Overview

This page provides an overview of Zach Turner’s contribution to the AESD Final Project Human Recognition using the Nvidia Jetson and Arduino BLE 33 Sense.

Video Outline

See Highlight Video

The video demonstrates:

• cross-compiling application from host machine to target board
• Serial communication between Nvidia Jetson and Arduino
• Logging serial messages from human recognition on Nvidia Jetson

Challenges

My most difficult challenge when implementing this project were:

• Creating the Yocto image to flash for the Jetson Nano
  • Difficulties with VM resources and virtual disk sizes
• Getting both boards to communicate on a pre-defined API
  • Debugging communication between two boards, only connected to one board
  • passing bytes between both boards at a given symbol rate and validating the message

Lessons Learned

The most important topics I learned from this project were:

• Serial communication using linux system calls are as simple as file read/writes because devices are files in the linux filesystem.
• Configuring a serial device from an appplication using the termio API.
• Creating a server/client communication ICD and implementing the read/write between subsystems to facilitate transitioning between states.