Home

Technology

How a walk in the field becomes a point cloud

Where the points come from, how the system knows where it was at the moment of measurement, and why a laser scanner alone is not enough for surveying. No jargon, in four steps.

Point cloud
Point cloud of a viaduct over a road
LiDAR
Point cloud of a street with buildings and cars

Step 1: LiDAR

Distance measured with light

LiDAR sends laser pulses and measures how long they take to come back after hitting an object. That time gives the distance to the surface the beam hit.

The BP-32 scanner has 32 beams rotating 20 times per second, which gives up to 640,000 measurements per second. Each point describes a piece of terrain, a building, a road, a kerb or a tree — anything the scanner can see within its 300 m range.

Step 2: position and orientation

The scanner has to know where it is and which way it is facing

Distance alone is not enough. For each point to land in the right place on the map, the system combines data from four sources at every moment.

GNSS/RTK

The device position in the project coordinate system, e.g. PL-2000, with RTK correction.

IMU

Motion, tilt and orientation of the sensor head, measured 1000 times per second.

SLAM

Compares consecutive parts of the scan and refines the walking trajectory.

LiDAR

Distances to objects, which build up the dense point cloud.

SLAM
Point cloud of buildings along a road

Step 3: SLAM

A route built from the scan itself

SLAM analyses consecutive parts of the point cloud and looks for shared features of the surroundings. This corrects the position and orientation of the moving scanner — especially where the GNSS signal is weaker: between buildings, under trees, next to infrastructure.

In a typical survey you do not need to set out calibration points, as you often do for drone or aircraft flights. On linear jobs you can measure a few GNSS check sections, but it is not required to start work.

The system calculates the trajectory automatically and works straight away in the target coordinate system.

Step 4: the result

Point cloud, panoramas and trajectory

A survey gives you more than a file of points: photos and 360° panoramas, the device orientation, the walking trajectory and a complete data set for processing.

After processing, the data goes into GeoOffice. There you work with the point cloud, panoramas, cross-sections, the map and reports in one program.

See GeoOffice
Survey result
Point cloud of an arch bridge

The numbers behind a survey

Values from the BP-32 specification. They show the scale of data collected during a single walk.

32
laser beams
640,000
points per second
20 Hz
LiDAR head rotation
0.5–300 m
scanner range
~2 cm
point cloud accuracy
1000 Hz
IMU sampling

Full BP-32 specifications

See a survey like this on real data

At a free demo we show the hardware and the software in action, on an example close to your own jobs.

Book a demo