Autonomous robots are being used to solve increasingly difficult real-world applications. These new uses are promising but highlight the limitations of current approaches. Advancing the state of the art to achieve them will require significant research on core problems in robotics, including motion estimation and motion planning.
This talk will present work to understand these problems and use this knowledge to design better algorithms. It will present both a brief look at multimotion estimation and a more detailed discussion on planning in continuous search spaces. This latter work unifies and extends informed, graph-based search (e.g., A*) and anytime, sampling-based planning (e.g., RRT*) to create informed, anytime sampling-based planning algorithms. These algorithms exploit universal properties of the robotic planning problem to perform better on many real-world robotic planning problems, especially in the presence of complex constraints or high state dimensions.
Jonathan Gammell is a Departmental Lecturer in Robotics in the Department of Engineering Science at the University of Oxford, where he founded and leads the Estimation, Search, and Planning (ESP) research group at the Oxford Robotics Institute (ORI). He holds a B.A.Sc. in Mechanical Engineering with a physics option from the University of Waterloo and a M.A.Sc. and Ph.D. in Aerospace Science & Engineering from the University of Toronto.
His research at ESP is focused on understanding the fundamental problems of robotics and autonomy and using this knowledge to develop theoretically well-founded algorithms. This work is tested on robots operating in complex environments, either independently or in collaboration with external partners, including NASA JPL, and is used widely in real-world robotic systems, from steerable catheters to full-sized autonomous helicopters.