SkillsCast

Real-time localisation and mapping for mobile robots

22nd August 2016 in London at CodeNode

There are 1 other SkillsCast available from Post Quantum Security and Visual SLAM

Mobile robots have gained much popularity recently, with applications ranging from domestic settings to industrial inspection and autonomous driving. In order to accomplish any useful task, however, robots need dedicated sensing capabilities paired with suitable algorithms for localisation inside a potentially unknown environment. The recent years have brought better and cheaper sensors as well as ever more computational power, e.g. GPUs, which have all driven the development of real-time algorithms towards higher accuracy and robustness, as well as very dense and large-scale map reconstructions. This talk will introduce the problem formulation of vision-based localisation and mapping, as well as related algorithms and challenges, such as real-time constraints under limited computational resources. Examples presented range from hand-held camera tracking to domestic ground robot and small drone operation.

YOU MAY ALSO LIKE:

Thanks to our sponsors

Real-time localisation and mapping for mobile robots

Stefan Leutenegger

Stefan Leutenegger is a Lecturer in the Dyson Robotics Lab, co-leading it with Prof. Andrew Davison. His research is centred around autonomous robot navigation. This includes localisation and mapping with a suite of sensors, most importantly cameras, to be processed efficiently to yield accurate results at real-time. In the past, he has mostly worked with Unmanned Aerial Systems (UAS), in order to allow them to fly autonomously and close to the ground. Stefan received a BSc and MSc in Mechanical Engineering from ETH Zurich in 2006, 2008, respectively, and a PhD in 2014, working at the Autonomous Systems Lab of ETH Zurich on “Unmanned Solar Airplanes: Design and Algorithms for Efficient and Robust Autonomous Operation”.

SkillsCast

Mobile robots have gained much popularity recently, with applications ranging from domestic settings to industrial inspection and autonomous driving. In order to accomplish any useful task, however, robots need dedicated sensing capabilities paired with suitable algorithms for localisation inside a potentially unknown environment. The recent years have brought better and cheaper sensors as well as ever more computational power, e.g. GPUs, which have all driven the development of real-time algorithms towards higher accuracy and robustness, as well as very dense and large-scale map reconstructions. This talk will introduce the problem formulation of vision-based localisation and mapping, as well as related algorithms and challenges, such as real-time constraints under limited computational resources. Examples presented range from hand-held camera tracking to domestic ground robot and small drone operation.

YOU MAY ALSO LIKE:

Thanks to our sponsors

About the Speaker

Real-time localisation and mapping for mobile robots

Stefan Leutenegger

Stefan Leutenegger is a Lecturer in the Dyson Robotics Lab, co-leading it with Prof. Andrew Davison. His research is centred around autonomous robot navigation. This includes localisation and mapping with a suite of sensors, most importantly cameras, to be processed efficiently to yield accurate results at real-time. In the past, he has mostly worked with Unmanned Aerial Systems (UAS), in order to allow them to fly autonomously and close to the ground. Stefan received a BSc and MSc in Mechanical Engineering from ETH Zurich in 2006, 2008, respectively, and a PhD in 2014, working at the Autonomous Systems Lab of ETH Zurich on “Unmanned Solar Airplanes: Design and Algorithms for Efficient and Robust Autonomous Operation”.