How LiDAR detects intruders in total darkness
Cameras need light. LiDAR doesn't. Here's how laser ranging builds a 3D picture of your campsite at night and tells a person apart from a swaying branch.
Ask most people how a security camera 'sees' at night and they'll mention infrared. That's true for vision — but vision alone is fragile. Fog, glare, deep shadow and clever movement all defeat a 2D image. LiDAR sidesteps the problem entirely by measuring distance with light, and it works just as well at midnight as at noon.
Light as a tape measure
LiDAR (Light Detection and Ranging) fires rapid pulses of eye-safe laser light and times how long each pulse takes to bounce back. Because the speed of light is constant, that time-of-flight converts directly into distance. Sweep those pulses across a scene thousands of times per second and you get a 'point cloud' — a live 3D map of everything around the van, accurate to a few centimeters.
LiDAR supplies its own light. Pitch black, moonless, fog — none of it matters, because the sensor isn't relying on the environment to illuminate the scene.
Telling a person from a branch
A point cloud is powerful because it carries shape and motion, not just brightness. A swaying branch is thin, anchored and moves in a predictable arc. A person is a dense, roughly human-sized volume that approaches with intent. Our on-device models classify clusters by size, shape and trajectory, so a deer browsing 12 meters away doesn't trigger the same response as someone closing in on the door.
- Range: know exactly how far away something is, in real units
- Volume: estimate the physical size of a moving object
- Trajectory: distinguish 'passing by' from 'approaching'
- Geometry: ignore flat surfaces, rain and small animals
Why pair it with a camera?
LiDAR knows there's a person-sized object two meters from your slider door. It can't read a face or a logo on a jacket. A night-vision camera adds that identifying detail but struggles with depth and darkness. Fuse the two and each covers the other's blind spot: LiDAR provides certainty about position and size, the camera provides recognizable footage. Together they're far more reliable than either alone.
A note on privacy and power
Because classification happens on the Jetson board inside your van, raw point clouds and video never have to leave the vehicle. That keeps your data private and means the system keeps thinking with zero signal — important when you're parked somewhere genuinely remote.
“Cameras tell you what something looks like. LiDAR tells you what it is and where. You want both watching the dark.”— Marcus Hale
Want to see how the pieces fit together end to end? Read our overview of how the full Turova system works, from sensor to alert.
