Home

Awesome

SS-Replan

Online observation, estimatation, planning, and control for a Franka Panda Robot operating in NVIDIA SRL's simulated kitchen environment.

Installation

SS-Replan supports both Python 2 and Python 3.

<!--* `sudo apt install cmake g++ make python ros-kinetic-trac-ik`-->

It's also possible to use TRAC-IK instead of IKFast; however it requires installing ROS ($ sudo apt install ros-kinetic-trac-ik).

<!--https://bitbucket.org/traclabs/trac_ik/src/master/-->

PyBullet Examples

<!--* `SS-Replan$ git pull --recurse-submodules`-->

<img src="https://img.youtube.com/vi/TvZqMDBZEnc/0.jpg" height="250">

<!--&emsp;-->

IsaacSim Examples

Executed using the IssacSim 3D robot simulation environment.

<img src="https://img.youtube.com/vi/XSZbCp0M1rw/0.jpg" height="250">

Real-World Examples

<img src="https://img.youtube.com/vi/-Jl6GtvtWb8/0.jpg" height="250">

<!-- https://developer.nvidia.com/isaac-sdk -->

Citation

Caelan R. Garrett, Chris Paxton, Tomás Lozano-Pérez, Leslie P. Kaelbling, Dieter Fox. Online Replanning in Belief Space for Partially Observable Task and Motion Problems, IEEE International Conference on Robotics and Automation (ICRA), 2020.

Publications

Videos

Resources

PDDLStream

<!-- * [SS-Replan Paper](https://arxiv.org/abs/1911.04577) -->

PDDL and FastDownward

PyBullet

<!--# Resources Please email Caelan Garrett at <caelan@mit.edu> for installation and usage help.-->