Differentiable robot model class

TODO

class differentiable_robot_model.robot_model.DifferentiableFrankaPanda(device=None)

Bases: differentiable_robot_model.robot_model.DifferentiableRobotModel

training: bool

class differentiable_robot_model.robot_model.DifferentiableKUKAiiwa(device=None)

Bases: differentiable_robot_model.robot_model.DifferentiableRobotModel

training: bool

class differentiable_robot_model.robot_model.DifferentiableRobotModel(urdf_path: str, name='', device=None)

Bases: torch.nn.modules.module.Module

Differentiable Robot Model

compute_endeffector_jacobian(*args, **kwargs)

compute_forward_dynamics(*args, **kwargs)

compute_forward_dynamics_old(*args, **kwargs)

compute_forward_kinematics(*args, **kwargs)

compute_inverse_dynamics(*args, **kwargs)

compute_lagrangian_inertia_matrix(*args, **kwargs)

compute_non_linear_effects(*args, **kwargs)

freeze_learnable_link_param(link_name: str, parameter_name: str)

get_joint_limits() → List[Dict[str, torch.Tensor]]

Returns: list of joint limit dict, containing joint position, velocity and effort limits

get_link_names() → List[str]

Returns: a list containing names for all links

iterative_newton_euler(*args, **kwargs)

make_link_param_learnable(link_name: str, parameter_name: str, parametrization: torch.nn.modules.module.Module)

print_learnable_params() → None

print the name and value of all learnable parameters

print_link_names() → None

print the names of all links

training: bool

unfreeze_learnable_link_param(link_name: str, parameter_name: str)

update_kinematic_state(*args, **kwargs)

class differentiable_robot_model.robot_model.DifferentiableTwoLinkRobot(device=None)

Bases: differentiable_robot_model.robot_model.DifferentiableRobotModel

training: bool

differentiable_robot_model.robot_model.tensor_check(function)

A decorator for checking the device of input tensors