diff --git a/pyreason_gym/pyreason_grid_world/graph/game_graph.graphml b/pyreason_gym/pyreason_grid_world/graph/game_graph.graphml
index dcfe857..6798475 100644
--- a/pyreason_gym/pyreason_grid_world/graph/game_graph.graphml
+++ b/pyreason_gym/pyreason_grid_world/graph/game_graph.graphml
@@ -7,12 +7,16 @@
   <key id="down" for="edge" attr.name="down" attr.type="long" />
   <key id="up" for="edge" attr.name="up" attr.type="long" />
   <key id="right" for="edge" attr.name="right" attr.type="long" />
+  <key id="blue-soldier-2" for="node" attr.name="blue-soldier-2" attr.type="long" />
+  <key id="red-soldier-2" for="node" attr.name="red-soldier-2" attr.type="long" />
   <key id="bullet" for="node" attr.name="bullet" attr.type="long" />
+  <key id="blue-soldier-1" for="node" attr.name="blue-soldier-1" attr.type="long" />
   <key id="teamBlue" for="node" attr.name="teamBlue" attr.type="long" />
   <key id="shootRightBlue" for="node" attr.name="shootRightBlue" attr.type="long" />
   <key id="shootLeftBlue" for="node" attr.name="shootLeftBlue" attr.type="long" />
   <key id="shootDownBlue" for="node" attr.name="shootDownBlue" attr.type="long" />
   <key id="shootUpBlue" for="node" attr.name="shootUpBlue" attr.type="long" />
+  <key id="red-soldier-1" for="node" attr.name="red-soldier-1" attr.type="long" />
   <key id="justDied" for="node" attr.name="justDied" attr.type="string" />
   <key id="teamRed" for="node" attr.name="teamRed" attr.type="long" />
   <key id="shootRightRed" for="node" attr.name="shootRightRed" attr.type="long" />
@@ -240,6 +244,7 @@
       <data key="shootRightRed">0</data>
       <data key="teamRed">1</data>
       <data key="justDied">0,0</data>
+      <data key="red-soldier-1">1</data>
     </node>
     <node id="blue-soldier-1">
       <data key="health">1</data>
@@ -253,6 +258,7 @@
       <data key="shootRightBlue">0</data>
       <data key="teamBlue">1</data>
       <data key="justDied">0,0</data>
+      <data key="blue-soldier-1">1</data>
     </node>
     <node id="red-bullet-1">
       <data key="teamRed">1</data>
@@ -262,6 +268,42 @@
       <data key="teamBlue">1</data>
       <data key="bullet">1</data>
     </node>
+    <node id="red-soldier-2">
+      <data key="health">1</data>
+      <data key="moveUp">0</data>
+      <data key="moveDown">0</data>
+      <data key="moveLeft">0</data>
+      <data key="moveRight">0</data>
+      <data key="shootUpRed">0</data>
+      <data key="shootDownRed">0</data>
+      <data key="shootLeftRed">0</data>
+      <data key="shootRightRed">0</data>
+      <data key="teamRed">1</data>
+      <data key="justDied">0,0</data>
+      <data key="red-soldier-2">1</data>
+    </node>
+    <node id="blue-soldier-2">
+      <data key="health">1</data>
+      <data key="moveUp">0</data>
+      <data key="moveDown">0</data>
+      <data key="moveLeft">0</data>
+      <data key="moveRight">0</data>
+      <data key="shootUpBlue">0</data>
+      <data key="shootDownBlue">0</data>
+      <data key="shootLeftBlue">0</data>
+      <data key="shootRightBlue">0</data>
+      <data key="teamBlue">1</data>
+      <data key="justDied">0,0</data>
+      <data key="blue-soldier-2">1</data>
+    </node>
+    <node id="red-bullet-2">
+      <data key="teamRed">1</data>
+      <data key="bullet">1</data>
+    </node>
+    <node id="blue-bullet-2">
+      <data key="teamBlue">1</data>
+      <data key="bullet">1</data>
+    </node>
     <edge source="0" target="1">
       <data key="right">1</data>
     </edge>
@@ -357,6 +399,9 @@
     <edge source="7" target="red-soldier-1">
       <data key="atLoc">1</data>
     </edge>
+    <edge source="7" target="red-soldier-2">
+      <data key="atLoc">1</data>
+    </edge>
     <edge source="8" target="9">
       <data key="right">1</data>
     </edge>
@@ -964,6 +1009,9 @@
     <edge source="56" target="blue-soldier-1">
       <data key="atLoc">1</data>
     </edge>
+    <edge source="56" target="blue-soldier-2">
+      <data key="atLoc">1</data>
+    </edge>
     <edge source="57" target="58">
       <data key="right">1</data>
     </edge>
@@ -1070,5 +1118,23 @@
     <edge source="blue-soldier-1" target="blue-bullet-1">
       <data key="bullet">1</data>
     </edge>
+    <edge source="red-soldier-2" target="red-base">
+      <data key="team">1</data>
+    </edge>
+    <edge source="red-soldier-2" target="7">
+      <data key="atLoc">1</data>
+    </edge>
+    <edge source="red-soldier-2" target="red-bullet-2">
+      <data key="bullet">1</data>
+    </edge>
+    <edge source="blue-soldier-2" target="blue-base">
+      <data key="team">1</data>
+    </edge>
+    <edge source="blue-soldier-2" target="56">
+      <data key="atLoc">1</data>
+    </edge>
+    <edge source="blue-soldier-2" target="blue-bullet-2">
+      <data key="bullet">1</data>
+    </edge>
   </graph>
 </graphml>
diff --git a/pyreason_gym/pyreason_grid_world/pyreason_grid_world.py b/pyreason_gym/pyreason_grid_world/pyreason_grid_world.py
index 3ce1ac0..218e668 100644
--- a/pyreason_gym/pyreason_grid_world/pyreason_grid_world.py
+++ b/pyreason_gym/pyreason_grid_world/pyreason_grid_world.py
@@ -11,14 +11,14 @@ def __init__(self, grid_size, num_agents_per_team):
 
         # Keep track of the next timestep to start
         self.next_time = 0
-        
+
         # Pyreason settings
         pr.settings.verbose = False
         pr.settings.atom_trace = False
         pr.settings.canonical = True
         pr.settings.inconsistency_check = False
         pr.settings.static_graph_facts = False
-        pr.settings.store_interpretation_changes = False
+        # pr.settings.store_interpretation_changes = True
         current_path = os.path.abspath(os.path.dirname(__file__))
 
         # Load the graph
@@ -49,20 +49,21 @@ def move(self, action):
                 fact_off = pr.fact_node.Fact(f'red_action_{i+1}_off', f'red-soldier-{i+1}', pr.label.Label(red_available_actions[a]), pr.interval.closed(0,0), self.next_time+1, self.next_time+1)
                 facts.append(fact_on)
                 facts.append(fact_off)
-    
+
         for i, a in enumerate(blue_team_actions):
             if a != 8:
                 fact_on = pr.fact_node.Fact(f'blue_action_{i+1}', f'blue-soldier-{i+1}', pr.label.Label(blue_available_actions[a]), pr.interval.closed(1,1), self.next_time, self.next_time)
                 fact_off = pr.fact_node.Fact(f'blue_action_{i+1}_off', f'blue-soldier-{i+1}', pr.label.Label(blue_available_actions[a]), pr.interval.closed(0,0), self.next_time+1, self.next_time+1)
                 facts.append(fact_on)
                 facts.append(fact_off)
-        
+
         self.interpretation = pr.reason(1, again=True, node_facts=facts)
+        # pr.save_rule_trace(self.interpretation)
         self.next_time = self.interpretation.time + 1
 
