News
TEDx talk: old news, but available on YouTube .
Paper: A new paper on Dexterous grasping of objects from a single view is about to appear in the International Journal of Robotics Research. The link gives you the originally submitted version.
Paper: A new paper on REBA: A Refinement-Based Architecture for Knowledge Representation and Reasoning in Robotics has appeared in the Journal of Artificial Intelligence Research. This provides a complete formal system for task planning in a way that combines the advantages of default logics (here Answer Set Programming) and Bayesian reasoning (POMDPs).
Moving: I have now moved to Amazon, Berlin.
Paper: New paper on tactile
exploration coming
out in IJHR.
Talk: I gave the Seelye Fellowship Lecture at the
University of Auckland.
Media: An interview with Radio New Zealand
on AI and Robotics.
Listen here.
Paper: I am proud to publish
my first philosophy
paper, with Mihaela
Popa-Wyatt, appearing in Philosophical Studies.
Paper: Read our paper on
learning to predict for manipulation, in Autonomous Robots.
Paper: Read our paper on robot task planning in
AI Journal.
Paper: Read our paper on dexterous grasp
learning in IJRR.
Paper: Read our paper on robot
task planning with non-monotonic logic and POMDPs in
IEEE TRO.
Talk: I gave a
public lecture called "Robot Life:
A User's Manual".
Research
We create algorithms that enable robots to work in uncertain and
unfamiliar worlds. I work on autonomous robot
planning, architectures for robot intelligence, robots that
learn, robot manipulation, machine vision and general machine learning. One goal of my
work is to endow a robot with explicit representations of
what it does and doesn't
know, and of how its knowledge changes
under the actions it can perform. This ability allows
robots to plan in challenging worlds where they know little.
I am a member of the Intelligent
Robotics Laboratory, and of CN-CR, a centre for
neuroscience and robotics. In my own laboratory I am lucky
enough to work with a wonderful team of about a dozen incredibly talented
researchers, most of whom work on dexterous robot manipulation.
Previously, I was the Project Coordinator for the FP7 funded project
PacMan on robot
manipulation. I also coordinated the CogX
project on robots that plan and learn in the face of
knowledge gaps. I worked in the project Strands on long term
autonomy and spatial temporal mapping, and I was part of
the projects GeRT
on robot manipulation, and CoSy on Cognitive
Robotics.
Publications
A locally maintained list, or my Google Scholar Profile.
Movies
For video updates on our work see my personal YouTube Channel, and the PacMan channel.
Media
Some recent appearances in the media.
Contact Details
jeremy dot l dot wyatt @
gmail dot com