Projects
Pfunk Lab
My research goal is to design robots which can develop the manipulation capabilities of human learners through exploration in natural environments. In particular, I am interested in manipulation of objects that the robot may have never seen before. To that end, I work on robot intuitive physics, a machine learning approach for efficiently modeling arbitrary world dynamics in terms of a space of latent physical quantities. Critically, and unlike previous approaches to parameter learning, my model representation adapts to the data observed by the robot. This provides an online bayesian perspective to system identification, and will (I hope!) pave the way to more sophisticated autonomous behavior for humanoid robots.
Humanoids Lab
PR2 Cart Pushing
PR2 Cart pushing: an intern project onat Willow Garage.
uNAMO
A hierarchical MDP solution to the problem of navigation among movable obstacles.
SPLinTER 1
Learning policies for 2D manipulation from closed-loop simulation with virtual vision
SPLinTER 2
Extends the manipulation abilities of Splinter 1.0 by learning models over a set action primitives, fed to a planner that can use them.
Lower-level robot stuff
Bug-Tracking!
My first project on the arm: visual servoing to track and capture a mobile target
Force-compliance
Force-compliance in workspace coordinates
Fun!!
Clodbuster RC car
A side project to hack an RC car that I did for fun
Helicopter Control
Another side project I did for fun to build my own UAV
2D mapping
Ever wonder how to image a plane for less than 100$ of embedded parts?
Box2D Friction Mod
Once for my research on object manipulation I had to reinvent the wheel (literally)
Motorcycle Fairing Repair
Ever wonder how to fix a motorcycle fairing for the same price you'd have to pay a pro?
About me