR (Light Detection and Ranging) technology has become a game-changer in various industries as sensors become more compact and accessible over the years. From NASA's Apollo 17 laser altimeter that weighed 23 kg, to the first commercially available, mass-produced real-time 3D LiDAR, and now, a fingertip-sized solid-state LiDAR chip developed by scientists at the University of California, Berkeley.
As LiDAR and simultaneous localization and mapping (SLAM) solutions move from research labs to industrial applications, industries including architecture and construction to autonomous vehicles and forestry management have benefitted from rich and relatively affordable data.
LiDAR Sensor AttributesLiDAR sensors emit light in the form of laser pulses to measure distances. They are known for their precision and ability to capture detailed 3D data. However, like any technology, they have their limitations and specific attributes to consider:
- Line of sight: LiDAR can't "see" through solid objects or heavy smoke. It's important to plan your scanning missions accordingly, ensuring that critical areas aren't obstructed.
- Reflective surfaces: Highly reflective surfaces like glass or metal can cause multiple reflections or "echoes," which can lead to inaccuracies in the data.
- Range: LiDAR sensors have a maximum scanning range, which may be affected by the reflectivity of surfaces scanned. Ensure you're scanning within this range to capture accurate data.
- Field of View (FOV): LiDAR sensors have different FOVs, and it's important to understand the FOV of your particular scanner, so you can position it in a way that allows it to capture the space at the correct angle.
- Plan and prepare: Thoroughly plan your scanning process, considering the specific environment and potential challenges. Clear the area of moving objects and people as much as possible. Keep your scans within 15 minutes and ensure loop closures for optimal results.
- Use targets and control networks: Deploy reference targets within your scanning area to aid in data registration.
- Post-processing: Invest in robust post-processing software like FJD Trion Model to clean and refine your point cloud data.
- Documentation: Keep detailed records of your scanning process, including scanning parameters and any adjustments made to account for challenges. Clearly name and label your scan tasks for easy organization.
Challenges to Consider and Best Practices for Addressing Them
As you scan, be mindful of the following factors that can affect data quality:
- Moving objects/people: LiDAR sensors are sensitive to movement. Any moving objects or people within the scanning area can result in inaccuracies. Ideally, clear the area before scanning, or point the scanner away as you encounter excessive movements. Door stoppers come in handy for securing doors.
- Narrow corridors: Tight spaces can limit the LiDAR's field of view, potentially causing data gaps or errors like drift. Carefully plan your scanning path in narrow areas, add features by placing chairs or tables if possible. It's best to scan corridors separately from your main scan tasks so you can isolate issues more easily.
- Glass and transparent surfaces: LiDAR pulses can pass through glass, making it challenging to capture data only on one side. Reflective surfaces like mirrors might "confuse" the LiDAR, and it's recommended to cover them before you scan.
- Large outdoor spaces: Open environments with few structures present challenges. Use a control network with control points measured by total stations or checkerboard targets, and use RTK-assisted mapping for optimal results.
- Transitioning between indoor and outdoor spaces: The change in lighting conditions, surfaces, can feature availability when transitioning from indoor to outdoor spaces can pose challenges. Move slowly and point the scanner so that it "sees" both the last part of the room and the new space you're entering.
Why Closing Loops MattersLoop closure is a critical concept in LiDAR scanning. It involves ensuring that the scanned environment's starting point matches the endpoint, creating a continuous loop with an overlap ranging from 5-10 meters.
Closing loops is essential for several reasons:
Data Integrity: It helps to validate the accuracy of the data. LiDAR sensors accumulate error as time goes on, and loop closures involve revisiting a known location during scanning, allowing the system to correct any accumulated errors, align the newly acquired data with previously captured data, and minimize drift.
Data Stitching: When scanning large areas, closing loops allows you to merge separate scans into a seamless point cloud.
Having some overlap between the two ends of the building, especially in the staircases, will ensure proper stitching.
Quality Control: Closing loops simplifies quality control. Overlapping datapoints make it easier to identify and rectify errors in your scans. For example, the following scan paths produced data with a good amount of overlap, even though the shape of the walking path was not exactly a circular loop.
- Evaluate the site ahead of the scan
- Avoid moving people or objects
- Initialize on a flat surface
- Close the loops (Big and Small)!
- Transition carefully between spaces
Here's a simple checklist that summarizes tips on how to set your scanning missions up for success.
LiDAR scanning is a powerful tool for creating accurate 3D models and capturing spatial data. By understanding LiDAR's limitations and following best practices, you can overcome challenges and ensure the quality and reliability of your data.
Watch Webinar to Learn from Our Expert User
To learn more, watch our webinar on how to set up your workflow for optimal scanning results. Hear from laser scanning expert and certified land surveyor Jakub Kałaska to learn about how he applied the best practices for BIM, stockpile measurement, and infrastructure surveying projects.