openai_ros.task_envs.parrotdrone package


openai_ros.task_envs.parrotdrone.parrotdrone_goto module

class openai_ros.task_envs.parrotdrone.parrotdrone_goto.ParrotDroneGotoEnv[source]

Bases: openai_ros.robot_envs.parrotdrone_env.ParrotDroneEnv


Make parrotdrone learn how to navigate to get to a point

__module__ = 'openai_ros.task_envs.parrotdrone.parrotdrone_goto'
_compute_reward(observations, done)[source]

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


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


The done can be done due to three reasons: 1) It went outside the workspace 2) It detected something with the sonar that is too close 3) It flipped due to a crash or something 4) It has reached the desired point


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


Sets the Robot in its init linear and angular speeds and lands the robot. Its preparing it to be reseted in the world.


Based on the orientation RPY given states if the drone has flipped


Calculates the distance from the current position to the desired point :param start_point: :return:

get_distance_from_point(pstart, p_end)[source]

Given a Vector3 Object, get distance from current position :param p_end: :return:

is_in_desired_position(current_position, epsilon=0.05)[source]

It return True if the current position is similar to the desired poistion


Check if the Drone is inside the Workspace defined


Detects if there is something too close to the drone front

Module contents