diff --git a/examples/rac/rac_data_collection_openarms_rtc.py b/examples/rac/rac_data_collection_openarms_rtc.py
index 9efa6d88f..030db0bc8 100644
--- a/examples/rac/rac_data_collection_openarms_rtc.py
+++ b/examples/rac/rac_data_collection_openarms_rtc.py
@@ -257,7 +257,7 @@ def get_actions_thread(
     policy,
     robot: RobotWrapper,
     robot_observation_processor,
-    action_queue: ActionQueue,
+    action_queue_holder: dict,
     shutdown_event: Event,
     cfg: RaCRTCConfig,
     policy_active: Event,
@@ -289,6 +289,7 @@ def get_actions_thread(
                 time.sleep(0.01)
                 continue
 
+            action_queue = action_queue_holder["queue"]
             if action_queue.qsize() <= get_actions_threshold:
                 current_time = time.perf_counter()
                 action_index_before_inference = action_queue.get_action_index()
@@ -429,11 +430,11 @@ def main(cfg: RaCRTCConfig):
     policy.eval()
     logger.info(f"Policy loaded: {policy.name}")
 
-    action_queue = ActionQueue(cfg.rtc)
+    action_queue_holder = {"queue": ActionQueue(cfg.rtc)}
 
     get_actions_t = Thread(
         target=get_actions_thread,
-        args=(policy, robot, obs_proc, action_queue, shutdown_event, cfg, policy_active, fps),
+        args=(policy, robot, obs_proc, action_queue_holder, shutdown_event, cfg, policy_active, fps),
         daemon=True,
         name="GetActions",
     )
@@ -462,7 +463,8 @@ def main(cfg: RaCRTCConfig):
         while recorded < cfg.dataset.num_episodes and not events["stop_recording"]:
             log_say(f"RaC episode {recorded + 1}", play_sounds=cfg.play_sounds)
 
-            action_queue = ActionQueue(cfg.rtc)
+            action_queue_holder["queue"] = ActionQueue(cfg.rtc)
+            action_queue = action_queue_holder["queue"]
             events["policy_paused"] = False
             events["correction_active"] = False
             events["start_correction"] = False