rofunc.learning.RofuncRL.tasks.isaacgymenv.franka_cube_stack#

1.  Module Contents#

1.1.  Classes#

FrankaCubeStackTask

1.2.  Functions#

axisangle2quat

compute_franka_reward

1.3.  API#

rofunc.learning.RofuncRL.tasks.isaacgymenv.franka_cube_stack.axisangle2quat(vec: Tensor, eps: float = 1e-06) → Tensor#
class rofunc.learning.RofuncRL.tasks.isaacgymenv.franka_cube_stack.FrankaCubeStackTask(cfg, rl_device, sim_device, graphics_device_id, headless, virtual_screen_capture, force_render)[source]#

Bases: rofunc.learning.RofuncRL.tasks.isaacgymenv.base.vec_task.VecTask

create_sim()[source]#
init_data()[source]#
compute_reward(actions)[source]#
compute_observations()[source]#
reset_idx(env_ids)[source]#
pre_physics_step(actions)[source]#
post_physics_step()[source]#
rofunc.learning.RofuncRL.tasks.isaacgymenv.franka_cube_stack.compute_franka_reward(reset_buf: Tensor, progress_buf: Tensor, actions: Tensor, states: Dict[str, Tensor], reward_settings: Dict[str, float], max_episode_length: float) → Tuple[Tensor, Tensor]#