# tf2.BufferCore::lookupTransform: Can't find transform from frame_a to frame_b, error: 'frame_b' passed to lookupTransform argument target_frame does not exist

- **ID:** `ros2/tf2-buffer-core-missing-transform-timeout`
- **Domain:** ros2
- **Category:** runtime_error
- **Verification:** ai_generated
- **Fix Rate:** 92%

## Root Cause

The target frame 'frame_b' has not been published to the tf2 tree, typically because the corresponding transform broadcaster node (e.g., robot_state_publisher or static_transform_publisher) is not running, not publishing, or the frame name is misspelled.

## Version Compatibility

| Version | Status | Introduced | Deprecated |
|---------|--------|------------|------------|
| Humble | active | — | — |
| Iron | active | — | — |
| Galactic | active | — | — |

## Workarounds

1. **Verify all frames are published: run 'ros2 run tf2_tools view_frames' or 'ros2 tf2_echo frame_a frame_b'. Check the output for missing frames. If frame_b is missing, add a static transform publisher: ros2 run tf2_ros static_transform_publisher 0 0 0 0 0 0 frame_a frame_b** (90% success)
   ```
   Verify all frames are published: run 'ros2 run tf2_tools view_frames' or 'ros2 tf2_echo frame_a frame_b'. Check the output for missing frames. If frame_b is missing, add a static transform publisher: ros2 run tf2_ros static_transform_publisher 0 0 0 0 0 0 frame_a frame_b
   ```
2. **Ensure robot_state_publisher or your custom broadcaster node is running and publishing transforms. Check with 'ros2 topic echo /tf_static' and 'ros2 topic echo /tf'. If no messages, verify the node's URDF or configuration.** (85% success)
   ```
   Ensure robot_state_publisher or your custom broadcaster node is running and publishing transforms. Check with 'ros2 topic echo /tf_static' and 'ros2 topic echo /tf'. If no messages, verify the node's URDF or configuration.
   ```
3. **Add a timeout and retry loop in your code: try: transform = tf_buffer.lookup_transform('frame_a', 'frame_b', rclpy.time.Time(), timeout=rclpy.duration.Duration(seconds=1.0)) except tf2.LookupException: # handle missing transform** (80% success)
   ```
   Add a timeout and retry loop in your code: try: transform = tf_buffer.lookup_transform('frame_a', 'frame_b', rclpy.time.Time(), timeout=rclpy.duration.Duration(seconds=1.0)) except tf2.LookupException: # handle missing transform
   ```

## Dead Ends

- **** — use_sim_time only synchronizes clock, not transforms. The simulation must explicitly publish transforms via robot_state_publisher or similar nodes. (80% fail)
- **** — Buffer size only affects how long transforms are stored, not whether they are published. If the frame is never broadcast, no buffer size will help. (95% fail)