Robots: An Uncertain Revolution

Posted 12 Sep 2008 at 10:16 UTC by mwaibel

The latest episode of Robots dives into the revolution brought on by the field of probabilistic robotics. Claudio Mattiussi, Senior Researcher at the EPFL's Laboratory of Intelligent Systems, discusses the advantages of using a probabilistic or Bayesian approach to perceiving the world, adding a touch of history, philosophy and projection. Rather than opposing good old fashion artificial intelligence (GOFAI) or Brooks' behavior based approach, Mattiussi proposes that the "uncertain" revolution caused by the probabilistic paradigm offers lots of space for middle ground and may be very good news for robotics. To see how good our listeners are at making their own probabilistic predictions Robots just launched a most uncertain robot contest to estimate the outcome of a mundane robot cleaning task.