openai_ros.task_envs.turtlebot3 package

Submodules

openai_ros.task_envs.turtlebot3.turtlebot3_world module

class openai_ros.task_envs.turtlebot3.turtlebot3_world.TurtleBot3WorldEnv

Bases: openai_ros.robot_envs.turtlebot3_env.TurtleBot3Env

__init__()

This Task Env is designed for having the TurtleBot3 in the turtlebot3 world closed room with columns. It will learn how to move around without crashing.

__module__ = 'openai_ros.task_envs.turtlebot3.turtlebot3_world'

_compute_reward(observations, done)

_get_obs()

Here we define what sensor data defines our robots observations To know which Variables we have acces to, we need to read the TurtleBot2Env API DOCS :return:

_init_env_variables()

Inits variables needed to be initialised each time we reset at the start of an episode. :return:

_is_done(observations)

_set_action(action)

This set action will Set the linear and angular speed of the turtlebot2 based on the action number given. :param action: The action integer that set s what movement to do next.

_set_init_pose()

Sets the Robot in its init pose

discretize_scan_observation(data, new_ranges)

Discards all the laser readings that are not multiple in index of new_ranges value.

get_vector_magnitude(vector)

It calculated the magnitude of the Vector3 given. This is usefull for reading imu accelerations and knowing if there has been a crash :return:

reward_range = None

We set the Observation space for the 6 observations cube_observations = [

round(current_disk_roll_vel, 0), round(y_distance, 1), round(roll, 1), round(pitch, 1), round(y_linear_speed,1), round(yaw, 1),

]

Module contents