robot_localization not fusing imu data

Question

According to the documentation in : http://docs.ros.org/en/noetic/api/robot_localization/html/state_estimation_nodes.html I was able to transform the imu data header fram from "imu_link" into "base_link" through a tf transformation in a node.

However, somehow my imu data is not being fused into the ekf_localization node. I mean I was able to see the yaw changes as I move my robot. But when I subscribe to the odometry/filtered, it still shows my yaw to be at 0 degree.

I have double checked with the raw data from the imu and verified it outputs data as shown below: The data below are my imu message after I move the robot. This which reflects the changes of quaternion in the orientation

---
header: 
  seq: 9520
  stamp: 
    secs: 1617336774
    nsecs:  71053028
  frame_id: "base_link"
orientation: 
  x: -873.0
  y: -496.0
  z: 0.0
  w: 16353.0
orientation_covariance: [-1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0]
angular_velocity: 
  x: 0.0
  y: 0.0
  z: 0.0
angular_velocity_covariance: [-1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0]
linear_acceleration: 
  x: 0.0
  y: 0.01
  z: -0.18
linear_acceleration_covariance: [-1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0]
---

Below is my configure file for the ekf kalman filter.

<include file="$(find roboteq_driver)/launch/controller.launch"/>
<include file="bno055.launch"/>
<node pkg="robot_localization" type="ekf_localization_node"
    name="robot_localization_ekf_node_continuous"
    clear_params="true">
    <param name="frequency" value="30" />   
    <param name="sensor_timeout" value="0.1"/>
    <param name="two_d_mode" value="true" />
    <param name="publish_tf" value="true" />
    <param name="odom_frame" value="odom" /> <!-- publish a new data to odom-->
    <param name="base_link_frame" value="base_link" />
    <param name="world_frame" value="odom" /> <!-- change here if we want to add map frame later-->
    <param name="odom0" value="/drive_odometry" />
    <rosparam param="odom0_config">
        [false,false, false, 
        false, false, false,
        true, true,false,
        false, false, false,
        false, false, false]</rosparam>
    <param name="odom0_differential" value="false"/>
   <param name="odom0_relative" value="false"/>
    <param name="odom0_nodelay" value="true"/>

    <param name="imu0" value="transformed_imu"/> 
    <rosparam param="imu0_config">
        [false,  false, false,
        false, false,false,
        false, false, false,
        false , false, true,
        true, false, false]  <!-- from IMU message acceleration-->  
    </rosparam>
    <param name="imu0_differential" value="false"/>
    <param name="imu0_relative" value="false"/>
    <param name="imu0_nodelay" value="false"/>
    <param name="imu0_remove_gravitational_acceleration" value="true"/>
    <param name="imu0_queue_size" value="10"/>

    <param name="use_control" value="false"/>
    <!--final odom message-->
    <remap from="/odometry/filtered" to="/odometry/filtered_odom"/>
    
</node>

Odometry message from : drive_odometry

header: 
  seq: 1392
  stamp: 
    secs: 1617336632
    nsecs: 897039890
  frame_id: "odom"
child_frame_id: "base_link"
pose: 
  pose: 
    position: 
      x: 10.4
      y: 4.43
      z: 0.0
    orientation: 
     x: -873.0
     y: -500.0
      z: 0.0
      w: 1240.0
  covariance: [0.5, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.5, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 1e-08, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1]
twist: 
  twist: 
    linear: 
      x: 0.0
      y: 0.0
      z: 0.0
    angular: 
      x: 0.0
      y: 0.0
      z: 0.0
  covariance: [0.5, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.5, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.4]
---

So In this case, according to the kalman filter in the library, since I only use the yaw velocity from IMU in the configure file to determine the current yaw position of the robot, I am expecting the yaw value output from odometry/filtered_odom to be about the same as the yaw value in the IMU message.

However, when I subscribed to the odometry/filtered_odom, I find the orientation to be: x: 0.0 y: 0.0 z: 0.0 w: 1.0

Note: I only allow the angular velocity of IMU to be fused with the filter in determining the yaw direction. I did not use the yaw message from the odom, because the Odom message is generated by my calculation based on the feedback from the wheel's encoder.