RRT* for Manipulation on the PR2

Alejandro Perez, Sertac Karaman, Matthew Walter, Alexander Shkolnik, Emilio Frazzoli and Seth Teller

Related Publication:

Perez, A., Karaman, S., Shkolnik, A., Frazzoli E., Teller, S. and Walter, M.,
“Asymptotically-optimal Manipulation Planning using Incremental Sampling-based Algorithms,”
in IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2011.[Pdf][Bibtex]

IROS Interactive Session Slides

Media Coverage

MIT News, Slashdot, New Scientist, New England Post, Gizmag, io9, AzoRobotics, Indian Express

Preliminary results slide (March 2011)

Time Lapse Images

Dual Arm Planning (12 DOF)

Execution of solutions for a grep-graps pose after 2000 iterations.

RRT

BT+RRT*

Videos

Single Arm Planning (5 DOF)

Execution of solutions for a goal pose (gripper under the table) after 4000 iterations.

RRT

RRT*

Single Arm Planning (7 DOF)

Execution of solutions for a pre-grasp pose after 1000 iterations.

RRT

BT+RRT*

Dual Arm Planning (12 DOF)

Execution of solutions for a grep-graps pose after 2000 iterations.

RRT

BT+RRT*

Execution of solutions for a pre-grasp pose after 4000 iterations.

RRT

RRT*

Execution of solutions for a goal pose (both grippers under the table) after 4000 iterations.

RRT-Connect + Smoothing

RRT* + Smoothing