||Shoudong Huang, Zhan Wang, Gamini Dissanayake, Udo Frese
||Iterated D-SLAM Map Joining -- Evaluating its performance in terms of consistency, accuracy and eﬃciency
||409 – 429
||Autonomous Robots, special issue on Characterizing Mobile Robot Localization and Mapping
Abstract This paper presents a new map joining al-
gorithm and a set of metrics for evaluating the perfor-
mance of mapping techniques.
The input to the new map joining algorithm is a
sequence of local maps containing the feature positions
and the ﬁnal robot pose in a local frame of reference.
The output is a global map containing the global po-
sitions of all the features but without any robot poses.
The algorithm builds on the D-SLAM mapping algo-
rithm  and uses iterations to improve the estimates
in the map joining step. So it is called Iterated D-SLAM
Map Joining (I-DMJ). Both simulation and experimen-
tal results show that the I-DMJ algorithm is eﬃcient
because the information matrix is exactly sparse and
the size of the state vector only depends on the number
The paper proposes metrics for quantifying the per-
formance of diﬀerent mapping algorithms focused on
evaluating their consistency, accuracy and eﬃciency.
The I-DMJ algorithm and a number of existing SLAM
algorithms are evaluated using the proposed metrics.
The simulation data sets and a preprocessed Victoria
Park data set used in this paper are made available to
enable interested researchers to compare their mapping
algorithms with I-DMJ.
29. 10. 2009