2019-01-22, 01:20 AM
I strongly suggest you attach the mpu9250 to a raspberry not an esp8266. But please feel free to experiment of course.
The reason is, the raspberry has the cpu power and algorithms already implemented to automatically calibrate and manage the mpu9250 efficiently.
If you were to say send the raw sensor data over wifi from the esp8266 it would incur lag which is a very bad thing. I definitely do not want to read the gyro, send it over wifi, then have the autopilot react to that. it would degrade performance. In fact, it is setup now, so that pypilot directly reads the mpu9250 in a real-time priority process and drives the motor directly with a latency less than 0.1 seconds, and distributes the data on a low priority process to other tools where latency is not an issue.
So depending on your use case, and what you are trying to accomplish it might make sense. I hope this is clear and please continue the discussion.
The reason is, the raspberry has the cpu power and algorithms already implemented to automatically calibrate and manage the mpu9250 efficiently.
If you were to say send the raw sensor data over wifi from the esp8266 it would incur lag which is a very bad thing. I definitely do not want to read the gyro, send it over wifi, then have the autopilot react to that. it would degrade performance. In fact, it is setup now, so that pypilot directly reads the mpu9250 in a real-time priority process and drives the motor directly with a latency less than 0.1 seconds, and distributes the data on a low priority process to other tools where latency is not an issue.
So depending on your use case, and what you are trying to accomplish it might make sense. I hope this is clear and please continue the discussion.