coop_assembly: Geometry generation of robotically assembled spatial structures.
News:
9-21-2019: This package contains materials for the FABRICATION-INFORMED DESIGN OF ROBOTICALLY ASSEMBLED STRUCTURES for the Design Modeling Symposium Berlin 2019.
See Getting Started for installation instructions.
First, make sure that you first follow the Installating sequence and motion planning stack section
of the Getting Started document.
Then, Run the regression search algorithm for find a construction plan for spatial bars.
python -m coop_assembly.planning.run -v -w
Issue python -m coop_assembly.planning.run -h to print out the helps.
And you should be able to see the following (press Enter when asked in the commandline) animation (click image for the youtube video):
First, follow the Installating PDDLStream (WIP) section
of the Getting Started document to set up PDDLStream and pyplanners.
This is a 2D additive construction domain, developed for testing our algorithms.
python -m examples.assembly2D.run -v -w -p 2D_tower_skeleton.json -a incremental_sa
The incremental_sa flag is the incremental algorithm in PDDLStream, using
the pyplanners for supporting
semantic attachments.
Issue python -m examples.assembly2D.run -h to print out the helps.
If run correctly, you should see an animation (click image for the youtube video) like the following (subject to changes) after a plan is found:
WIP.
Sometimes things don't go as expected. Checkout the troubleshooting documentation for answers to the most common issues you might bump into.
This package was initiated by Stefana Parascho <parascho@princeton.edu> @stefanaparascho at the CREATE lab, Princeton University. The sequence and motion planning scripts are developed by Yijiang Huang <yijiangh@mit.edu> @yijiangh with collaborators.