GRIP Raspberry Pi 3

GRIP generated code to track the 2016 FRC tower on a Raspberry Pi using an ip camera

##Set up the raspberry pi

There are two options

1. Follow [this] (http://www.pyimagesearch.com/2016/04/18/install-guide-raspberry-pi-3-raspbian-jessie-opencv-3/) tutorial

• Skip everything inside step 4 related to virtual environments

• Change openCV version from 3.1.0 to whatever the latest version is (Currently using 3.2.0)

• Install pynetworktables

  $ pip3 install pynetworktables

2. Download and flash the image on GitHub

##Generating and editing your code

1. Write your GRIP pipeline to process the image
2. Export the code in Python

• tools -> Generate Code

3. Edit the exported code

• Change class name to "GripPipeline"

  class GripPipeline:

• Add a second parameter "source0" to the process function

  def process(self, source0):

• Remove "self.__" from input of first step in the pipeline in process function (in this case resize)

  #Step Resize_Image0:
  self.__resize_image_input = source0

##Modifying sample_vision.py

1. Make sure the import statement is correct for your generated pipeline

from "PIPELINE_FILE_NAME" import "CLASS_NAME"

2. Change NetworkTables IP address to that of the roboRIO

   NetworkTable.setIPAddress("10.TE.AM.XX") 

3. Pick video input method

• If using an IP camera change address of the camera

  cap = cv2.VideoCapture("http://10.TE.AM.3/mjpg/video.mjpg")

##References

1. https://github.com/WPIRoboticsProjects/GRIP-code-generation/tree/master/python
2. https://github.com/Frc2481/paul-bunyan/blob/master/Camera/main.py
3. https://github.com/robotpy/pynetworktables
4. https://github.com/WPIRoboticsProjects/GRIP/wiki/Running-GRIP-on-a-Raspberry-Pi-2