     def get_obs(self):
         observation = {'red_team': [], 'blue_team': [], 'red_bullets': [], 'blue_bullets': []}
-        
+
         # Gather bullet info for red and blue bullets
         (red_bullet_positions, blue_bullet_positions), (red_bullet_directions, blue_bullet_directions), (red_killed_who, blue_killed_who) = self._get_bullet_info()
         for red_pos, red_dir in zip(red_bullet_positions, red_bullet_directions):
@@ -99,14 +100,14 @@ def get_obs(self):
             observation['blue_team'].append({'pos': np.array(blue_pos_coords, dtype=np.int32), 'health': np.array([blue_health], dtype=np.float32), 'killed': list(blue_killed_who[i-1])})
 
         return observation
-    
+
     def get_obstacle_locations(self):
         # Return the coordinates of all the mountains in the grid to be able to draw them
         relevant_edges = [edge for edge in self.interpretation.edges if edge[1]=='mountain']
         obstacle_positions = [int(edge[0]) for edge in relevant_edges]
         obstacle_positions_coords = np.array([[pos%self.grid_size, pos//self.grid_size] for pos in obstacle_positions])
         return obstacle_positions_coords
-    
+
     def get_base_locations(self):
         # Return the locations of the two bases
         relevant_edges = [edge for edge in self.interpretation.edges if 'base' in edge[0]]
@@ -127,7 +128,8 @@ def _get_bullet_info(self):
         positions = (red_bullet_positions_coords, blue_bullet_positions_coords)
 
         # Get info about who killed whom. Stored in the form a list for every agent: (red-killer: [blue-casualties]) or (blue-killer: [red-casualties])
-        kill_info_edges = [edge for edge in self.interpretation.edges if pr.label.Label('killed') in self.interpretation.interpretations_edge[edge].world]
+        kill_info_edges = [edge for edge in self.interpretation.edges if pr.label.Label('killed') in self.interpretation.interpretations_edge[edge].world
+                           and self.interpretation.interpretations_edge[edge].world[pr.label.Label('killed')] == pr.interval.closed(1, 1)]
         kill_info_edges = sorted(kill_info_edges, key=lambda x: int(x[0][-1]))
         red_killed_who_tuple = [(int(edge[0][-1]), int(edge[1][-1])) for edge in kill_info_edges if 'red' in edge[0]]
         blue_killed_who_tuple = [(int(edge[0][-1]), int(edge[1][-1])) for edge in kill_info_edges if 'blue' in edge[0]]
diff --git a/pyreason_gym/pyreason_grid_world/yamls/rules.yaml b/pyreason_gym/pyreason_grid_world/yamls/rules.yaml
index 5580c7e..adccc8e 100644
--- a/pyreason_gym/pyreason_grid_world/yamls/rules.yaml
+++ b/pyreason_gym/pyreason_grid_world/yamls/rules.yaml
@@ -45,28 +45,26 @@ bullet_hit_just_died_red_rule:
 bullet_hit_just_died_off_blue_rule:
     target: justDied
     target_criteria:
-        - [health, 0.1, 1]
+        - [health, 0, 0]
+        - [justDied, 1, 1]
         - [teamBlue, 1, 1]
     delta_t: 1
-    immediate: true
     neigh_criteria:
         - [edge, [target, loc], atLoc, [1,1]]
         - [edge, [loc, bullet], atLoc, [1,1]]
-        - [edge, [loc, bullet], life, [0.1, 1]]
         - [node, [bullet], teamRed, [1,1]]
     ann_fn: [0, 0]
 
 bullet_hit_just_died_off_red_rule:
     target: justDied
     target_criteria:
-        - [health, 0.1, 1]
+        - [health, 0, 0]
+        - [justDied, 1, 1]
         - [teamRed, 1, 1]
     delta_t: 1
-    immediate: true
     neigh_criteria:
         - [edge, [target, loc], atLoc, [1,1]]
         - [edge, [loc, bullet], atLoc, [1,1]]
-        - [edge, [loc, bullet], life, [0.1, 1]]
         - [node, [bullet], teamBlue, [1,1]]
     ann_fn: [0, 0]
 
@@ -79,7 +77,7 @@ bullet_hit_rule:
     immediate: true
     neigh_criteria:
     ann_fn: [0, 0]
-
+    
 add_kill_blue_info_rule:
     target:
     target_criteria:
@@ -108,6 +106,14 @@ add_kill_red_info_rule:
     edges: [target, player, killed]
     ann_fn: [1, 1]
 
+remove_kill_info_rule:
+    target: killed
+    target_criteria:
+        - [killed, 1, 1]
+    delta_t: 1
+    neigh_criteria:
+    ann_fn: [0, 0]
+
 remove_bullet_after_hit_blue_rule:
     target: life
     target_criteria:
@@ -122,6 +128,20 @@ remove_bullet_after_hit_blue_rule:
 
     ann_fn: [0, 0]
 
+# Sometimes a bullet is scheduled to move to a location twice before it has actually moved
+# This prevents the bullet from re-appearing
+remove_bullet_after_hit_blue_rule_2:
+    target: life
+    target_criteria:
+        - [life, 0, 0]
+    delta_t: 1
+    neigh_criteria:
+        - [node, [player], justDied, [1,1]]
+        - [node, [target], teamRed, [1,1]]
+        - [node, [player], teamBlue, [1,1]]
+
+    ann_fn: [0, 0]
+
 remove_bullet_after_hit_red_rule:
     target: life
     target_criteria:
@@ -136,11 +156,28 @@ remove_bullet_after_hit_red_rule:
 
     ann_fn: [0, 0]
 
+# Sometimes a bullet is scheduled to move to a location twice before it has actually moved
+# This prevents the bullet from re-appearing
+remove_bullet_after_hit_red_rule_2:
+    target: life
+    target_criteria:
+        - [life, 0, 0]
+    delta_t: 1
+    neigh_criteria:
+        - [node, [player], justDied, [1,1]]
+        - [node, [target], teamBlue, [1,1]]
+        - [node, [player], teamRed, [1,1]]
+
+    ann_fn: [0, 0]
+
+
+# Blue-agent-1
 # Moving rules
 # Move UP
-move_up_rule_1:
+move_up_rule_1_b1:
     target:
     target_criteria:
+        - [blue-soldier-1, 1, 1]
         - [moveUp, 1, 1]
         - [health, 0.1, 1]
     delta_t: 0
@@ -153,9 +190,10 @@ move_up_rule_1:
     edges: [target, newLoc, atLoc]
     ann_fn: [0.5,0.5]
 
-move_up_rule_2:
+move_up_rule_2_b1:
     target:
     target_criteria:
+        - [blue-soldier-1, 1, 1]
         - [moveUp, 1, 1]
         - [health, 0.1, 1]
     delta_t: 0
@@ -168,34 +206,37 @@ move_up_rule_2:
     edges: [newLoc, target, atLoc]
     ann_fn: [0.5,0.5]
 
-move_up_change_prev_rule_1:
+move_up_change_prev_rule_1_b1:
     target: atLoc
     target_criteria:
         - [atLoc, 1, 1]
     delta_t: 0
     immediate: true
     neigh_criteria:
+        - [node, [source], blue-soldier-1, [1,1]]
         - [edge, [target, newLoc], up, [1,1]]
         - [edge, [source, newLoc], atLoc, [0.5,0.5]]
 
     ann_fn: [0,0]
 
-move_up_change_prev_rule_2:
+move_up_change_prev_rule_2_b1:
     target: atLoc
     target_criteria:
         - [atLoc, 1, 1]
     delta_t: 0
     immediate: true
     neigh_criteria:
+        - [node, [target], blue-soldier-1, [1,1]]
         - [edge, [source, newLoc], up, [1,1]]
         - [edge, [newLoc, target], atLoc, [0.5,0.5]]
 
     ann_fn: [0,0]
 
 # Move DOWN
-move_down_rule_1:
+move_down_rule_1_b1:
     target:
     target_criteria:
+        - [blue-soldier-1, 1, 1]
         - [moveDown, 1, 1]
         - [health, 0.1, 1]
     delta_t: 0
@@ -208,9 +249,10 @@ move_down_rule_1:
     edges: [target, newLoc, atLoc]
     ann_fn: [0.5,0.5]
 
