furbrain/SAP5 Axes Configuration question

[gerowtj]

@furbrain

I am studying your code for the SAP and have some really basic questions. I've started by trying to understand the raw data that is read from the BM1422 and LSM6DS3 sensors and how the sensor orientation and polarity is manipulated during calibration.

The sensors.c file is where I am currently looking.
Can you describe the purpose of these numbers (3,4,5,4,0,2)? in static const uint8_t V1_1_AXES[] = {3, 4, 5, 4, 0, 2};
I'm assuming they are used to determine how the actual sensors are mounted on the PCB and oriented to the World View.

I see that the V1_1_AXES values are placed in the config.axes.accel and config.axes.mag data structures and would look like:

config.axes.accel[0] = 3
config.axes.accel[1] = 4
config.axes.accel[2] = 5
config.axes.mag[0] = 4
config.axes.mag[1] = 0
config.axes.mag[2] = 2

The values are then used to determine sign and axis in the void sensors_raw_adjust_axes function.

sign = config.axes.accel[i]>=3 ? -1 : 1;
axis = config.axes.accel[i] % 3;

The result of the code would be:

config.axes.accel[0] = 3    Sign = -1   Axes = 0
config.axes.accel[1] = 4    Sign = -1   Axes = 1
config.axes.accel[2] = 5    Sign = -1   Axes = 2
config.axes.mag[0] = 4    Sign = -1   Axes = 1
config.axes.mag[1] = 0    Sign = 1   Axes = 0
config.axes.mag[2] = 2    Sign = 1   Axes = 2

So, the sensor data is read via I2C communication in the order X, Y, Z into the raw sensor structure.
The config.axes data is used to modify the polarity of the data and the order the data is stored in the sensors data structure.
The raw data that was read from the sensors is then multiplied by the sign for the "axes" number.

It looks like the accel and gyro axis raw data are read in the same order that they are needed in the sensors.accel and sensors.gyro sturcture.
However, the mag raw sensor data looks like it swaps the X and Y raw data so that it is stored in the sensors.mag structure as Y, X, Z

As for the sign, it looks like the accel and gyro values are all flipped, ie multiplied by -1. The mag Y is flipped but the X and Z raw values are left alone (multiple by 1).

I attached a pdf with some notes trying to understand the sensor chip orientation versus the units/laser pointer orientation.
Any guidance would be appreciated.

SAP Oriention of Chips and Axis.pdf

[furbrain]

Looks like you've understood this all correctly. The only possible issue is that I have my Y-axis (in device coordinates) as going to the right with the laser pointing away from you. What I'm actually wanting is the direction of gravity (which is opposite to the direction of acceleration), hence all the accelerometer values are flipped. Code is a bit ambiguous on this point - apologies.

Phil

[gerowtj]

Hi Phil,
Thanks for the quick reply.
One more question, can you describe the orientation of the magnometer (bm1422) and accel/gyro (lsm6ds3) sensors?
Unfortunately i can't open any of the hardware files to see the pcb design.
How are the sensors mounted on the pcb and then whats the orientation of the pcb in relation to the laser?

Thanks
Tom

[furbrain]

The PCB files are oriented with the display up and the laser printing away from you. Both sensor chips are oriented with the pin 1 mark at top left, and are mounted on the upper surface of the board. I can send you pdfs of the board layout if you want. Phil

…

On Tue, 15 Mar 2022, 23:53 gerowtj, ***@***.***> wrote: Hi Phil, Thanks for the quick reply. One more question, can you describe the orientation of the magnometer (bm1422) and accel/gyro (lsm6ds3) sensors? Unfortunately i can't open any of the hardware files to see the pcb design. How are the sensors mounted on the pcb and then whats the orientation of the pcb in relation to the laser? Thanks Tom — Reply to this email directly, view it on GitHub <#3 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAW3PXPBHPFULNZ6G2QI2NDVAEPJFANCNFSM5QTSLEDA> . Triage notifications on the go with GitHub Mobile for iOS <https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675> or Android <https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub>. You are receiving this because you were mentioned.Message ID: ***@***.***>

[gerowtj]

A pdf would be great. I'll send you a description of what I'm working on when i get home after 10pm ( US Eastern Standard Time) Thanks Tom On Tue, Mar 15, 2022, 8:11 PM Phil Underwood ***@***.***> wrote:

