[Bug Report] : `/angular_position` publisher not operating at desired speed

[Legohead259]

Describe the bug The /angular_position publisher is supposed to publish the hall effect sensor position at 100 Hz. In testing with the FOC algorithm and MicroROS, this frequency drops to ~38 Hz.

To Reproduce Steps to reproduce the behavior:

1. Boot board
2. Initialize connection to board through Serial ROS2 agent (baudrate=115200)
3. Execute ros2 topic hz /angular_position

Expected behavior The report from the ros2 topic hz command should be ~100 Hz, not ~38

[Legohead259]

After commit 0bf950c, publisher is still publishing at ~38 Hz. Note that the publisher was originally running at 100 Hz and this frequency does not change after adding/removing services or FOC code. Something else is going on....

[Legohead259]

When all functionality of the callback method is removed, the publish frequency is 100 Hz, as expected. The timer functions properly, there is just delay in the callback. Delay is due to several factors:

• Transactions on the I2C bus take time and the "polling" method of data acquisition lowers sample rate
  • Was able to speed up by removing an erroneous transaction, sensor.getAngleEn() from the callback (~10 ms)
  • Increased bus speed to FAST (400 kHz) (~5 ms)
• Sensor sample rate is slow (~30 ms)
• Sensor by default does 32x supersampling to reduce HF noise (1x ~10 ms)

It is recommended to switch sensor to interrupt mode instead of polling method

• /angular_position topic will be published to when the sensor reports new data ready
  • This will also reduce the timing loads on the CPU
• If position needs to be used, it can be stored in a global variable in memory

For now, expect a maximum sample rate of ~40 Hz. THIS IS NORMAL FUTURE BRAIDAN