Cooperation with humans is a requirement for the next generation of robots. Reason about what robots senses and knows, share and acquire knowledge in human voice interaction are capabilities required for the success of this cooperation. Instead of traditional SLAM (Simultaneous Localization and Mapping) methods, which do not interpret sensor information other than at the geometric level, these capabilities requires an environment map representation similar to the human.
Here it is described a new hybrid (metric and semantic) localization and mapping extension layer to the traditional SLAM algorithms. HySeLAM ( "hybrid semantic localization mapping system" is the name of this new extension.
Here you can find previous versions of hyselam code. At this moment we don't publish the latest versions, but if you are interested work or test the latest version, please send an email to fbnsantos (aAT) fbnsantos (Dott) com
This is a demo of SeloVis - Visual Place Recognition!
Filipe Neves dos Santos graduated with a degree in Electrical Engineering since 2003. He then pursued graduate studies at the Instituto Superior Técnico of Universidade Técnica Lisboa, completing a M.Sc. degree in Electrical Engineering - Automation and Systems in 2006. In 2004, he also worked as an assistant lecturer in the Electrical Engineering Department of the Instituto Superior the Engenharia do Porto. From 2007 to 2010, he started a high tech startup company for autonomous system monitoring. Since 2010, he has been a Ph.D. student and researcher at Engineering Faculty of University of Porto. His main research areas are in Process Control and Robotics, localisation of autonomous vehicles and Human-Robotics Interaction.