Robot Mapping and Navigation with Learning and Sensor Fusion - Symposium 2024

In this talk I will focus on multi-sensor state estimation and 3D mapping methods for dirty, dark and dusky environments - underground mines, natural environments such as forests and construction sites. Fusing vision, inertial, lidar and kinematic sensing creates a variety of algorithmic challenges but also promises redundancy and complementarity.
Additionally by leveraging learning, a robot or sensor system can better understand its surroundings and avoid degenerate failure modes. A variety of demonstrations will be presented on closed loop quadrupeds, drones and handheld mapping systems.
This talk was given at the Hi! PARIS Symposium 2024 by Maurice Fallon, an Associate Professor at University of Oxford. He leads the Dynamic Robot Systems Group which focuses on perception, mapping and navigation and focuses on dynamic robots (quadrupeds, handheld systems and drones). Originally from Ireland, Maurice did his PhD on audio source tracking in at the University of Cambridge. He was a post doc working on marine navigation in John Leonard’s Marine Robotics Group after that. From 2012-2015 he was the perception lead of MIT’s team in the DARPA Robotics Challenge developing state estimation algorithms for the Boston Dynamics Atlas robot. He moved to Oxford in 2017 to take up a prestigious Royal Society University Research Fellowship. He has been a PI on several large UK and EU collaborative projects including deploying platforms measuring radiation maps of Chernobyl nuclear power plant, exploring mines and creating digital models of forests.