Flight Companion Computer (FCPC) Development & Fork

The Flight Companion Computer, FCPC, which at this stage is basically a Raspberry pi CM4 with a HAT, was intended as system for both handling user interfacing & experience, and handling systems data, for an Arrow single seat ultralight.

It’s current functions are:
Read and record Battery, Motor, Flight Controller and Joystick data.
Create data log files.
Host displays with a pilot user interface.
Send telemetry data over a wireless network.

Currently, especially hardware wise, the system is a bit of a hodgepodge with functionality spread across multiple circuit boards and several undue operational limitation such as range. As such to be both more robust and cost effective production wise several improvements could be made:

A PT4 and Onwards FCPC

However, the long slog of actually making the damn thing to do its job and play nice with the Electronic Speed Controllers (ESCs), Battery Managment Systems (BMSs), as well as interacting with the Embention Team on integrating their Flight Controller system, …and in general… , as well as the difficulties even they faced with ESC interactions, flagged up the difficulty of multi system, multi protocal interfacing.

Each one basically has it own logic to grok which requires its own unique expertise. Time, effort annoyance, and This was only dealing with 4 or 5 systems !

Imagine having to deal with potentially dozens of unique elements all speaking a different language.

This led to thoughs of “What if there was a way to bridge this process ?” “What else could such a system be used for ?” “How much value could it add ?”

There are multiple paths to this but cue:

RoboDaqc: General Purpose Data Acquisition & Control (an FCPC Fork) WIP

1 Like

I am not sure if I fully understand. Is this a suggestion on how to make the companion computer more robust? Why consider it a fork of the FCPC and not just an upgrade for PT4? Additionally, do you think you will be running into many of the same issues as the first go around when changing to the Nvidia Jetson NX?

There are 2 bridging off points described here.

The first, in the PT4 FCPC, presentation is a direct continuation of the PT3 system.

The second, RoboDAQC is a concept for a more powerful potential extension to a stand alone general purpose system / product for robotics aviation industry etc.

The idea is that the issues with integration we faced with the escs, battery fc etc are general industry wide issues faced in aviation or robotics or biotech etc when integrating various sensors or systems. So if there was some device which could help expidite the process then it would be very useful indeed

Key Project Milestones and Deliverables for a Proposed PT4 FCPC

  1. Base Board ~ $ 1200
    RPI CM4 breakout Board with Harness and Peripheral Connectors and plugin modules for functionality

  2. CAN Module ~ $ 800
    Waveshare hat functional replicated on a plugin module

  3. RS-485 Module ~ $ 800
    RS-485 transceiver implemented on a plugin module

  4. Custom ISO ~ $ 250
    ISO with dependencies and prerequisites preinstalled and able to be loaded onto the RPI via an SD card

  5. Network Architecture ~ $ 1200
    Retool current networking functionality with tbd,e.g google, streamr , pub-sub sdk

  6. New UI ~ $ 900
    Retool current UI with PyQt

  7. Otto Implementation ~ $ 250
    Test otto joyticks

Document

1 Like