pychoreo 0.3.0

sequence and motion planning for robotic spatial extrusion

pychoreo is a sequence and motion planning engine that allow you to print the following cool structures (and many more!) with ease:

Note :pushpin:

• In the summer of 2019, pychoreo will be integrated into the compas_fab infrastructure. Stay tuned! :beers:
• The ROS implementation of choreo can be found here: choreo.

Main features

• feature

Documentation

Coming soon!

Requirements

Installation

Credits

If you use this work, please consider citing as follows:

@article{huang2018automated,

title={Automated sequence and motion planning for robotic spatial extrusion of 3D trusses}, author={Huang, Yijiang and Garrett, Caelan R and Mueller, Caitlin T}, journal={Construction Robotics}, volume={2}, number={1-4}, pages={15–39}, year={2018}, publisher={Springer}}

Algorithms behind Choreo:

• Automated sequence and motion planning for robotic spatial extrusion of 3D trusses, Constr Robot (2018) 2:15-39, Arxiv-1810.00998

Applications of Choreo:

• Robotic extrusion of architectural structures with nonstandard topology, RobArch 2018, RobArch paper link
• Spatial extrusion of Topology Optimized 3D Trusses, IASS 2018, IASS paper link

Related repos

Task and Motion Planning

• https://github.com/caelan/pb-construction
• https://github.com/caelan/pddlstream

Computational design and digital fabrication

• https://github.com/compas-dev/compas_fab

Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog and this project adheres to Semantic Versioning.

TODO

• storing ee_poses in CapVert is not necessary, should think of a way to get around this.
• storing joint data in a continuous array may not be necessary, since we are using nested list to describe subprocesses anyway
• need to regulate the use of ik_joints or ik_joint_names for user interfaces

0.3.0

Added

0.2.0

Added

• SparseLadderGraph completed
• export planned trajectory for extrusion
• add parsing function for visualizing saved extrusion trajectories
• from_data methods for Trajectory and subclasses
• tagging print processes with ground/creation/connect in the test function
• infinite pose sampler added for extrusion case when using sparse ladder graph to solve
• Added max_valence_extrusion_direction_routing to extrusion.utils
• Added reverse_flags info to add_collision_fns_from_seq and extrusion’s test
• Added start_conf parameter to SparseLadderGraph.extract_solution and solve_ladder_graph_from_cartesian_process_list to allow minimizing ladder graph with respect to a given start configuration
• Added picknplace.transition_planner
• Added target_conf attribute to CartesianProcess to allow using snap_sols when sample_ik_sols is called. This is essential for robots with large joint limits, e.g. UR.

Minor

• is_any_empty utility function for checking ik sol list of lists
• reset_ee_pose_gen_fn for easier resetting generator
• Added print_table model in the mit_3-412_workspace URDF/SRDF

Removed

• Removed PicknPlaceBufferTrajectory’s ee_attachments and attachments attributes
• Removed picknplace.planner_interface (which is there only as an archive)

Fixed

• fix nested empty list detection bug in is_any_empty
• add disabled_collisions argument to the extrusion transition_planner
• Fixed min_z to base_point model transformation in extrusion.parsing

Changed

• extrusion export save lin_path’s poses as 4x4 tform matrix (there’s some disagreement in quaterion in compas.Frame.from_quat?)
• move extrusion test fixtures into a separate fixture module
• ladder graph interface broken into from_cartesian_process_list, from_cartesian_process, from_poses to increase code reuse
• Changed sub_process_ids specification in prune_ee_feasible_directions
• Changed Trajectory to have ee_attachments and attachments attributes natively
• Changed Trajectory’s from_data, making it raise ValueError when robot body cannot be found in pybullet
• Changed MoveTrajectory to have element_id attributes natively
• Changed picknplace.visualization to reload and manually assign pybullet bodies to ensure objects get matched correctly
• Changed build_picknplace_cartesian_process_seq to inject ee_attach info before passing into ladder graph solver, and tag element attachment after solving is finished.

0.1.1

Added

• cartesian process class for modeling general linear movement in the workspace
• ladder graph interface using the Cartesian process class
• Trajectory class for modeling result trajectory in different contexts (inherited classes)
• display_trajectories for extrusion
• some simple exceptions added for LadderGraph and DAGSearch
• subprocess modeling to have a more detailed control over Cartesian process modeling
• add exhaust_iter method to CartisianProcess which resets the generator
• add template class GenFn for generating functions
• add PrintBufferTrajectory to model approach/retreat trajectories

Changed

• move transition planning to application context.
• conform to the latest pybullet_planning

Removed

• assembly_datastructure
• the old extrusion.run module, moved to the test file

Fixed

Deprecated

0.0.1

Added

• Initial version