GraphSLAM

In robotics, GraphSLAM is a Simultaneous localization and mapping algorithm which uses sparse information matrices produced by generating a graph of observation interdependencies (two observations are related if they contain data about the same landmark).[1]

See also

References

  1. ^ Thrun, S.; Burgard, W.; Fox, D. (2005). Probabilistic Robotics. Cambridge: The MIT Press. ISBN 0262201623.