Minor Documentation and Comment Improvements (#2147)

* Update projects.rst

* Update test_envs.py

---------

Co-authored-by: Antonin RAFFIN <[email protected]>

Author: leopardracer

docs/misc/projects.rst

@@ -270,7 +270,7 @@ A Deep Reinforcement Learning agent for a humanoid robot that learns to recover
 
 The agent is trained using the MuJoCo physics engine. Real world experiments are conducted on the 
 Sigmaban humanoid robot, a small-sized humanoid designed by the *Rhoban Team* to compete in the RoboCup Kidsize League.
-The results, detailled in the paper and the video, show that the agent is able to recover from 
+The results, detailed in the paper and the video, show that the agent is able to recover from 
 various external disturbances and stand up in a few seconds.
 
 | Authors: Marc Duclusaud, Clément Gaspard, Grégoire Passault, Mélodie Daniel, Olivier Ly

tests/test_envs.py

@@ -31,7 +31,7 @@
 @pytest.mark.parametrize("env_id", ["CartPole-v1", "Pendulum-v1"])
 def test_env(env_id):
     """
-    Check that environmnent integrated in Gym pass the test.
+    Check that environment integrated in Gym pass the test.
 
     :param env_id: (str)
     """