updating obs and reward to fix backboard bounce logic

Constants_demo.py

@@ -2,7 +2,7 @@
 from pretrained_agents.KBB.kbb import KBB
 
 FRAME_SKIP = 4
-TIME_HORIZON = 2  # horizon in seconds
+TIME_HORIZON = 3  # horizon in seconds
 T_STEP = FRAME_SKIP / 120   # real time per rollout step
 ZERO_SUM = False
 STEP_SIZE = 500_000

CoyoteObs.py

@@ -17,7 +17,7 @@
 from gym import Space
 from gym.spaces import Tuple, Box
 
-from rlgym.utils.common_values import BOOST_LOCATIONS, BALL_RADIUS
+from rlgym.utils.common_values import BOOST_LOCATIONS, BALL_RADIUS, BACK_WALL_Y
 
 from numba import njit
 
@@ -266,12 +266,18 @@ def pre_step(self, state: GameState):
 
         # for double tap
         if self.doubletap_indicator:
-            if state.ball.position[2] < BALL_RADIUS * 2 and 0.55 * self.prev_ball_vel[2] \
-                    < state.ball.linear_velocity[2] > 0.65 * self.prev_ball_vel[2]:
+            touched = False
+            for player in state.players:
+                if player.ball_touched:
+                    touched = True
+            ball_bounced_ground = self.prev_ball_vel[2] * state.ball.linear_velocity[2] < 0
+            ball_near_ground = state.ball.position[2] < BALL_RADIUS * 2
+            if not touched and ball_near_ground and ball_bounced_ground:
                 self.floor_bounce = True
-            elif 0.55 * self.prev_ball_vel[1] < state.ball.linear_velocity[1] > 0.65 * \
-                    self.prev_ball_vel[1] and \
-                    abs(state.ball.position[1]) > 4900 and state.ball.position[2] > 500:
+
+            ball_bounced_backboard = self.prev_ball_vel[1] * state.ball.linear_velocity[1] < 0
+            ball_near_wall = abs(state.ball.position[1]) > (BACK_WALL_Y - BALL_RADIUS * 2)
+            if not touched and ball_near_wall and ball_bounced_backboard:
                 self.backboard_bounce = True
             self.prev_ball_vel = np.array(state.ball.linear_velocity)
 

learner_demo.py

@@ -51,10 +51,10 @@
         ent_coef=0.01,
     )
 
-    run_id = "demo_run1.00"
+    run_id = "demo_run1.02"
     wandb.login(key=os.environ["WANDB_KEY"])
     logger = wandb.init(dir="./wandb_store",
-                        name="Demo_Run1.00",
+                        name="Demo_Run1.02",
                         project="Opti",
                         entity="kaiyotech",
                         id=run_id,
@@ -136,11 +136,11 @@
         disable_gradient_logging=True,
     )
 
-    # alg.load("Demo_saves/Opti_1683473991.4737124/Opti_5080/checkpoint.pt")
+    alg.load("Demo_saves/Opti_1683728019.129012/Opti_2140/checkpoint.pt")
 
     alg.agent.optimizer.param_groups[0]["lr"] = logger.config.actor_lr
     alg.agent.optimizer.param_groups[1]["lr"] = logger.config.critic_lr
 
-    alg.freeze_policy(50)
+    # alg.freeze_policy(50)
 
     alg.run(iterations_per_save=logger.config.save_every, save_dir="Demo_saves")

rewards.py

@@ -286,13 +286,27 @@ def pre_step(self, state: GameState):
         self.orange_touch_timer += 1
         self.kickoff_timer += 1
         # for double tap
-        if state.ball.position[2] < BALL_RADIUS * 2 and 0.55 * self.last_state.ball.linear_velocity[2] \
-                < state.ball.linear_velocity[2] > 0.65 * self.last_state.ball.linear_velocity[2]:
+        # if state.ball.position[2] < BALL_RADIUS * 2 and 0.55 * self.last_state.ball.linear_velocity[2] \
+        #         < state.ball.linear_velocity[2] > 0.65 * self.last_state.ball.linear_velocity[2]:
+        #     self.floor_bounce = True
+        # elif 0.55 * self.last_state.ball.linear_velocity[1] < state.ball.linear_velocity[1] > 0.65 * \
+        #         self.last_state.ball.linear_velocity[1] and \
+        #         abs(state.ball.position[1]) > 4900 and state.ball.position[2] > 500:
+        #     self.backboard_bounce = True
+        touched = False
+        for player in state.players:
+            if player.ball_touched:
+                touched = True
+        ball_bounced_ground = self.last_state.ball.linear_velocity[2] * state.ball.linear_velocity[2] < 0
+        ball_near_ground = state.ball.position[2] < BALL_RADIUS * 2
+        if not touched and ball_near_ground and ball_bounced_ground:
             self.floor_bounce = True
-        elif 0.55 * self.last_state.ball.linear_velocity[1] < state.ball.linear_velocity[1] > 0.65 * \
-                self.last_state.ball.linear_velocity[1] and \
-                abs(state.ball.position[1]) > 4900 and state.ball.position[2] > 500:
+
+        ball_bounced_backboard = self.last_state.ball.linear_velocity[1] * state.ball.linear_velocity[1] < 0
+        ball_near_wall = abs(state.ball.position[1]) > (BACK_WALL_Y - BALL_RADIUS * 2)
+        if not touched and ball_near_wall and ball_bounced_backboard:
             self.backboard_bounce = True
+
         # for aerial
         # player who last touched is now on the ground, don't allow jumping back up to continue "dribble"
         if self.blue_touch_timer < self.orange_touch_timer and state.players[self.blue_toucher].on_ground: