Tutorials: SLAM algorithms

 

SLAM: Map types vs. algorithms

Not all SLAM algorithms fit any kind of observation (sensor data) and produce any map type. The following table summarizes what algorithms (of those implemented in MRPT) fit what situation.

↓Observations \ Maps → Occupancy grid
(Class
COccupancyGridMap2D)
Point maps
(Class
CPointsMap)
Landmark map
(class depends
on SLAM method)
Graph
of pose
constraints
2D laser scanner
(CObservation2DRangeScan)
X X
1D sonar range readings
(CObservationRange)
X X X
Features / Landmarks
(SLAM method-dependent)
X X X
Range-only sensors
(CObservationBeaconRanges)
X X X
Relative poses X X X