Options for inverse kinematics solver. More...

Public Attributes
Scalar	tolerance = 1e-6
	Relative tolerance for cost function value.

Scalar	xtol_rel = 1e-12
	Relative tolerance for NLopt optimization variables.

int	max_iterations = 5e3
	Maximum number of iterations for optimization.

InverseKinematicsMethod	method = InverseKinematicsMethod::JACOBIAN
	Inverse kinematics solver method.

Scalar	step_size = 1e-1
	Step size.

nlopt::algorithm	algorithm = nlopt::LD_SLSQP
	NLopt optimization algorithm to use.

Eigen::Matrix< Scalar, 6, 6 >	K = Eigen::Matrix<Scalar, 6, 6>::Identity()
	Weighting matrix for pose error.

Eigen::Matrix< Scalar, nq, nq >	W = 1e-3 * Eigen::Matrix<Scalar, nq, nq>::Identity()
	Weighting matrix for joint displacement from initial configuration.

Scalar	initial_damping = 1e-2
	Inital damping factor for Levenberg-Marquardt.

Scalar	damping_decrease_factor = 1e-1
	Damping decrease factor for Levenberg-Marquardt.

Scalar	damping_increase_factor = 2
	Damping increase factor for Levenberg-Marquardt.

int	num_particles = 5e2
	Number of particles in the swarm.

Scalar	omega = 1e-1
	Inertia weight (omega) for updating particle velocities.

Scalar	c1 = 1e-1
	Cognitive weight (c1) for updating particle velocities.

Scalar	c2 = 1e-1
	Social weight (c2) for updating particle velocities.

Scalar	init_position_scale = 1e-1
	Scale for initializing particle positions around the initial guess.

Detailed Description

template<typename Scalar, int nq>
struct tinyrobotics::InverseKinematicsOptions< Scalar, nq >

Options for inverse kinematics solver.

The documentation for this struct was generated from the following file:

Public Attributes