-move_down_rule_2:
+move_down_rule_2_b1:
     target:
     target_criteria:
+        - [blue-soldier-1, 1, 1]
         - [moveDown, 1, 1]
         - [health, 0.1, 1]
     delta_t: 0
@@ -223,34 +265,37 @@ move_down_rule_2:
     edges: [newLoc, target, atLoc]
     ann_fn: [0.5,0.5]
 
-move_down_change_prev_rule_1:
+move_down_change_prev_rule_1_b1:
     target: atLoc
     target_criteria:
         - [atLoc, 1, 1]
     delta_t: 0
     immediate: true
     neigh_criteria:
+        - [node, [source], blue-soldier-1, [1,1]]
         - [edge, [target, newLoc], down, [1,1]]
         - [edge, [source, newLoc], atLoc, [0.5,0.5]]
 
     ann_fn: [0,0]
 
-move_down_change_prev_rule_2:
+move_down_change_prev_rule_2_b1:
     target: atLoc
     target_criteria:
         - [atLoc, 1, 1]
     delta_t: 0
     immediate: true
     neigh_criteria:
+        - [node, [target], blue-soldier-1, [1,1]]
         - [edge, [source, newLoc], down, [1,1]]
         - [edge, [newLoc, target], atLoc, [0.5,0.5]]
 
     ann_fn: [0,0]
 
 # Move LEFT
-move_left_rule_1:
+move_left_rule_1_b1:
     target:
     target_criteria:
+        - [blue-soldier-1, 1, 1]
         - [moveLeft, 1, 1]
         - [health, 0.1, 1]
     delta_t: 0
@@ -263,9 +308,10 @@ move_left_rule_1:
     edges: [target, newLoc, atLoc]
     ann_fn: [0.5,0.5]
 
-move_left_rule_2:
+move_left_rule_2_b1:
     target:
     target_criteria:
+        - [blue-soldier-1, 1, 1]
         - [moveLeft, 1, 1]
         - [health, 0.1, 1]
     delta_t: 0
@@ -278,34 +324,53 @@ move_left_rule_2:
     edges: [newLoc, target, atLoc]
     ann_fn: [0.5,0.5]
 
-move_left_change_prev_rule_1:
+move_left_change_prev_rule_1_b1:
     target: atLoc
     target_criteria:
         - [atLoc, 1, 1]
     delta_t: 0
     immediate: true
     neigh_criteria:
+        - [node, [source], blue-soldier-1, [1,1]]
         - [edge, [target, newLoc], left, [1,1]]
         - [edge, [source, newLoc], atLoc, [0.5,0.5]]
 
     ann_fn: [0,0]
 
-move_left_change_prev_rule_2:
+move_left_change_prev_rule_2_b1:
     target: atLoc
     target_criteria:
         - [atLoc, 1, 1]
     delta_t: 0
     immediate: true
     neigh_criteria:
+        - [node, [target], blue-soldier-1, [1,1]]
         - [edge, [source, newLoc], left, [1,1]]
         - [edge, [newLoc, target], atLoc, [0.5,0.5]]
 
     ann_fn: [0,0]
 
 # Move RIGHT
-move_right_rule_1:
+move_right_rule_1_b1:
+    target:
+    target_criteria:
+        - [blue-soldier-1, 1, 1]
+        - [moveRight, 1, 1]
+        - [health, 0.1, 1]
+    delta_t: 0
+    immediate: true
+    neigh_criteria:
+        - [edge, [target, oldLoc], atLoc, [1,1]]
+        - [edge, [oldLoc, newLoc], right, [1,1]]
+        - [node, [newLoc], blocked, [0,0]]
+
+    edges: [target, newLoc, atLoc]
+    ann_fn: [0.5,0.5]
+
+move_right_rule_2_b1:
     target:
     target_criteria:
+        - [blue-soldier-1, 1, 1]
         - [moveRight, 1, 1]
         - [health, 0.1, 1]
     delta_t: 0
@@ -315,16 +380,716 @@ move_right_rule_1:
         - [edge, [oldLoc, newLoc], right, [1,1]]
         - [node, [newLoc], blocked, [0,0]]
 
