Loop Closures

A loop closure is when you visit the same location twice in a dataset. Loop closures are an important part of scanning that help ensure the accuracy of the created dataset. Your earlier original path is shown as a solid line on the user interface. Cross this line to create a loop closure.

• Ideally, you should make a loop closure every 15 - 30 m. We also recommend making larger loop closures, such as returning to your starting point at the end of a dataset. Or, when scanning multiple floors in one dataset, returning to a visited location on the floor you started on.

• If loop closures are not possible we recommend retracing your path in the opposite direction.

• Loop closures are calculated during data processing. You will see the benefits of loop closures in your processed data, but not on the user interface of your device.

• If possible, perform loop closures between indoors and outdoors.

Shown below loop closures in corridors.

Loop closure over multiple floors

Scanning Large Spaces

When scanning large open spaces, we recommend that you move in structured rows. That way the lasers do not lose their point of reference. Walk in parallel to the shortest walls and try to map until the whole quality map is completely covered with color.

Example of Rows

Walk in rows with about 3m distance between the rows. Close the loops for better data quality.

360° Field of View

The device can "hold on" to structures behind it, if not hidden by scanner’s head. In some cases, like long corridors, it can improve SLAM if the device is held sideways so that the scanner is able to see the beginning and end of a corridor.”