Talk (30 min): Inside a robot’s brain
Deep dive on how to make autonomous robots think
Join online: viscon.ethz.ch/livestream
Navigating from A to B is a trivial task for humans, even when it comes to dynamic or crowded environments. But does this also apply to autonomous mobile robots? In this talk we present the main challenges that robots face in the real world and give insights into the most recent algorithmic solutions to tackle them. From computer vision to motion planning and controls, we will outline the main technology components that constitute an autonomous robot’s brain.
Autonomous mobile robots will soon be carrying out repetitive, unhygienic and dangerous work in airports, railway stations, construction sites or department stores. This way robots contribute to making human workplaces more interesting and safer again thanks to their support - all while boosting operational efficiency.
For this important mission, autonomous robots need to be able to move independently and safely in all environments. This requires tackling a number of technological challenges such as navigating in complex environments shared with people, achieving precise positioning and safe motions even in the presence of dynamic obstacles. In addition, mobile robots have limited compute capabilities, thus requiring the development of highly efficient and sophisticated algorithms.
In this talk we give an introduction to the main software components that enable autonomous mobile navigation. To name only a few, we will answer the following questions: How to make robots precisely estimate their positions and maintain digital maps of their environments? How to make robots differentiate between static and dynamic obstacles? How to make robots plan and execute safe but efficient paths in complex environments?