haibot-rosgymbullet 0.5.7

DiffBot Ros-Gym-Bullet Environment

(*) DiffBot RosGymBullet instruction:

• reset(): state constructor, returns the corresponding observation for each initialized state;
• step(): action selection function, considered as input vsf used for the environment, instructing the kernel to act in the environment to go to the new state. This function returns 4 values: +) observation: corresponds to the current state; +) reward: the reward received from interacting with the environment when performing an action; +) done: when the state limit is reached, if True, the simulation ends, False resets the simulation; +) info: a dictionary that returns additional information about the returned environment.
• render(): display simulation window of Gym environment;
• close(): close the simulation window.