A 3D spinning LiDAR sensor helps robots and self-driving vehicles measure the shape of the world around them. It sends out laser pulses, measures how long they take to return, and uses that timing to calculate distance. By spinning rapidly, the sensor scans in all directions and builds a 360-degree map.
This lets a machine detect roads, walls, trees, people, and other vehicles even when it does not know the environment ahead of time.
Inside the sensor, many laser channels are aimed at slightly different vertical angles, so each rotation captures stacked layers of distance measurements. Each measured return becomes a point in 3D space, and millions of points can form a point cloud of the surrounding scene. Software then filters, groups, and tracks these points to identify obstacles and plan safe motion.
LiDAR is especially useful because it gives direct distance measurements with high spatial detail, though it can be affected by rain, fog, dirt, and reflective surfaces.
Key Facts
- Distance from time of flight: d = ct/2, where c is the speed of light and t is the round-trip travel time.
- A spinning 3D LiDAR creates a 360-degree horizontal field of view by rotating the laser and detector assembly.
- Multiple laser channels at different vertical angles produce stacked scan lines that form a 3D point cloud.
- Angular resolution describes the spacing between laser measurements, with smaller angles giving finer detail.
- Point cloud coordinates can be written as x = r cos(theta) cos(phi), y = r sin(theta) cos(phi), z = r sin(phi).
- Robots use LiDAR data for obstacle detection, localization, mapping, and path planning.
Vocabulary
- LiDAR
- LiDAR is a sensing method that uses laser light to measure distances to objects.
- Time of Flight
- Time of flight is the measured time between sending a laser pulse and receiving its reflection.
- Point Cloud
- A point cloud is a set of 3D points that represents the surfaces detected by a sensor.
- Angular Resolution
- Angular resolution is the smallest angle between two separate measurements that a sensor can distinguish.
- SLAM
- SLAM is simultaneous localization and mapping, a method robots use to build a map while estimating their own position.
Common Mistakes to Avoid
- Forgetting the factor of 2 in d = ct/2. The laser travel time is for the trip to the object and back, so using d = ct gives twice the true distance.
- Assuming LiDAR directly understands objects. A LiDAR measures points in space, and separate algorithms are needed to classify cars, pedestrians, walls, or road edges.
- Confusing horizontal field of view with vertical field of view. A spinning sensor may see 360 degrees around itself horizontally, but its vertical coverage is limited by the number and angles of laser channels.
- Treating every point as equally reliable. Reflections from glass, water, dark materials, rain, or fog can reduce accuracy or create missing and noisy data.
Practice Questions
- 1 A LiDAR pulse returns after 80 ns. Using c = 3.0 x 10^8 m/s, calculate the distance to the object.
- 2 A spinning LiDAR rotates at 10 revolutions per second and records 1800 horizontal samples per rotation. What is the horizontal angular spacing between samples in degrees?
- 3 A robot sees a dense flat wall in its LiDAR point cloud and a small cluster of moving points crossing in front of it. Explain how the robot could use this information differently for mapping and obstacle avoidance.