pyroboplan.ik package

Inverse kinematics module.

This module contains solvers and utilities for inverse kinematics (IK).

Submodules

pyroboplan.ik.differential_ik module

Utilities for differential IK.

class pyroboplan.ik.differential_ik.DifferentialIk(model, collision_model=None, data=None, collision_data=None, visualizer=None, options=<pyroboplan.ik.differential_ik.DifferentialIkOptions object>)

Bases: object

Differential IK solver.

This is a numerical IK solver that uses the manipulator’s Jacobian to take first-order steps towards a solution. It contains several of the common options such as damped least squares (Levenberg-Marquardt), random restarts, and nullspace projection.

Some good resources:

• https://motion.cs.illinois.edu/RoboticSystems/InverseKinematics.html

• https://homes.cs.washington.edu/~todorov/courses/cseP590/06_JacobianMethods.pdf

• https://www.cs.cmu.edu/~15464-s13/lectures/lecture6/iksurvey.pdf

• http://www.diag.uniroma1.it/deluca/rob2_en/02_KinematicRedundancy_1.pdf

solve(target_frame, target_tform, init_state=None, nullspace_components=[], verbose=False)

Solves an IK query.

Parameters:

• target_frame (str) – The name of the target frame in the model.

• target_tform (pinocchio.SE3) – The desired transformation of the target frame in the model.

• init_state (array-like, optional) – The initial state to solve from. If not specified, a random initial state will be selected.

• nullspace_components (list[function], optional) – An optional list of nullspace components to use when solving. These components must take the form lambda model, q: component(model, q, <other_args>).

• verbose (bool, optional) – If True, prints additional information to the console.

Returns:

A list of joint configuration values with the solution, if one was found. Otherwise, returns None.

Return type:

array-like or None

class pyroboplan.ik.differential_ik.DifferentialIkOptions(max_iters=200, max_retries=10, max_translation_error=0.001, max_rotation_error=0.001, damping=0.001, min_step_size=0.1, max_step_size=0.5, ignore_joint_indices=[], joint_weights=None, rng_seed=None)

Bases: object

Options for differential IK.

pyroboplan.ik.nullspace_components module

Library of common nullspace components for inverse kinematics.

pyroboplan.ik.nullspace_components.collision_avoidance_nullspace_component(model, data, collision_model, collision_data, q, dist_padding=0.05, gain=1.0)

Returns a collision avoidance nullspace component.

These calculations are based off the following resources:

• https://typeset.io/pdf/a-collision-free-mpc-for-whole-body-dynamic-locomotion-and-1l6itpfk.pdf

• https://laas.hal.science/hal-04425002

Parameters:

• model (pinocchio.Model) – The model from which to generate a random state.

• data (pinocchio.Data) – The model data to use for collision distance checks.

• collision_model (pinocchio.GeometryModel) – The model with which to check collision distances.

• collision_data (pinocchio.GeometryData) – The collision model data to use for collision distance checks.

• q (array-like) – The joint configuration for the model.

• dist_padding (float) – The distance padding, in meters, on the collision distances. For example, a distance padding of 0.1 means collisions have an influence 10 cm away from actual collision..

• gain (float, optional) – A gain to modify the relative weight of this term.

Returns:

An array containing the collision avoidance nullspace terms.

Return type:

array-like

pyroboplan.ik.nullspace_components.joint_center_nullspace_component(model, q, gain=1.0)

Returns a joint centering nullspace component.

Parameters:

• model (pinocchio.Model) – The model from which to generate a random state.

• q (array-like) – The joint configuration for the model.

• gain (float, optional) – A gain to modify the relative weight of this term.

Returns:

An array containing the joint centering nullspace terms.

Return type:

array-like

pyroboplan.ik.nullspace_components.joint_limit_nullspace_component(model, q, gain=1.0, padding=0.0)

Returns a joint limits avoidance nullspace component.

Parameters:

• model (pinocchio.Model) – The model from which to generate a random state.

• q (array-like) – The joint configuration for the model.

• gain (float, optional) – A gain to modify the relative weight of this term.

• padding (float, optional) – Optional padding around the joint limits.

Returns:

An array containing the joint space avoidance nullspace terms.

Return type:

array-like

pyroboplan.ik.nullspace_components.zero_nullspace_component(model, q)

Returns a zero nullspace component, which is effectively a no-op.

Parameters:

• model (pinocchio.Model) – The model from which to generate a random state.

• q (array-like) – The joint configuration for the model. Not used, but required to match the function interface.

Returns:

An array of zeros whose length is the number of joint variables in the model.

Return type:

array-like