We consider range-sensing mobile robots operating under significant localization uncertainty that are tasked with exploring and mapping unknown environments. This talk will describe an approach that seeks a suitable tradeoff between exploring such environments quickly and efficiently, while ensuring that the maps constructed are accurate. We propose to do so using virtual maps, an abstraction which allows a mobile robot to predict the mapping outcomes that will result from a wide variety of possible sensing actions. A novel active perception framework will be described that leverages virtual maps to guide a robot's planning and decision-making as it explores unknown environments under localization uncertainty. Examples will be discussed involving lidar-equipped unmanned ground vehicles and sonar-equipped underwater robots, which include experimental outcomes of fully autonomous exploration in both settings.
Dr. Brendan Englot received S.B., S.M. and Ph.D. degrees in Mechanical Engineering from the Massachusetts Institute of Technology in 2007, 2009 and 2012, respectively. At MIT, he studied motion planning for surveillance and inspection applications, deploying his algorithms on an underwater robot to inspect Navy and Coast Guard ships. During 2012-2014, Brendan was with United Technologies Research Center in East Hartford, Connecticut, where he was a Research Scientist and Principal Investigator in the Autonomous and Intelligent Robotics Laboratory and a technical contributor to the Sikorsky Autonomous Research Aircraft. At Stevens Institute of Technology, he directs the Robust Field Autonomy Lab, which focuses on robust autonomous navigation solutions for robots operating in harsh and unstructured environments. Brendan received an NSF CAREER award in 2017 and an ONR Young Investigator award in 2020. In 2018, he was appointed the Geoffrey S. Inman Endowed Junior Professor of Mechanical Engineering at Stevens Institute of Technology.