…

The PCB files are oriented with the display up and the laser printing away from you. Both sensor chips are oriented with the pin 1 mark at top left, and are mounted on the upper surface of the board. I can send you pdfs of the board layout if you want. Phil On Tue, 15 Mar 2022, 23:53 gerowtj, ***@***.***> wrote: > Hi Phil, > Thanks for the quick reply. > One more question, can you describe the orientation of the magnometer > (bm1422) and accel/gyro (lsm6ds3) sensors? > Unfortunately i can't open any of the hardware files to see the pcb design. > How are the sensors mounted on the pcb and then whats the orientation of > the pcb in relation to the laser? > > Thanks > Tom > > — > Reply to this email directly, view it on GitHub > <#3 (reply in thread) >, > or unsubscribe > < https://github.com/notifications/unsubscribe-auth/AAW3PXPBHPFULNZ6G2QI2NDVAEPJFANCNFSM5QTSLEDA > > . > Triage notifications on the go with GitHub Mobile for iOS > < https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 > > or Android > < https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub >. > > You are receiving this because you were mentioned.Message ID: > ***@***.***> > — Reply to this email directly, view it on GitHub <#3 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AIY5CW632I4TTN6ZGQZNEN3VAERJLANCNFSM5QTSLEDA> . Triage notifications on the go with GitHub Mobile for iOS <https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675> or Android <https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub>. You are receiving this because you authored the thread.Message ID: ***@***.***>

[gerowtj]

First, I am not a programmer. I can read code, google a bunch, copy and paste, and generally get something to work. I've been playing with Arduino's for a few years doing very simple projects, data logging of temperature, weigh scales, etc. Recently I've been trying to build a very basic laser disto for cave survey. I've built a few prototypes but keep running into issues with IMU stability and getting a good calibration. I've tried the MPU9250 IMU and recently a Sentral IMU. Neither has given good results for me.

I've now moved onto a Seeeduino Xiao controller, and an IMU from Pololu, AltIMU10-v5. This looks a lot more promising once I get the calibration figured out.

The AltIMU10 has the accel, gyro and magnometer all on one board. This is why I'm asking questions about your sensor orientation in order to try an get the AltIMU10 oriented in a similar manner, or modify the software Axes Configuration to reflect the AltIMU10 orientation.

If there is a better place to discuss this project please let me know. Also, if you don't have time for my questions please let me know, I don't want to be a bother. Eventually I'll figure it out and when I get something developed enough I'll post the code in a public repository.

Thanks again for your time,
Tom

[furbrain]

Hi Tom, here is a picture of the pony PCB. The magnetometer is U11 and the accelerometer is U12. Notice how I've put the magnetometer near the edge of the board. This means that there is limited impact from any stray currents on the board. I've also attached a paper I wrote on the calibration process, it's got a few good references for calibration techniques. I'd love to hear how you get on!

…

On Wed, 16 Mar 2022, 04:14 gerowtj, ***@***.***> wrote: First, I am not a programmer. I can read code, google a bunch, copy and paste, and generally get something to work. I've been playing with Arduino's for a few years doing very simple projects, data logging of temperature, weigh scales, etc. Recently I've been trying to build a very basic laser disto for cave survey. I've built a few prototypes but keep running into issues with IMU stability and getting a good calibration. I've tried the MPU9250 IMU and recently a Sentral IMU <https://www.tindie.com/products/onehorse/ultimate-sensor-fusion-solution-lsm6dsm-lis2md/>. Neither has given good results for me. I've now moved onto a Seeeduino <https://www.seeedstudio.com/Seeeduino-XIAO-Arduino-Microcontroller-SAMD21-Cortex-M0+-p-4426.html>Xiao controller, and an IMU from Pololu, AltIMU10-v5 <https://www.pololu.com/product/2739>. This looks a lot more promising once I get the calibration figured out. The AltIMU10 has the accel, gyro and magnometer all on one board. This is why I'm asking questions about your sensor orientation in order to try an get the AltIMU10 oriented in a similar manner, or modify the software Axes Configuration to reflect the AltIMU10 orientation. If there is a better place to discuss this project please let me know. Also, if you don't have time for my questions please let me know, I don't want to be a bother. Eventually I'll figure it out and when I get something developed enough I'll post the code in a public repository. Thanks again for your time, Tom — Reply to this email directly, view it on GitHub <#3 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAW3PXMYDJKQPPL4PIVC4WDVAFNYPANCNFSM5QTSLEDA> . Triage notifications on the go with GitHub Mobile for iOS <https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675> or Android <https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub>. You are receiving this because you were mentioned.Message ID: ***@***.***>