Observation Space Reduction for POMDPs
Using K-means Analysis and Principal Component Analysis, we attempt
to perform feature reduction directly on the observation matrix for
POMDPs. The reduced observation matrices allow for faster planning
of larger POMDPs, but with minimal loss to the expected reward. |
Active Learning for POMDPs
POMDP research has typically been focused on planning using existing
POMDP models. The goal of this project is to use active learning to
discover the parameters of the model itself. By requesting
information from an oracle, a POMDP model is refined over time to
better approximate the real world. |
Smartwheeler Autonomous Wheelchair
The goal Smartwheeler project is
to design an construct an autonomous wheelchair. Starting with a standard electric
wheelchair, sensors and an onboard computer were added. This will act as a platform
for testing other robotics related research.
|
DARPA MARS
My focus in the project is on probabilistic planning and execution in
behavior based systems. By applying POMDPs (Partially Observable
Markov Decision Processes), plans can be created which help
behavior-based robotic system deal with uncertainty. This uncertainty
may come from noise in the environment or sensor inaccuracy. |
Autonomous Pinball Using Neural Networks, we are trying
to learn low-level control for a physical pinball machine. The
pinball machine is interfaced to the computer through a circuit which
controls the flippers. A camera tracks the ball in play. This
forces us to deal with issues such as real time processing, noise
from the environment, and noise from the sensors, as well as
learning a complex space. |
Continuous Localization
As a summer internship at the Naval Reserach Lab on applying
the continuous localization technique on a robot to be used in
outdoor environments. Continuous localization is a map-based
localization technique initially developed on Nomads. Current work
is in porting it to an ATRV Jr, a robot designed for outdoor
enviroments. This involved integrating various sensors to help with
localization, experimentation in small, high resolution indoor
environments, and experimentation in large, low resolution outdoor
environments. |
|
There's more, I just haven't finished updating this new version of the page. Consult
my old page for a more thorough description of my previous research.
|