Learn how to set up Visual Inertial Odometry (VIO) on a Pixhawk-based drone using a Jetson Nano and Intel T265 Realsense camera for indoor vision positioning. Follow this comprehensive 29-minute tutorial that covers hardware assembly, software installation, and configuration. Build a MAVLink cable, mount the Jetson Nano and T265 camera, install necessary software including ROS Melodic and dependencies, and configure QGroundControl for wireless communication. Explore PX4 EKF2 parameters for VIO and witness a live flight test demonstrating VIO altitude and position modes. Access provided resources, including a simplified software installation guide, 3D printing files for mounts, and links to relevant GitHub repositories.
PX4 VIO Setup with Jetson Nano and Intel T265 - Auterion VIO Fork
Overview
Syllabus
Intro
Quick VIO Position Hold Demo
Building the MAVLink Cable
Building the MAVLink Cable
Top Down View of VIO Drone
Powering Jetson Nano from Drone
3D Printed Jetson Nano Mounts
3D Printed T265 Mount
Burning Jetson Nano Image
Auterion VIO Code
VIO Gist Shortcut
ROS Melodic and Dependency Installation
Install librealsense
Install Python Catkin Tools
QGroundControl Wireless Configuration
Verify MAVLink Data Comms
Review MAVROS Launch Files
PX4 EKF2 Params for VIO
Launching VIO
Begin VIO Flight Test
VIO Altitude Flight Mode
VIO Position Flight Mode
VIO Yawing
Conclusion
Taught by
Dennis Baldwin