+    edges: [newLoc, target, atLoc]
+    ann_fn: [0.5,0.5]
+
+move_right_change_prev_rule_1_b1:
+    target: atLoc
+    target_criteria:
+        - [atLoc, 1, 1]
+    delta_t: 0
+    immediate: true
+    neigh_criteria:
+        - [node, [source], blue-soldier-1, [1,1]]
+        - [edge, [target, newLoc], right, [1,1]]
+        - [edge, [source, newLoc], atLoc, [0.5,0.5]]
+
+    ann_fn: [0,0]
+
+move_right_change_prev_rule_2_b1:
+    target: atLoc
+    target_criteria:
+        - [atLoc, 1, 1]
+    delta_t: 0
+    immediate: true
+    neigh_criteria:
+        - [node, [target], blue-soldier-1, [1,1]]
+        - [edge, [source, newLoc], right, [1,1]]
+        - [edge, [newLoc, target], atLoc, [0.5,0.5]]
+
+    ann_fn: [0,0]
+
+# Blue-agent-2
+# Moving rules
+# Move UP
+move_up_rule_1_b2:
+    target:
+    target_criteria:
+        - [blue-soldier-2, 1, 1]
+        - [moveUp, 1, 1]
+        - [health, 0.1, 1]
+    delta_t: 1
+    immediate: true
+    neigh_criteria:
+        - [edge, [target, oldLoc], atLoc, [1,1]]
+        - [edge, [oldLoc, newLoc], up, [1,1]]
+        - [node, [newLoc], blocked, [0,0]]
+
+    edges: [target, newLoc, atLoc]
+    ann_fn: [0.5,0.5]
+
+move_up_rule_2_b2:
+    target:
+    target_criteria:
+        - [blue-soldier-2, 1, 1]
+        - [moveUp, 1, 1]
+        - [health, 0.1, 1]
+    delta_t: 1
+    immediate: true
+    neigh_criteria:
+        - [edge, [target, oldLoc], atLoc, [1,1]]
+        - [edge, [oldLoc, newLoc], up, [1,1]]
+        - [node, [newLoc], blocked, [0,0]]
+
+    edges: [newLoc, target, atLoc]
+    ann_fn: [0.5,0.5]
+
+move_up_change_prev_rule_1_b2:
+    target: atLoc
+    target_criteria:
+        - [atLoc, 1, 1]
+    delta_t: 0
+    immediate: true
+    neigh_criteria:
+        - [node, [source], blue-soldier-2, [1,1]]
+        - [edge, [target, newLoc], up, [1,1]]
+        - [edge, [source, newLoc], atLoc, [0.5,0.5]]
+
+    ann_fn: [0,0]
+
+move_up_change_prev_rule_2_b2:
+    target: atLoc
+    target_criteria:
+        - [atLoc, 1, 1]
+    delta_t: 0
+    immediate: true
+    neigh_criteria:
+        - [node, [target], blue-soldier-2, [1,1]]
+        - [edge, [source, newLoc], up, [1,1]]
+        - [edge, [newLoc, target], atLoc, [0.5,0.5]]
+
+    ann_fn: [0,0]
+
+# Move DOWN
+move_down_rule_1_b2:
+    target:
+    target_criteria:
+        - [blue-soldier-2, 1, 1]
+        - [moveDown, 1, 1]
+        - [health, 0.1, 1]
+    delta_t: 1
+    immediate: true
+    neigh_criteria:
+        - [edge, [target, oldLoc], atLoc, [1,1]]
+        - [edge, [oldLoc, newLoc], down, [1,1]]
+        - [node, [newLoc], blocked, [0,0]]
+
+    edges: [target, newLoc, atLoc]
+    ann_fn: [0.5,0.5]
+
+move_down_rule_2_b2:
+    target:
+    target_criteria:
+        - [blue-soldier-2, 1, 1]
+        - [moveDown, 1, 1]
+        - [health, 0.1, 1]
+    delta_t: 1
+    immediate: true
+    neigh_criteria:
+        - [edge, [target, oldLoc], atLoc, [1,1]]
+        - [edge, [oldLoc, newLoc], down, [1,1]]
+        - [node, [newLoc], blocked, [0,0]]
+
+    edges: [newLoc, target, atLoc]
+    ann_fn: [0.5,0.5]
+
+move_down_change_prev_rule_1_b2:
+    target: atLoc
+    target_criteria:
+        - [atLoc, 1, 1]
+    delta_t: 0
+    immediate: true
+    neigh_criteria:
+        - [node, [source], blue-soldier-2, [1,1]]
+        - [edge, [target, newLoc], down, [1,1]]
+        - [edge, [source, newLoc], atLoc, [0.5,0.5]]
+
+    ann_fn: [0,0]
+
+move_down_change_prev_rule_2_b2:
+    target: atLoc
+    target_criteria:
+        - [atLoc, 1, 1]
+    delta_t: 0
+    immediate: true
+    neigh_criteria:
+        - [node, [target], blue-soldier-2, [1,1]]
+        - [edge, [source, newLoc], down, [1,1]]
+        - [edge, [newLoc, target], atLoc, [0.5,0.5]]
+
+    ann_fn: [0,0]
+
+# Move LEFT
+move_left_rule_1_b2:
+    target:
+    target_criteria:
+        - [blue-soldier-2, 1, 1]
+        - [moveLeft, 1, 1]
+        - [health, 0.1, 1]
+    delta_t: 1
+    immediate: true
+    neigh_criteria:
+        - [edge, [target, oldLoc], atLoc, [1,1]]
+        - [edge, [oldLoc, newLoc], left, [1,1]]
+        - [node, [newLoc], blocked, [0,0]]
+
+    edges: [target, newLoc, atLoc]
+    ann_fn: [0.5,0.5]
+
+move_left_rule_2_b2:
+    target:
+    target_criteria:
+        - [blue-soldier-2, 1, 1]
+        - [moveLeft, 1, 1]
+        - [health, 0.1, 1]
+    delta_t: 1
+    immediate: true
+    neigh_criteria:
+        - [edge, [target, oldLoc], atLoc, [1,1]]
+        - [edge, [oldLoc, newLoc], left, [1,1]]
+        - [node, [newLoc], blocked, [0,0]]
+
+    edges: [newLoc, target, atLoc]
+    ann_fn: [0.5,0.5]
+
+move_left_change_prev_rule_1_b2:
+    target: atLoc
+    target_criteria:
+        - [atLoc, 1, 1]
+    delta_t: 0
+    immediate: true
+    neigh_criteria:
+        - [node, [source], blue-soldier-2, [1,1]]
+        - [edge, [target, newLoc], left, [1,1]]
+        - [edge, [source, newLoc], atLoc, [0.5,0.5]]
+
+    ann_fn: [0,0]
+
+move_left_change_prev_rule_2_b2:
+    target: atLoc
+    target_criteria:
+        - [atLoc, 1, 1]
+    delta_t: 0
+    immediate: true
+    neigh_criteria:
+        - [node, [target], blue-soldier-2, [1,1]]
+        - [edge, [source, newLoc], left, [1,1]]
+        - [edge, [newLoc, target], atLoc, [0.5,0.5]]
+
+    ann_fn: [0,0]
+
+# Move RIGHT
+move_right_rule_1_b2:
+    target:
+    target_criteria:
+        - [blue-soldier-2, 1, 1]
+        - [moveRight, 1, 1]
+        - [health, 0.1, 1]
+    delta_t: 1
+    immediate: true
+    neigh_criteria:
+        - [edge, [target, oldLoc], atLoc, [1,1]]
+        - [edge, [oldLoc, newLoc], right, [1,1]]
+        - [node, [newLoc], blocked, [0,0]]
+
     edges: [target, newLoc, atLoc]
     ann_fn: [0.5,0.5]
 
