For efficient long-term operation, robots need to adapt to the changes characteristic of natural and human-populated environments. We will present methods which enable robots to build and maintain spatio-temporal models of environmental dynamics and predict the future states of the environment they operate in. We will demonstrate how these capabilities contribute to the lifelong learning of socially intelligent behaviours, improve the efficiency of robot operation and increase their acceptance by the people they work with. We will focus on human-centric representations of time, and methods which can introduce the notion of dynamics to the environmental models used by state-of-the-art robots.
The discussion after the presentation will be chaired by Dr. Michael Londesborough.
You can ask Tomáš Krajník questions on the day via https://app.sli.do/event/oxnj5d20
Tomáš Krajník works in the mobile robotics domain with a particular focus on long-term mobile robot navigation in changing environments. He has proposed several methods of representing the uncertainty of changing environments leading to the gradual improvement of mobile robot efficiency over time. He has designed and implemented software libraries for reliable visual navigation and tracking, which have been used by the roboticists of NASA, EPFL, KIT, AIT and many others. He actively cooperates with several UK universities and industries as well as with research institutes across the globe.
He also contributed to the success of the Czech Technical University team during the most prestigious international robotics competitions – earlier this year their drones and robots won two gold and one silver medal at MBZIRC (The Mohamed Bin Zayed International Robotics Challenge) and at the same time Tomáš Krajník’s students from the Centre for Robotics and Autonomous Systems earned the highest score among self-funded teams and ranked third overall at the DARPA SubT Challenge, (Defense Advanced Research Projects Agency) where they demonstrated a multirobotic system for searching the areas affected by a theoretical disaster.
Michael Londesborough is a British scientist working at the Institute of Inorganic Chemistry of the Academy of Sciences of the Czech Republic, where he focuses on research into boron compounds. He has won three prizes for Czech / Slovak young inorganic chemists, collaborates on projects with the British Council in Prague and on popularization projects for the Academy of Sciences of the Czech Republic, Czech Television and the National Technical Museum. He focuses on making science accessible to the general public, teaching young people, and strengthening international relations between young scientists (FameLab competition). In 2009 the Academy of Sciences of the Czech Republic awarded him the Vojtěch Náprstek Medal for popularizing science.
Streamed live on the Czech Centre London Facebook page
Part of the AI SCIENCE CAFÉ SERIES 2020
Artificial Intelligence is the new phenomenon of the 21st century and has evolved out of the realm of science fiction and at best, a few technologically advanced minds and IT professionals, to everyday topics of discussion and one of the key drivers of the future progress of humankind. The objective of the AI Science Café series, which will consist of five presentations by high profile, world acclaimed experts, is to introduce, present and publicly discuss various topics that impact our professional as well as our personal lives. As AI topics clearly generate more questions than answers, the series will strive to unveil trends and expectations from the inevitable social and technological changes which the future will bring.
Part of the SCIENCE CAFÉ SERIES
Science Café is a series of popular science presentations and talks by the best Czech scientists and innovators of today presented by the Czech Centre London. Inspired by the phenomenon of Cafe Scientifique which was first established in the UK in 1998 and spread across the world as a place where anyone could come to discuss the latest ideas in science and technology over a cup of coffee or a glass of wine, Science Café is an informal forum for the discussion of current work and interesting scientific issues.
The AI Science Cafe Series is kindly supported by AVAST
Partner: MBZIRC 2020, DARPA Subterranean Challange 2020
