03-13-2023 12:29 PM
I used the development board to sample data in a clean environment and in a jar with some cheese inside. After that i successfully created the ai algorithm and I managed to successfully load the configuration into the bsec_config_state.ino. The board with 8 sensors and the flashed code is then able to give me the percentages right.
After that i loaded the configuration into my custom code for the Raspberry Pi Pico using the bsec2.2.0.0 library, but when asking for the gas estimates i always obtain gas_estimate_1 at 100%, no matter in which environment i put the sensor.
I followed the workflow of the integration guide:
The sample rate is set to scan, is there anyone that already faced this kind of issue even if on other devices?
Solved! Go to Solution.
03-24-2023 08:45 AM
Hi Prop4et,
The versions of the BME dev kit software, BME AI-Studio, BSEC sftware must match, or there will be a version mismatch error. Also, what version of BME AI-Sudio is your config generated with?
03-24-2023 02:08 PM
I don't get any version mismatch error on this because i know that the generated configuration with BME AI Studio must match the library version used in the implementation. So i generate a BSEC2.2.0.0 configuration and i use the BSEC2.2.0.0 library. After what you told me i switched to BSEC2.4.0.0 and i generated the BSEC2.4.0.0 configuration from the same algorithm.
I am using BME AI Studio v2.0.0.
To reinforce what is my problem:
I have a single BME688 sensor on an Adafruit board connected to a Raspberry Pi Pico with I2C that runs the BSEC2.2.0.0 (and after your advice i switched to 2.4.0.0 making all the required adjustments). I take samples with the devkit and generate a configuration with BME AI Studio v2.0.0. for the selected bsec library. I then load the .c file (i also tried hardcoding the variable to avoid loading problems) and pass it to the bsec_set_configuration and check that the result of the operation is ok. Then when i go to the sensing part the result is stuck on 100% for the GAS_ESTIMATE_1 value even if i change the condition.
Loading the same (always the right version) .c file in the .ino example for the devkit board gives me the right results when doing the sensing.
The workflow of the code that i've written for the Raspberry Pi Pico is the one present in a previous answer on this topic
03-29-2023 10:46 AM
A short follow up on the topic.
There was a bug in my code that didn't allow for the get data to retrieve the right values, and the estimeates weren't changing because of that.
After having fixed that i'm having the same problem exposed in this post.
I tried running the basic_config_state.ino on the devkit and with SENS_NUM = 0 the results are right
If i change SENS_NUM to 5 you can see how the sensor needs a couple of readings to get to the correct result
With my configuration running with an adafruit board with the sensor installed on it and a raspberry pi pico this doesn't even happen, the readings are wrong, even if i switch from bsec2.2.0.0 to bsec 2.4.0.0
Did anybody face the same issue and solved it even if with another board?
05-05-2023 03:57 PM
Hi Prop4et,
BME688 dev kit used 8 BME688 sensors, did you loaded the configuration into your custom code for the Raspberry Pi Pico with 1 BME688 sensor?
05-09-2023 11:33 AM
Hi Prop4et,
We tested the config file you used, which can run normally with BSEC 2.4.0.0.
If you used BSEC 2.2.0.0 with your config file, which would report some error codes.