-move_right_rule_2:
+move_right_rule_2_b2:
     target:
     target_criteria:
+        - [blue-soldier-2, 1, 1]
         - [moveRight, 1, 1]
         - [health, 0.1, 1]
+    delta_t: 1
+    immediate: true
+    neigh_criteria:
+        - [edge, [target, oldLoc], atLoc, [1,1]]
+        - [edge, [oldLoc, newLoc], right, [1,1]]
+        - [node, [newLoc], blocked, [0,0]]
+
+    edges: [newLoc, target, atLoc]
+    ann_fn: [0.5,0.5]
+
+move_right_change_prev_rule_1_b2:
+    target: atLoc
+    target_criteria:
+        - [atLoc, 1, 1]
     delta_t: 0
     immediate: true
+    neigh_criteria:
+        - [node, [source], blue-soldier-2, [1,1]]
+        - [edge, [target, newLoc], right, [1,1]]
+        - [edge, [source, newLoc], atLoc, [0.5,0.5]]
+
+    ann_fn: [0,0]
+
+move_right_change_prev_rule_2_b2:
+    target: atLoc
+    target_criteria:
+        - [atLoc, 1, 1]
+    delta_t: 0
+    immediate: true
+    neigh_criteria:
+        - [node, [target], blue-soldier-2, [1,1]]
+        - [edge, [source, newLoc], right, [1,1]]
+        - [edge, [newLoc, target], atLoc, [0.5,0.5]]
+
+    ann_fn: [0,0]
+
+# red-agent-1
+# Moving rules
+# Move UP
+move_up_rule_1_r1:
+    target:
+    target_criteria:
+        - [red-soldier-1, 1, 1]
+        - [moveUp, 1, 1]
+        - [health, 0.1, 1]
+    delta_t: 0
+    immediate: true
+    neigh_criteria:
+        - [edge, [target, oldLoc], atLoc, [1,1]]
+        - [edge, [oldLoc, newLoc], up, [1,1]]
+        - [node, [newLoc], blocked, [0,0]]
+
+    edges: [target, newLoc, atLoc]
+    ann_fn: [0.5,0.5]
+
+move_up_rule_2_r1:
+    target:
+    target_criteria:
+        - [red-soldier-1, 1, 1]
+        - [moveUp, 1, 1]
+        - [health, 0.1, 1]
+    delta_t: 0
+    immediate: true
+    neigh_criteria:
+        - [edge, [target, oldLoc], atLoc, [1,1]]
+        - [edge, [oldLoc, newLoc], up, [1,1]]
+        - [node, [newLoc], blocked, [0,0]]
+
+    edges: [newLoc, target, atLoc]
+    ann_fn: [0.5,0.5]
+
+move_up_change_prev_rule_1_r1:
+    target: atLoc
+    target_criteria:
+        - [atLoc, 1, 1]
+    delta_t: 0
+    immediate: true
+    neigh_criteria:
+        - [node, [source], red-soldier-1, [1,1]]
+        - [edge, [target, newLoc], up, [1,1]]
+        - [edge, [source, newLoc], atLoc, [0.5,0.5]]
+
+    ann_fn: [0,0]
+
+move_up_change_prev_rule_2_r1:
+    target: atLoc
+    target_criteria:
+        - [atLoc, 1, 1]
+    delta_t: 0
+    immediate: true
+    neigh_criteria:
+        - [node, [target], red-soldier-1, [1,1]]
+        - [edge, [source, newLoc], up, [1,1]]
+        - [edge, [newLoc, target], atLoc, [0.5,0.5]]
+
+    ann_fn: [0,0]
+
+# Move DOWN
+move_down_rule_1_r1:
+    target:
+    target_criteria:
+        - [red-soldier-1, 1, 1]
+        - [moveDown, 1, 1]
+        - [health, 0.1, 1]
+    delta_t: 0
+    immediate: true
+    neigh_criteria:
+        - [edge, [target, oldLoc], atLoc, [1,1]]
+        - [edge, [oldLoc, newLoc], down, [1,1]]
+        - [node, [newLoc], blocked, [0,0]]
+
+    edges: [target, newLoc, atLoc]
+    ann_fn: [0.5,0.5]
+
+move_down_rule_2_r1:
+    target:
+    target_criteria:
+        - [red-soldier-1, 1, 1]
+        - [moveDown, 1, 1]
+        - [health, 0.1, 1]
+    delta_t: 0
+    immediate: true
+    neigh_criteria:
+        - [edge, [target, oldLoc], atLoc, [1,1]]
+        - [edge, [oldLoc, newLoc], down, [1,1]]
+        - [node, [newLoc], blocked, [0,0]]
+
+    edges: [newLoc, target, atLoc]
+    ann_fn: [0.5,0.5]
+
+move_down_change_prev_rule_1_r1:
+    target: atLoc
+    target_criteria:
+        - [atLoc, 1, 1]
+    delta_t: 0
+    immediate: true
+    neigh_criteria:
+        - [node, [source], red-soldier-1, [1,1]]
+        - [edge, [target, newLoc], down, [1,1]]
+        - [edge, [source, newLoc], atLoc, [0.5,0.5]]
+
+    ann_fn: [0,0]
+
+move_down_change_prev_rule_2_r1:
+    target: atLoc
+    target_criteria:
+        - [atLoc, 1, 1]
+    delta_t: 0
+    immediate: true
+    neigh_criteria:
+        - [node, [target], red-soldier-1, [1,1]]
+        - [edge, [source, newLoc], down, [1,1]]
+        - [edge, [newLoc, target], atLoc, [0.5,0.5]]
+
+    ann_fn: [0,0]
+
+# Move LEFT
+move_left_rule_1_r1:
+    target:
+    target_criteria:
+        - [red-soldier-1, 1, 1]
+        - [moveLeft, 1, 1]
+        - [health, 0.1, 1]
+    delta_t: 0
+    immediate: true
+    neigh_criteria:
+        - [edge, [target, oldLoc], atLoc, [1,1]]
+        - [edge, [oldLoc, newLoc], left, [1,1]]
+        - [node, [newLoc], blocked, [0,0]]
+
+    edges: [target, newLoc, atLoc]
+    ann_fn: [0.5,0.5]
+
+move_left_rule_2_r1:
+    target:
+    target_criteria:
+        - [red-soldier-1, 1, 1]
+        - [moveLeft, 1, 1]
+        - [health, 0.1, 1]
+    delta_t: 0
+    immediate: true
+    neigh_criteria:
+        - [edge, [target, oldLoc], atLoc, [1,1]]
+        - [edge, [oldLoc, newLoc], left, [1,1]]
+        - [node, [newLoc], blocked, [0,0]]
+
+    edges: [newLoc, target, atLoc]
+    ann_fn: [0.5,0.5]
+
+move_left_change_prev_rule_1_r1:
+    target: atLoc
+    target_criteria:
+        - [atLoc, 1, 1]
+    delta_t: 0
+    immediate: true
+    neigh_criteria:
+        - [node, [source], red-soldier-1, [1,1]]
+        - [edge, [target, newLoc], left, [1,1]]
+        - [edge, [source, newLoc], atLoc, [0.5,0.5]]
+
+    ann_fn: [0,0]
+
+move_left_change_prev_rule_2_r1:
+    target: atLoc
+    target_criteria:
+        - [atLoc, 1, 1]
+    delta_t: 0
+    immediate: true
+    neigh_criteria:
+        - [node, [target], red-soldier-1, [1,1]]
+        - [edge, [source, newLoc], left, [1,1]]
+        - [edge, [newLoc, target], atLoc, [0.5,0.5]]
+
+    ann_fn: [0,0]
+
+# Move RIGHT
+move_right_rule_1_r1:
+    target:
+    target_criteria:
+        - [red-soldier-1, 1, 1]
+        - [moveRight, 1, 1]
+        - [health, 0.1, 1]
+    delta_t: 0
+    immediate: true
+    neigh_criteria:
+        - [edge, [target, oldLoc], atLoc, [1,1]]
+        - [edge, [oldLoc, newLoc], right, [1,1]]
+        - [node, [newLoc], blocked, [0,0]]
+
+    edges: [target, newLoc, atLoc]
+    ann_fn: [0.5,0.5]
+
+move_right_rule_2_r1:
+    target:
+    target_criteria:
+        - [red-soldier-1, 1, 1]
+        - [moveRight, 1, 1]
+        - [health, 0.1, 1]
+    delta_t: 0
+    immediate: true
