Methods for salient frescoes selection based mobile robots navigation

In this paper we consider the problem of non-metric/qualitative environment representation for frescoes based autonomous robot navigation. Within this framework, the problem of selecting meaningful set of frescoes and some possible solutions are analysed and compared. Experimental results using six different approaches (resemblance, barycentre, Hamming and Levenshtein distances, cross-correlation and selforganizing feature map neural network) are also provided