New IMU BerryGPS-IMUv4

[cortinaman]

a quick search shows the LSM6DSL.  It should be supported although this is not the usual sensor.   I tested and the gyroscope has a higher noise level than icm20948/mpu9255 so I would prefer that sensor if possible instead.

What does i2cdetect -y 1 show?

     0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f
00:          -- -- -- -- -- -- -- -- -- -- -- -- --
10: -- -- -- -- -- -- -- -- -- -- -- -- 1c -- -- --
20: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
30: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
40: -- -- 42 -- -- -- -- -- -- -- -- -- -- -- -- --
50: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
60: -- -- -- -- -- -- -- -- -- -- 6a -- -- -- -- --
70: -- -- -- -- -- -- -- 77

Where from BerryGPS-IMU v4 documentation:
0x1c = LSM6DSL accelerometer (not detectable from pypilot, even after procedure above and reinstalling pypilot)
0x1e = LIS3MDL magnetometer (not detected by i2cdetect above)
0x42 = GPS chip (this is a whole other kettle of fish)
0x6a = LSM6DSL gyro (not detectable from pypilot, even after procedure above and reinstalling pypilot)
0x77 = BM388 pressure/temperature sensor (these are visible in i2c app and connected to signalk)

Which version of rtimulib2 is installed?    It should detect it.    Can you stop all pypilot processes and services and perhap install rtimulib2 from source and run the basic drive demo?

It could be something very simple like the sensors set to the secondary address or something like that.   It should be supported.

try single stepping RTIMUSettings::discoverIMU  in gdb
especially line 213:

           } else if (result == LSM6DS33_ID || result == LSM6DSL_ID || result == ISM330DHCX_ID) {

and check the value of "result"

Sorry, my mistake.
I was mistaken about the i2c adresses, 0x1e should not be there. Correct addresses are:

• 0x6A for the gyroscope and accelerometer
  
  
• 0x1C for the magnetometer
  
  
• 0x77 for the pressure sensor
  
  
• 0x42 for GPS
  

I was more worried about pypilot being broken, so I started from scratch again.
This time I checked on pypilot after each step, after step 14 pypilot detected the imu and magnetometer so I stopped.

1. sudo apt-get update
   the above may fail, still continue
   
2. sudo apt-get install libqt4-dev cmake -y
   
3. git clone https://github.com/seandepagnier/RTIMULib2.git
   
4. git clone https://github.com/pypilot/pypilot.git
   
5. cd RTIMULib2/Linux/
   
6. mkdir build
   
7. cd build/
   
8. cmake ..
   
9. make -j4
   
10. sudo make install
    
11. sudo ldconfig
    
12. cd ../python/
    
13. python3 setup.py build
    
14. sudo python3 setup.py install
    

I.e. I did not perform the final two steps.

1. cd ~/pypilot/
   
2. sudo python3 setup.py install
   run the above command twice
   

Pypilot runs and data are sent to signalk. Same with temp. and press from I2C app.

Maybe it's time to clone the sd card before I continue with gps and PICAN-M.