Refresh the dof properties of the actor in the given environments, i.e. dof friction, damping, armature

Creates environments: 1. loads the robot URDF/MJCF asset, 2. For each environment 2.1 creates the environme

Adds a ground plane to the simulation, sets friction and restitution based on the cfg.

Adds a heightfield terrain to the simulation, sets parameters based on the cfg.

Adds a triangle mesh terrain to the simulation, sets parameters based on the cfg. #

Samples heights of the terrain at required points around each robot. The points are offset by the base's position and rotated by the base

Randommly select commands of some environments Args: env_ids (List[int]): Environments ids for which new commands are needed

Randommly select commands of some environments Args: env_ids (List[int]): Environments ids for which new commands are needed

Resets DOF position and velocities of selected environmments Positions are randomly selected within 0.5:1.5 x default positions. Velo

Resets ROOT states position and velocities of selected environmments Sets base position based on the curriculum Selects rando

Implements the game-inspired curriculum. Args: env_ids (List[int]): ids of environments being reset

Creates the training algorithm either from a registered namme or from the provided config file. Args: env (isaacgym.VecTaskPyth

Creates an environment either from a registered namme or from the provided config file. Args: name (string): Name of a registere

random add lag

Reset some environments. Calls self._reset_dofs(env_ids), self._reset_root_states(env_ids), and self._resample_commands(env_ids)

Set camera position and direction

Implements a curriculum of increasing commands Args: env_ids (List[int]): ids of environments being reset

Apply force and torque on fix body

Compute torques from actions. Actions can be interpreted as position or velocity targets given to a PD controller, or directly as scaled

Draws visualizations for dubugging (slows down simulation a lot). Default behaviour: draws height measurement points

Sets environment origins. On rough terrain the origins are defined by the terrain platforms. Otherwise create a grid.

Sets a vector used to scale the noise added to the observations. [NOTE]: Must be adapted when changing the observations structure

Initialize torch tensors which will contain simulation states and processed quantities

Returns points at which the height measurments are sampled (in base frame) Returns: [torch.Tensor]: Tensor of shape (num_envs, s

Parse simulation, reward, command, terrain, random configuration

Callback called before computing terminations, rewards, and observations Default behaviour: Compute ang vel command based on target and h

Prepares a list of reward functions, which will be called to compute the total reward. Looks for self._reward_<REWARD_NAME>, where <REWAR

Callback allowing to store/change/randomize the DOF properties of each environment. Called During environment creation. Base

Callback allowing to store/change/randomize the rigid shape properties of each environment. Called During environment creation.

Random pushes the robots. Emulates an impulse by setting a randomized base velocity.

Random pushes the robots. Emulates an impulse by setting a randomized base velocity.

Check if environments need to be reset

Compute rewards Calls each reward function which had a non-zero scale (processed in self._prepare_reward_function()) adds eac

Creates simulation, terrain and evironments

Extracts an observation from the mujoco data structure

Initialize torch tensors for random properties

Calculates torques from position commands

check terminations, compute observations and rewards calls self._post_physics_step_callback() for common computations calls

Randomise some of the rigid body properties of the actor in the given environments, i.e. sample the mass, centre of mass position, fricti