poetry add git+ssh://git@github.com:rocsys/research-ktree.git#0.3.1 # 0.3.1 as example (modify this for different versions)The library allows the creation of a static (and soon dynamic) kinematic tree based on pose transformations (Set in a configuration yaml)
base_frame: yaskawa_base
end_effector_frame: conn
kinematics_chain:
- parent: yaskawa_base
child: yaskawa_eff
joint:
axis: x
type: prismatic
pose:
x_mm: 1.0
y_m: 0.002
z_mm: 3.0
rx_deg: 180.0
ry_rad: 0.0
rz_deg: 0.0
- parent: yaskawa_eff
child: checkered_board
pose:
x_mm: 4.0
y_mm: 5.0
z_mm: 6.0
rx_deg: 0.0
ry_deg: 0.0
rz_deg: 0.0
- parent: cam
child: checkered_board
pose:
x_mm: 7.0
y_mm: 8.0
z_mm: 9.0
rx_deg: 0.0
ry_deg: 0.0
rz_deg: 0.0
- parent: cam
child: conn
pose:
x_mm: 10.0
y_mm: 11.0
z_mm: 12.0
rx_deg: 0.0
ry_deg: 0.0
rz_deg: 0.0- Pose transformations must contain clear references to parent and child frames and optionally if the transformation involves an active joint (set via the joint axis-type object).
To use the library just import it and load the config file into the object
from ktree import KinematicsConfig, KinematicsTree, Pose
kc = KinematicsConfig.parse(Path("./test/config.yaml"))
kt = KinematicsTree(config=kc)
end_eff_in_base = kt.get_transformation(parent=kc.base, child=kc.end_effector)
print(end_eff_in_base)
kt.update_transformation(parent="cam", child="conn", pose=Pose)
print(end_eff_in_base)