+    neigh_criteria:
+        - [edge, [target, oldLoc], atLoc, [1,1]]
+        - [edge, [oldLoc, newLoc], right, [1,1]]
+        - [node, [newLoc], blocked, [0,0]]
+
+    edges: [newLoc, target, atLoc]
+    ann_fn: [0.5,0.5]
+
+move_right_change_prev_rule_1_r1:
+    target: atLoc
+    target_criteria:
+        - [atLoc, 1, 1]
+    delta_t: 0
+    immediate: true
+    neigh_criteria:
+        - [node, [source], red-soldier-1, [1,1]]
+        - [edge, [target, newLoc], right, [1,1]]
+        - [edge, [source, newLoc], atLoc, [0.5,0.5]]
+
+    ann_fn: [0,0]
+
+move_right_change_prev_rule_2_r1:
+    target: atLoc
+    target_criteria:
+        - [atLoc, 1, 1]
+    delta_t: 0
+    immediate: true
+    neigh_criteria:
+        - [node, [target], red-soldier-1, [1,1]]
+        - [edge, [source, newLoc], right, [1,1]]
+        - [edge, [newLoc, target], atLoc, [0.5,0.5]]
+
+    ann_fn: [0,0]
+
+
+# red-agent-2
+# Moving rules
+# Move UP
+move_up_rule_1_r2:
+    target:
+    target_criteria:
+        - [red-soldier-2, 1, 1]
+        - [moveUp, 1, 1]
+        - [health, 0.1, 1]
+    delta_t: 1
+    immediate: true
+    neigh_criteria:
+        - [edge, [target, oldLoc], atLoc, [1,1]]
+        - [edge, [oldLoc, newLoc], up, [1,1]]
+        - [node, [newLoc], blocked, [0,0]]
+
+    edges: [target, newLoc, atLoc]
+    ann_fn: [0.5,0.5]
+
+move_up_rule_2_r2:
+    target:
+    target_criteria:
+        - [red-soldier-2, 1, 1]
+        - [moveUp, 1, 1]
+        - [health, 0.1, 1]
+    delta_t: 1
+    immediate: true
+    neigh_criteria:
+        - [edge, [target, oldLoc], atLoc, [1,1]]
+        - [edge, [oldLoc, newLoc], up, [1,1]]
+        - [node, [newLoc], blocked, [0,0]]
+
+    edges: [newLoc, target, atLoc]
+    ann_fn: [0.5,0.5]
+
+move_up_change_prev_rule_1_r2:
+    target: atLoc
+    target_criteria:
+        - [atLoc, 1, 1]
+    delta_t: 0
+    immediate: true
+    neigh_criteria:
+        - [node, [source], red-soldier-2, [1,1]]
+        - [edge, [target, newLoc], up, [1,1]]
+        - [edge, [source, newLoc], atLoc, [0.5,0.5]]
+
+    ann_fn: [0,0]
+
+move_up_change_prev_rule_2_r2:
+    target: atLoc
+    target_criteria:
+        - [atLoc, 1, 1]
+    delta_t: 0
+    immediate: true
+    neigh_criteria:
+        - [node, [target], red-soldier-2, [1,1]]
+        - [edge, [source, newLoc], up, [1,1]]
+        - [edge, [newLoc, target], atLoc, [0.5,0.5]]
+
+    ann_fn: [0,0]
+
+# Move DOWN
+move_down_rule_1_r2:
+    target:
+    target_criteria:
+        - [red-soldier-2, 1, 1]
+        - [moveDown, 1, 1]
+        - [health, 0.1, 1]
+    delta_t: 1
+    immediate: true
+    neigh_criteria:
+        - [edge, [target, oldLoc], atLoc, [1,1]]
+        - [edge, [oldLoc, newLoc], down, [1,1]]
+        - [node, [newLoc], blocked, [0,0]]
+
+    edges: [target, newLoc, atLoc]
+    ann_fn: [0.5,0.5]
+
+move_down_rule_2_r2:
+    target:
+    target_criteria:
+        - [red-soldier-2, 1, 1]
+        - [moveDown, 1, 1]
+        - [health, 0.1, 1]
+    delta_t: 1
+    immediate: true
+    neigh_criteria:
+        - [edge, [target, oldLoc], atLoc, [1,1]]
+        - [edge, [oldLoc, newLoc], down, [1,1]]
+        - [node, [newLoc], blocked, [0,0]]
+
+    edges: [newLoc, target, atLoc]
+    ann_fn: [0.5,0.5]
+
+move_down_change_prev_rule_1_r2:
+    target: atLoc
+    target_criteria:
+        - [atLoc, 1, 1]
+    delta_t: 0
+    immediate: true
+    neigh_criteria:
+        - [node, [source], red-soldier-2, [1,1]]
+        - [edge, [target, newLoc], down, [1,1]]
+        - [edge, [source, newLoc], atLoc, [0.5,0.5]]
+
+    ann_fn: [0,0]
+
+move_down_change_prev_rule_2_r2:
+    target: atLoc
+    target_criteria:
+        - [atLoc, 1, 1]
+    delta_t: 0
+    immediate: true
+    neigh_criteria:
+        - [node, [target], red-soldier-2, [1,1]]
+        - [edge, [source, newLoc], down, [1,1]]
+        - [edge, [newLoc, target], atLoc, [0.5,0.5]]
+
+    ann_fn: [0,0]
+
+# Move LEFT
+move_left_rule_1_r2:
+    target:
+    target_criteria:
+        - [red-soldier-2, 1, 1]
+        - [moveLeft, 1, 1]
+        - [health, 0.1, 1]
+    delta_t: 1
+    immediate: true
+    neigh_criteria:
+        - [edge, [target, oldLoc], atLoc, [1,1]]
+        - [edge, [oldLoc, newLoc], left, [1,1]]
+        - [node, [newLoc], blocked, [0,0]]
+
+    edges: [target, newLoc, atLoc]
+    ann_fn: [0.5,0.5]
+
+move_left_rule_2_r2:
+    target:
+    target_criteria:
+        - [red-soldier-2, 1, 1]
+        - [moveLeft, 1, 1]
+        - [health, 0.1, 1]
+    delta_t: 1
+    immediate: true
+    neigh_criteria:
+        - [edge, [target, oldLoc], atLoc, [1,1]]
+        - [edge, [oldLoc, newLoc], left, [1,1]]
+        - [node, [newLoc], blocked, [0,0]]
+
+    edges: [newLoc, target, atLoc]
+    ann_fn: [0.5,0.5]
+
+move_left_change_prev_rule_1_r2:
+    target: atLoc
+    target_criteria:
+        - [atLoc, 1, 1]
+    delta_t: 0
+    immediate: true
+    neigh_criteria:
+        - [node, [source], red-soldier-2, [1,1]]
+        - [edge, [target, newLoc], left, [1,1]]
+        - [edge, [source, newLoc], atLoc, [0.5,0.5]]
+
+    ann_fn: [0,0]
+
+move_left_change_prev_rule_2_r2:
+    target: atLoc
+    target_criteria:
+        - [atLoc, 1, 1]
+    delta_t: 0
+    immediate: true
+    neigh_criteria:
+        - [node, [target], red-soldier-2, [1,1]]
+        - [edge, [source, newLoc], left, [1,1]]
+        - [edge, [newLoc, target], atLoc, [0.5,0.5]]
+
+    ann_fn: [0,0]
+
+# Move RIGHT
+move_right_rule_1_r2:
+    target:
+    target_criteria:
+        - [red-soldier-2, 1, 1]
+        - [moveRight, 1, 1]
+        - [health, 0.1, 1]
+    delta_t: 1
+    immediate: true
+    neigh_criteria:
+        - [edge, [target, oldLoc], atLoc, [1,1]]
+        - [edge, [oldLoc, newLoc], right, [1,1]]
+        - [node, [newLoc], blocked, [0,0]]
+
+    edges: [target, newLoc, atLoc]
+    ann_fn: [0.5,0.5]
+
+move_right_rule_2_r2:
+    target:
+    target_criteria:
+        - [red-soldier-2, 1, 1]
+        - [moveRight, 1, 1]
+        - [health, 0.1, 1]
+    delta_t: 1
+    immediate: true
     neigh_criteria:
         - [edge, [target, oldLoc], atLoc, [1,1]]
         - [edge, [oldLoc, newLoc], right, [1,1]]
@@ -333,25 +1098,27 @@ move_right_rule_2:
     edges: [newLoc, target, atLoc]
     ann_fn: [0.5,0.5]
 
