updated action behavior at curve. deviation left and right not allowed at curve

examples/training_navigation.py

@@ -112,7 +112,7 @@ for trials in range(1, n_trials + 1):
             action = agent.act(np.array(obs[a]), eps=eps)
             action_prob[action] += 1
             action_dict.update({a: action})
-            env.obs_builder.util_print_obs_subtree(tree=obs[a], num_features_per_node=5)
+            #env.obs_builder.util_print_obs_subtree(tree=obs[a], num_features_per_node=5)
         # Environment step
         next_obs, all_rewards, done, _ = env.step(action_dict)
         for a in range(env.number_of_agents):

flatland/core/env_observation_builder.py

@@ -340,7 +340,7 @@ class TreeObsForRailEnv(ObservationBuilder):
 
             elif num_transitions == 0:
                 # Wrong cell type, but let's cover it and treat it as a dead-end, just in case
-                print("WRONG CELL TYPE detected in tree-search (0 transitions possible)")
+                print("WRONG CELL TYPE detected in tree-search (0 transitions possible) at cell",position[0], position[1] )
                 last_isTerminal = True
                 break
 

flatland/envs/rail_env.py

@@ -998,7 +998,7 @@ class RailEnv(Environment):
 
         for i in range(len(self.agents_handles)):
             handle = self.agents_handles[i]
-
+            transition_isValid = None
             if handle not in action_dict:
                 continue
 
@@ -1027,9 +1027,13 @@ class RailEnv(Environment):
                 movement = direction
                 if action == 1:
                     movement = direction - 1
+                    if nbits <= 2:
+                        transition_isValid == False
+
                 elif action == 3:
                     movement = direction + 1
-
+                    if nbits <= 2:
+                        transition_isValid == False
                 if movement < 0:
                     movement += 4
                 if movement >= 4:
@@ -1089,6 +1093,7 @@ class RailEnv(Environment):
                 else:
                     new_cell_isValid = False
 
+                if transition_isValid == None:
                     transition_isValid = self.rail.get_transition(
                         (pos[0], pos[1], direction),
                         movement) or is_deadend