Updated documentation of observation format in Simple World Comm (#1212)

pettingzoo/mpe/simple_world_comm/simple_world_comm.py

@@ -30,11 +30,11 @@
 In particular, the good agents reward, is -5 for every collision with an adversary, -2 x bound by the `bound` function described in simple_tag, +2 for every collision with a food, and -0.05 x minimum distance to any food. The adversarial agents are rewarded +5 for collisions and -0.1 x minimum
 distance to a good agent. s
 
-Good agent observations: `[self_vel, self_pos, landmark_rel_positions, other_agent_rel_positions, other_agent_velocities, self_in_forest]`
+Good agent observations: `[self_vel, self_pos, landmark_rel_positions, other_agent_rel_positions, self_in_forest, other_agent_velocities]`
 
 Normal adversary observations:`[self_vel, self_pos, landmark_rel_positions, other_agent_rel_positions, other_agent_velocities, self_in_forest, leader_comm]`
 
-Adversary leader observations: `[self_vel, self_pos, landmark_rel_positions, other_agent_rel_positions, other_agent_velocities, leader_comm]`
+Adversary leader observations: `[self_vel, self_pos, landmark_rel_positions, other_agent_rel_positions, other_agent_velocities, self_in_forest, leader_comm]`
 
 *Note that when the forests prevent an agent from being seen, the observation of that agents relative position is set to (0,0).*