-move_right_change_prev_rule_1:
+move_right_change_prev_rule_1_r2:
     target: atLoc
     target_criteria:
         - [atLoc, 1, 1]
     delta_t: 0
     immediate: true
     neigh_criteria:
+        - [node, [source], red-soldier-2, [1,1]]
         - [edge, [target, newLoc], right, [1,1]]
         - [edge, [source, newLoc], atLoc, [0.5,0.5]]
 
     ann_fn: [0,0]
 
-move_right_change_prev_rule_2:
+move_right_change_prev_rule_2_r2:
     target: atLoc
     target_criteria:
         - [atLoc, 1, 1]
     delta_t: 0
     immediate: true
     neigh_criteria:
+        - [node, [target], red-soldier-2, [1,1]]
         - [edge, [source, newLoc], right, [1,1]]
         - [edge, [newLoc, target], atLoc, [0.5,0.5]]
 
diff --git a/tests/multi_agent_shoot/gen_graph_test.py b/tests/multi_agent_shoot/gen_graph_test.py
index 893ace2..e22e816 100644
--- a/tests/multi_agent_shoot/gen_graph_test.py
+++ b/tests/multi_agent_shoot/gen_graph_test.py
@@ -72,10 +72,12 @@ def generate_graph(grid_dim, num_agents_per_team, base_loc, start_loc, obstacle_
     # =======================================================================
     # Initialize players health, action choice and team
     for i in range(1, num_agents_per_team + 1):
+        dict1 = {f'red-soldier-{i}': 1}
+        dict2 = {f'blue-soldier-{i}': 1}
         g.add_node(f'red-soldier-{i}', health=1, moveUp=0, moveDown=0, moveLeft=0, moveRight=0, shootUpRed=0,
-                   shootDownRed=0, shootLeftRed=0, shootRightRed=0, teamRed=1, justDied='0,0')
+                   shootDownRed=0, shootLeftRed=0, shootRightRed=0, teamRed=1, justDied='0,0', **dict1)
         g.add_node(f'blue-soldier-{i}', health=1, moveUp=0, moveDown=0, moveLeft=0, moveRight=0, shootUpBlue=0,
-                   shootDownBlue=0, shootLeftBlue=0, shootRightBlue=0, teamBlue=1, justDied='0,0')
+                   shootDownBlue=0, shootLeftBlue=0, shootRightBlue=0, teamBlue=1, justDied='0,0', **dict2)
         # Teams
         g.add_edge(f'red-soldier-{i}', 'red-base', team=1)
         g.add_edge(f'blue-soldier-{i}', 'blue-base', team=1)
@@ -97,7 +99,7 @@ def generate_graph(grid_dim, num_agents_per_team, base_loc, start_loc, obstacle_
 
