Reliable Collaborative Low-complexity 3D Indoor Pedestrian Navigation


Maria Garcia Puyol (PhD Student, Institute of Communications and Navigation Oberpfaffenhofen, DLR, Wessling)


FootSLAM is a form of simultaneous localization and mapping (SLAM) for pedestrians that addresses both the problems of indoor positioning and mapping. It is based on human odometry (aka pedestrian dead reckoning), e.g. with a foot-mounted inertial sensor and uses a Rao-Blackwellized particle filter to estimate the building layout and the pedestrian's position and orientation. FootSLAM divides the 2D space into a grid of uniform and adjacent hexagons and maps the number of times each particle crosses the edges of the hexagons it visits. The algorithm explores many different odometry error hypotheses, settling on those which best fits the previous pose history. Hypotheses in which the pedestrian revisits certain areas in the environment are rewarded and over time a reliable 2D map of essentially walkable areas" is constructed. We can include data sets obtained from other walks in the same building in the form of a prior map, helping the convergence of the algorithm. This has led to the development of FeetSLAM, the collaborative form, whereby different walks in the same building are used to obtain a combined map in an offline process. FootSLAM shows a computational complexity that grows quadratically with time and it is only able to map in 2D. During the talk, a new complexity-reduced FootSLAM implementation with 3D mapping abilities will be shown. Furthermore, different methods to reduce the computational complexity of FeetSLAM's combination of maps step will be proposed to make it a scalable technology suitable for wide adoption of collaborative indoor pedestrian navigation."


María Garcia Puyol was born in Malaga, Spain. She began to study Telecommunications Engineering at the University of Malaga in September 2005. During the academic year 2008/2009 Maria studied at the Polytechnic of Milan, Italy, as part of the Erasmus exchange program. Between November 2010 and June 2011 she did her Diploma Thesis at the Institute of Communications and Navigation of the German Aerospace Center (DLR), obtaining the Mark of Excellence Award. Since October 2011 she has been carrying out her doctoral studies at DLR this time as part of the DLR-DAAD fellowship program. In April 2012 María obtained the Best Academic Record Award and the Best Diploma Thesis Award by the University of Malaga.