- assembled and tested first v5-cam board - defective CAM receiver part
- Use internal antenna for the CAM-M8C GNSS Module
- revise CAM-M8C layout to get a cleaner ground plane
- add 30 ohm resistors on CAM-M8C I2C lines to reduce noise
- revise MPU-6050 PCB layout to more closely match the datasheet
- add header lines for SWD programming, switch to 2.0 mm spacing on header
I decided to revise the SAM variant PCB to use more modern ICs:
- Change GNSS to the u-blox SAM-M10Q (the M10Q is pin compatible with the M8Q)
- Change IMU to TDK InvenSense ICM-42688
- Add hardware interrupt lines from the IMU
- Change barometric sensor to the Bosch BMP-390
- Switch to a USB-C connector
I have resolved the buffering issue in the GNSS reporting stream and continue to work on developing analysis tools using Octave.
I have collected the first log files from several jumps. GPS data is usable, although there are intermittent u-blox buffering errors interspersed in these logs. I have been exploring the feasibility of pose reconstruction from IMU log data. I am using GNU Octave for this investigation. The code resides in the ./analysis subdirectory. It should be considered entirely experimental at this stage - don't trust the accuracy of any of that code based on where I am today.