Telemetry Research [MSXV]
- Vinai Birbal (Deactivated)
- Forest Zhou
Telemetry has been deleted due to lack of available resources. Data will be logged locally for EOD analysis.
Current System
The current hardware for the telemetry system for MS14 is based around a Raspberry Pi 3+ with a SIM7600X 4G/GPS Hat for GPS and LTE connectivity. It collects data from the CAN bus and GPS module and transmits this data to the chase car through an external Wi-Fi antenna and using an LTE antenna to a cloud server. Both the GPS antenna and the LTE antenna are connected to the SIM7600X 4G/GPS Hat for the Raspberry Pi 3+ as shown below.
The current system is described in the confluence pages below:
I also found this page on the telemetry system for MS12 helpful in getting an idea of the requirements:
Possible Improvements
· Design custom antenna – This was suggested initially but was deemed too difficult/not worth the effort. The current idea is to use off-the-self antennas. It would be cheaper if we can reuse the ones that were bought for MS14 but new ones could also be looked at.
· Replace Raspberry Pi 3+ with custom MCU board - The Raspberry Pi 3+ is relatively power hungry (around 2.5 W for normal operations). A custom MCU board should be less than 1W so it should reduce power consumption.
The SIM7600X 4G/GPS Hat is compatible with STM32 host boards so this should be doable. Additionally, getting to reuse the SIM7600X 4G/GPS Hat and GPS and LTE antennas would save money. However, networking with a custom MCU board might prove challenging.
· Off-the-shelf solutions – There are several off-the-shelf solutions that should meet our requirements. In general, these would be more expensive but would need less work. Below is a list of possible options that I found but more research will need to be done if we decide to go with this route:
o OriginIoT™ cellular IoT system:
From the OEM: “The OriginIoT™ is a miniature cellular IoT system that enables the speedy development of IoT products without writing embedded code and without knowledge of RF engineering. … The multi-purpose OriginIoT™ system can accommodate peripheral devices such as sensors or other components through UART, SPI, I2C, or GPIO and combines cellular communications module according to customer choice, with superior positional accuracy of standalone GNSS. Peripheral devices are configured over a Web interface, eliminating additional embedded FW efforts.”
https://origingps.com/wp-content/uploads/2022/01/OriginIoT%E2%84%A2-System-Datasheet-Rev-2.5.pdf
Couldn’t find information on the cost/if it is available for single order on website.
o GPy with Pysense 2.0 X :
The GPy is “the only LTE CAT M1 / NB1 and MicroPython combo micro controller on the market today …. the GPy offers WiFI, BLE and cellular CAT-M1/NB”.
https://pycom.io/product/gpy-multipack/
This would cost around $150 including the GNSS expansion board and all needed antennas. Power consumption should be less than the Pi.
o Notecard:
From the OEM: “The Notecard is a device-to-cloud data pump that reduces the complexity of building connected solutions with secure, reliable cellular. As an embeddable system on module (SoM), you can use the Notecard with any microcontroller for both greenfield and retrofit projects using your own design, or one of our custom-designed Notecarriers.”
The Notecard with the Notecarrier-AF with onboard Cellular and GPS antennas and Host MCU development board would cost about $130:https://shop.blues.io/collections/swan/products/notecarrier-swan .
We could also just buy the Notecard and/or Notecarrier and design our own MUC host board : https://shop.blues.io/collections/notecarrier/products/carr-ae
Overall, this seems like it might be slight cheaper than the other off-the-shelf options mentioned above. Also there seems to be good documentation: GitHub - blues/note-hardware
Notecard Datasheet - NOTE-WBEX-500 - Blues Developers
· Custom Board:
This would involve designing our own custom board around something like the SARA-R510M8S module for LTE and GNSS: https://www.mouser.ca/pdfDocs/SARA-R5_DataSheet_UBX-19016638.pdf .
This would be the most difficult option and involve some RF design for the antenna parts. Also it would not necessarily be that much cheaper than the off-the -self solutions and certainly more expensive than just replacing the Raspberry Pi 3+.
Conclusion
Considering the options mentioned above I think that the easiest and cheapest improvement would be to just replace the Raspberry Pi 3+ with a custom MCU board. A new off-shelf-solution or a custom board would also be potentially viable but these would come with increased cost and, in the case of the custom board, increased difficulty. These options both also offer potentially better performance and more features. Finally, on the antenna, I think that it would be nice if we could use smaller/onboard antennas like those found in some of the off-the-self solutions above instead of the large antennas that we currently use for LTE and GPS.