This paper uses the smoothing and mapping framework to solve the SLAM problem in indoor environments; focusing on how some key issues such as feature extraction and data association can be handled by applying probabilistic techniques. For feature extraction, an odds ratio approach to find multiple lines from laser scans is proposed, this criterion allows to decide which model must be merged and to output the best number of models. In addition, to solve the data association problem a method based on the segments of each line is proposed. Experimental results show that high quality indoor maps can be obtained from noisy data
@article{arxiv.1402.4907,
title = {Line Maps in Cluttered Environments},
author = {Leonardo Romero and Carlos Lara},
journal= {arXiv preprint arXiv:1402.4907},
year = {2014}
}
Comments
Advances in Artificial Intelligence Lecture Notes in Computer Science Volume 6437, 2010, pp 154-165