Skip to content

Code base for off-board vision processing on a Raspberry Pi.

Notifications You must be signed in to change notification settings

NicholsSchool/2019-RaspberryPi

Repository files navigation

2019-RaspberryPi

FRC Team 4930's codebase for off-board vision processing on a Raspberry Pi for 2019 Deep Space, based off of the Java demo included in FRC's Raspberry Pi image. We use OpenCV to determine the robot's pose relative to the retroreflective tape on the cargo ship and the rocket.

vision demo

Cool Features

  • Image contour analysis
  • Sub-pixel corner detection
  • 6D pose esimation with 3D-2D point correspondence
  • Camera offset correction
  • Headings and distances to multiple custom waypoints

Building & Deploying

Java 11 is required to build. Set your path and/or JAVA_HOME environment variable appropriately.

  1. Run "./gradlew build"

On the rPi web dashboard:

  1. Make the rPi writable by selecting the "Writable" tab
  2. In the rPi web dashboard Application tab, select the "Uploaded Java jar" option for Application
  3. Click "Browse..." and select the "2019-RaspberryPi-all.jar" file in your desktop project directory in the build/libs subdirectory
  4. Click Save

The application will be automatically started. Console output can be seen by enabling console output in the Vision Status tab.

About

Code base for off-board vision processing on a Raspberry Pi.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published