 def main():
     ## Red is first then Blue
-    generate_graph(grid_dim=8, num_agents_per_team=2, base_loc=[7, 56], start_loc=[[2,16], [5, 21]],
+    generate_graph(grid_dim=8, num_agents_per_team=2, base_loc=[7, 56], start_loc=[[0,16], [5, 21]],
                    obstacle_loc=[26, 27, 34, 35, 36, 44])
 
 
diff --git a/tests/multi_agent_shoot/test.py b/tests/multi_agent_shoot/test.py
index 171c484..93c8da2 100644
--- a/tests/multi_agent_shoot/test.py
+++ b/tests/multi_agent_shoot/test.py
@@ -7,42 +7,63 @@
 
 action = {
     'red_team': [7,7],
-    'blue_team': [1,3]
+    'blue_team': [6,6]
 }
 obs = env.step(action)
 print(obs)
 time.sleep(1)
 action = {
-    'red_team': [0,0],
-    'blue_team': [1,0]
+    'red_team': [2,8],
+    'blue_team': [1,8]
 }
 obs = env.step(action)
 print(obs)
 time.sleep(1)
 action = {
-    'red_team': [1,2],
-    'blue_team': [1,0]
+    'red_team': [2,2],
+    'blue_team': [6,6]
 }
 obs = env.step(action)
 print(obs)
 time.sleep(1)
 action = {
-    'red_team': [1,2],
-    'blue_team': [1,0]
+    'red_team': [2,8],
+    'blue_team': [1,8]
 }
 obs = env.step(action)
 print(obs)
 time.sleep(1)
 action = {
     'red_team': [2,2],
-    'blue_team': [1,0]
+    'blue_team': [1,1]
+}
+obs = env.step(action)
+print(obs)
+time.sleep(1)
+action = {
+    'red_team': [2,8],
+    'blue_team': [1,8]
+}
+obs = env.step(action)
+print(obs)
+time.sleep(1)
+action = {
+    'red_team': [2,2],
+    'blue_team': [1,1]
+}
+obs = env.step(action)
+print(obs)
+time.sleep(1)
+action = {
+    'red_team': [2,8],
+    'blue_team': [1,8]
 }
 obs = env.step(action)
 print(obs)
 time.sleep(1)
 action = {
     'red_team': [2,2],
-    'blue_team': [1,0]
+    'blue_team': [1,1]
 }
 obs = env.step(action)
 print(obs)
diff --git a/tests/non_markov_map/gen_graph_test.py b/tests/non_markov_map/gen_graph_test.py
new file mode 100644
index 0000000..c5860d7
--- /dev/null
+++ b/tests/non_markov_map/gen_graph_test.py
@@ -0,0 +1,107 @@
+import networkx as nx
+
+
+def generate_graph(grid_dim, num_agents_per_team, base_loc, start_loc, obstacle_loc):
+    # Check parameters
+    assert len(base_loc) == 2, 'There are only two bases--supply two locations'
+    assert len(start_loc) == 2, 'There are only two teams--supply lists of start positions for each team in a nested list'
+    assert len(start_loc[0]) == num_agents_per_team and len(start_loc[1]) == num_agents_per_team, 'Supply correct number of start locations'
+
+    g = nx.DiGraph()
+
+    # Game variables
+    game_height = grid_dim
+    game_width = grid_dim
+
+    # =======================================================================
+    # Add a node for each grid location in the game
+    nodes = list(range(0, game_height * game_width))
+    g.add_nodes_from([str(node) for node in nodes], blocked='0,0')
+
+    # =======================================================================
+    # Add edges connecting each of the grid nodes. Add up, down, left, right attributes to correct edges
+    # Right edges
+    for node in g.nodes:
+        if (int(node) + 1) % game_width != 0:
+            g.add_edge(node, str(int(node) + 1), right=1)
+
+    # Left edges
+    for node in g.nodes:
+        if int(node) % game_width != 0:
+            g.add_edge(node, str(int(node) - 1), left=1)
+
+    # Up edges
+    for node in g.nodes:
+        if (int(node) // game_width) + 1 != game_height:
+            g.add_edge(node, str(int(node) + game_width), up=1)
+
+    # Down edges
+    for node in g.nodes:
+        if int(node) // game_width != 0:
+            g.add_edge(node, str(int(node) - game_width), down=1)
+
+    # Add edges between border nodes and end nodes to demarcate the end of the grid world. Bullets will disappear after it crosses this end
+    g.add_node('end', blocked=1)
+    # Bottom border
+    for i in range(game_width):
+        g.add_edge(f'{i}', 'end', down=1)
+    # Top border
+    for i in range(game_width):
+        g.add_edge(f'{i + game_width * (game_height - 1)}', 'end', up=1)
+    # Left border
+    for i in range(game_height):
+        g.add_edge(f'{i * game_width}', 'end', left=1)
+    # Right border
+    for i in range(game_height):
+        g.add_edge(f'{i * game_width + game_width - 1}', 'end', right=1)
+
+    # =======================================================================
+    # Add the bases and connect them to the correct location: bottom right and top left
+    g.add_node('red-base')
+    g.add_node('blue-base')
+    g.add_edge('red-base', str(base_loc[0]), atLoc=1)
+    g.add_edge('blue-base', str(base_loc[1]), atLoc=1)
+
+    # =======================================================================
+    # Add mountains and obstacle attributes
+    mountain_loc = obstacle_loc
+    g.add_node('mountain', isMountain=1)
+    for i in mountain_loc:
+        g.add_edge(str(i), 'mountain', atLoc=1)
+
+    # =======================================================================
+    # Initialize players health, action choice and team
+    for i in range(1, num_agents_per_team + 1):
+        dict1 = {f'red-soldier-{i}': 1}
+        dict2 = {f'blue-soldier-{i}': 1}
+        g.add_node(f'red-soldier-{i}', health=1, moveUp=0, moveDown=0, moveLeft=0, moveRight=0, shootUpRed=0,
+                   shootDownRed=0, shootLeftRed=0, shootRightRed=0, teamRed=1, justDied='0,0', **dict1)
+        g.add_node(f'blue-soldier-{i}', health=1, moveUp=0, moveDown=0, moveLeft=0, moveRight=0, shootUpBlue=0,
+                   shootDownBlue=0, shootLeftBlue=0, shootRightBlue=0, teamBlue=1, justDied='0,0', **dict2)
+        # Teams
+        g.add_edge(f'red-soldier-{i}', 'red-base', team=1)
+        g.add_edge(f'blue-soldier-{i}', 'blue-base', team=1)
+        # Soldier Start Locations (dual edge)
+        g.add_edge(f'red-soldier-{i}', str(start_loc[0][i - 1]), atLoc=1)
+        g.add_edge(f'blue-soldier-{i}', str(start_loc[1][i - 1]), atLoc=1)
+        g.add_edge(str(start_loc[0][i - 1]), f'red-soldier-{i}', atLoc=1)
+        g.add_edge(str(start_loc[1][i - 1]), f'blue-soldier-{i}', atLoc=1)
+        # Bullets
+        g.add_node(f'red-bullet-{i}', teamRed=1, bullet=1)
+        g.add_node(f'blue-bullet-{i}', teamBlue=1, bullet=1)
+        g.add_edge(f'red-soldier-{i}', f'red-bullet-{i}', bullet=1)
+        g.add_edge(f'blue-soldier-{i}', f'blue-bullet-{i}', bullet=1)
+
+    # =======================================================================
+
+    nx.write_graphml_lxml(g, 'pyreason_gym/pyreason_grid_world/graph/game_graph.graphml', named_key_ids=True)
+
+
+def main():
+    ## Red is first then Blue
+    generate_graph(grid_dim=8, num_agents_per_team=2, base_loc=[7, 56], start_loc=[[7,7], [56, 56]],
+                   obstacle_loc=[26, 27, 34, 35, 36, 44])
+
+
+if __name__ == '__main__':
+    main()
diff --git a/tests/non_markov_test/gen_graph_test.py b/tests/non_markov_test/gen_graph_test.py
new file mode 100644
index 0000000..7e2b699
--- /dev/null
+++ b/tests/non_markov_test/gen_graph_test.py
@@ -0,0 +1,107 @@
+import networkx as nx
+
+
+def generate_graph(grid_dim, num_agents_per_team, base_loc, start_loc, obstacle_loc):
+    # Check parameters
+    assert len(base_loc) == 2, 'There are only two bases--supply two locations'
+    assert len(start_loc) == 2, 'There are only two teams--supply lists of start positions for each team in a nested list'
+    assert len(start_loc[0]) == num_agents_per_team and len(start_loc[1]) == num_agents_per_team, 'Supply correct number of start locations'
+
+    g = nx.DiGraph()
+
+    # Game variables
+    game_height = grid_dim
+    game_width = grid_dim
+
+    # =======================================================================
+    # Add a node for each grid location in the game
+    nodes = list(range(0, game_height * game_width))
+    g.add_nodes_from([str(node) for node in nodes], blocked='0,0')
+
+    # =======================================================================
+    # Add edges connecting each of the grid nodes. Add up, down, left, right attributes to correct edges
+    # Right edges
+    for node in g.nodes:
+        if (int(node) + 1) % game_width != 0:
+            g.add_edge(node, str(int(node) + 1), right=1)
+
+    # Left edges
+    for node in g.nodes:
+        if int(node) % game_width != 0:
+            g.add_edge(node, str(int(node) - 1), left=1)
+
+    # Up edges
+    for node in g.nodes:
+        if (int(node) // game_width) + 1 != game_height:
+            g.add_edge(node, str(int(node) + game_width), up=1)
+
+    # Down edges
+    for node in g.nodes:
+        if int(node) // game_width != 0:
+            g.add_edge(node, str(int(node) - game_width), down=1)
+
+    # Add edges between border nodes and end nodes to demarcate the end of the grid world. Bullets will disappear after it crosses this end
+    g.add_node('end', blocked=1)
+    # Bottom border
+    for i in range(game_width):
+        g.add_edge(f'{i}', 'end', down=1)
+    # Top border
+    for i in range(game_width):
+        g.add_edge(f'{i + game_width * (game_height - 1)}', 'end', up=1)
+    # Left border
+    for i in range(game_height):
+        g.add_edge(f'{i * game_width}', 'end', left=1)
+    # Right border
+    for i in range(game_height):
+        g.add_edge(f'{i * game_width + game_width - 1}', 'end', right=1)
+
+    # =======================================================================
+    # Add the bases and connect them to the correct location: bottom right and top left
+    g.add_node('red-base')
+    g.add_node('blue-base')
+    g.add_edge('red-base', str(base_loc[0]), atLoc=1)
+    g.add_edge('blue-base', str(base_loc[1]), atLoc=1)
+
+    # =======================================================================
+    # Add mountains and obstacle attributes
+    mountain_loc = obstacle_loc
+    g.add_node('mountain', isMountain=1)
+    for i in mountain_loc:
+        g.add_edge(str(i), 'mountain', atLoc=1)
+
+    # =======================================================================
+    # Initialize players health, action choice and team
+    for i in range(1, num_agents_per_team + 1):
+        dict1 = {f'red-soldier-{i}': 1}
+        dict2 = {f'blue-soldier-{i}': 1}
+        g.add_node(f'red-soldier-{i}', health=1, moveUp=0, moveDown=0, moveLeft=0, moveRight=0, shootUpRed=0,
+                   shootDownRed=0, shootLeftRed=0, shootRightRed=0, teamRed=1, justDied='0,0', **dict1)
+        g.add_node(f'blue-soldier-{i}', health=1, moveUp=0, moveDown=0, moveLeft=0, moveRight=0, shootUpBlue=0,
+                   shootDownBlue=0, shootLeftBlue=0, shootRightBlue=0, teamBlue=1, justDied='0,0', **dict2)
+        # Teams
+        g.add_edge(f'red-soldier-{i}', 'red-base', team=1)
+        g.add_edge(f'blue-soldier-{i}', 'blue-base', team=1)
+        # Soldier Start Locations (dual edge)
+        g.add_edge(f'red-soldier-{i}', str(start_loc[0][i - 1]), atLoc=1)
+        g.add_edge(f'blue-soldier-{i}', str(start_loc[1][i - 1]), atLoc=1)
+        g.add_edge(str(start_loc[0][i - 1]), f'red-soldier-{i}', atLoc=1)
+        g.add_edge(str(start_loc[1][i - 1]), f'blue-soldier-{i}', atLoc=1)
+        # Bullets
+        g.add_node(f'red-bullet-{i}', teamRed=1, bullet=1)
+        g.add_node(f'blue-bullet-{i}', teamBlue=1, bullet=1)
+        g.add_edge(f'red-soldier-{i}', f'red-bullet-{i}', bullet=1)
+        g.add_edge(f'blue-soldier-{i}', f'blue-bullet-{i}', bullet=1)
+
+    # =======================================================================
+
+    nx.write_graphml_lxml(g, 'pyreason_gym/pyreason_grid_world/graph/game_graph.graphml', named_key_ids=True)
+
+
+def main():
+    ## Red is first then Blue
+    generate_graph(grid_dim=8, num_agents_per_team=2, base_loc=[7, 56], start_loc=[[48,56], [55, 63]],
+                   obstacle_loc=[26, 27, 34, 35, 36, 44])
+
+
+if __name__ == '__main__':
+    main()
diff --git a/tests/non_markov_test/test.py b/tests/non_markov_test/test.py
new file mode 100644
index 0000000..6cffbfe
--- /dev/null
+++ b/tests/non_markov_test/test.py
@@ -0,0 +1,38 @@
+import pyreason_gym
+import gym
+import time
+
+env = gym.make('PyReasonGridWorld-v0', render_mode='human', num_agents_per_team=2)
+obs = env.reset()
+print("\n")
+print("Starting Experiment with init obs:")
+print(obs)
+print("\n")
+action = {
+            'red_team':  [7, 7],
+            'blue_team': [2, 2]
+        }
+obs = env.step(action)
+print(obs)
+print("\n")
+time.sleep(1)
+for ss in range(6):
+    print("Step", ss)
+    if ss % 2 == 0:
+        print("Taking both actions")
+        action = {
+            'red_team':  [2, 2],
+            'blue_team': [2, 2]
+        }
+    else:
+        print("Taking first agent actions")
+        action = {
+            'red_team':  [2,2],
+            'blue_team': [8,2]
+        }
+    obs = env.step(action)
+    print(obs)
+    print("\n")
+    time.sleep(1)
+
+env.close()
\ No newline at end of file
diff --git a/tests/same_location_shoot_all_dir/test.py b/tests/same_location_shoot_all_dir/test.py
index e6dacad..1215b20 100644
--- a/tests/same_location_shoot_all_dir/test.py
+++ b/tests/same_location_shoot_all_dir/test.py
@@ -2,7 +2,7 @@
 import gym
 import time
 
-env = gym.make('PyReasonGridWorld-v0', render_mode='human')
+env = gym.make('PyReasonGridWorld-v0', render_mode='human', num_agents_per_team=1)
 obs = env.reset()
 action = {
     'red_team': [3],
@@ -39,11 +39,4 @@
 obs = env.step(action)
 print(obs)
 time.sleep(1)
-action = {
-    'red_team': [2],
-    'blue_team': [1]
-}
-obs = env.step(action)
-print(obs)
-time.sleep(1)
 env.close()
\ No newline at end of file