Disclosure — some links on this page go to Amazon, and we earn a commission if you buy through them. It costs you nothing extra and it does not change what we recommend or in what order. The full disclosure.
LiDAR spins a laser to measure the room and works in complete darkness, but the turret makes the robot taller. vSLAM builds the map from camera images, allowing a lower body but needing light. Solid-state sensing gets most of LiDAR's accuracy with no turret, which is why the thinnest current flagships use it.
The three approaches
LiDAR
vSLAM (camera)
Solid-state / dToF
How it maps
A spinning laser measures distance in a 360-degree sweep
A camera tracks visual features and infers position
Fixed time-of-flight sensors and structured light, no moving turret
Works in the dark
Yes — it supplies its own light
Poorly. In an unlit room it is navigating from memory
Yes
Body height cost
The turret adds height, typically to about 100mm
None from navigation itself
None — the thinnest models here use it
Typical failure
Mirrors and floor-to-ceiling glass confuse the returns
Dim rooms, plain walls, and rooms that have been rearranged
Newer, so less field history; relies on sensor cleanliness
Privacy question
No camera required for mapping
A camera is fundamental to how it works
Often no camera for mapping, though many models add one for obstacles
Examples here
Roborock PreciSense; Shark 360-degree LiDAR
iRobot PrecisionVision
Roborock StarSight; ECOVACS dToF with AIVI 3D
Compiled from the manufacturers' published navigation descriptions on 2026-09-01. Every entry is linked to its source in the table below.
Navigation system by model, as each maker names it
23 of 23 models publish this figure. Every number below links to the page it came from.
Compiled from the manufacturers’ own published material on 2026-09-01. We have not measured any of these figures ourselves — see how we pick. Found an error? Tell us and we will correct it.
Why the turret matters more than the technology
For years the practical difference between LiDAR and camera navigation was not map quality but height. A spinning turret has to sit above the body, which pushed LiDAR robots to around 100mm — Roborock's Qrevo Curv is 104mm — while camera-based Roombas came in around 86mm. That is the difference between reaching under a sofa and stopping at it.
Solid-state sensing removes the trade-off. Roborock's Saros 10R uses StarSight rather than a turret and is 80mm tall, and the Qrevo CurvX retracts its turret to reach the same figure. If your dust problem is under low furniture, this is now the specification to shop on rather than the mapping technology itself.
Obstacle avoidance is not the same system
It is worth separating these, because the marketing does not. Mapping tells the robot where the walls are. Obstacle avoidance stops it eating a charging cable, and it usually runs on separate hardware — a forward camera, a structured-light projector, or both. A robot can have excellent LiDAR mapping and poor obstacle avoidance, and several do. See obstacle avoidance.
If cameras concern you
A vSLAM robot cannot navigate without its camera, so there is no setting that turns it off. A LiDAR or dToF robot maps without one, though many still carry a camera for obstacle recognition — which usually can be disabled at the cost of the avoidance feature. What each system actually collects is on what robot vacuums know about your house.
Quick picks
Ranked on the published specifications below. Tap a row to jump to the full write-up.
#
Product
Best for
Max climb
Price
1
RoborockNo photo
Roborock Qrevo CurvPublishes both a standard-threshold and a double-layer threshold figure, which is the number most sill-blocked buyers are actually shopping for.
iRobot Roomba j9+The vacuum-only j9+ is the model whose obstacle avoidance is specifically marketed against pet waste, which is a failure mode with a very high cost.
40mm (1.6in) double-layer, 30mm (1.2in) standardRoborock's own wording is that it can overcome standard thresholds up to 3cm and navigate double-layer thresholds up to 4cm. Stated maxima assume a clean square edge; a rounded or worn sill behaves differently.Source: Roborock Qrevo Curv product page — threshold crossing (retrieved 2026-09-01)
LiDAR maps more accurately and works in the dark. vSLAM allows a lower body and no turret. If your rooms are dim, LiDAR. If your problem is furniture clearance, look at solid-state models rather than either.
+ Robot vacuum lidar vs camera: which is more private?
LiDAR and dToF map without a camera, so a model that uses them for both mapping and obstacle detection collects no images. A vSLAM robot cannot work without its camera. Many LiDAR models still add a camera for obstacles — check what you are actually buying.
+ Do LiDAR robot vacuums work in the dark?
Yes, completely. The laser supplies its own light, which is one of LiDAR's real advantages — a camera-based robot in an unlit room is navigating from memory rather than observation.
+ What is gyroscope navigation?
The cheapest mapping approach: the robot estimates position from wheel rotations and a gyroscope, with no external sensing. It produces a rough map that drifts over a long run. None of the models on this page uses it as their primary system.
+ Why does my robot map a room that does not exist?
Usually a mirror or a large pane of glass. A LiDAR return from a reflective surface reads as open space beyond it, so the map grows a phantom room. Blocking the reflective surface at robot height during the first mapping run fixes it.
