GoPiGo Micro-controller Capable of Computer Vision processing?


#1

I am intending to create a crude autonomous robot using the GoPiGo Robot Starter Kit and a Raspberry Pi Camera. I intend to code ROS packages in C++ and OpenCV to perform crude computer vision and autonomously navigate the robot.

Are these things possible or will I run into problems or limitations? Any insight from you guys would be greatly appreciated :slight_smile:

Do you think the following requirements are met by the GoPiGo Robot Starter Kit and a Raspberry Pi Camera?

  • Install and run ROS on the GoPiGo Raspberry Pi micro-controller
  • Perform computer vision video processing with not too high latency. Do you think the micro-controller can handle a ROS package/packages that will be processing the video stream (goal is to process/handle ~25 frames/second).
  • API to move the GoPiGo wheels, turn etc. How is this setup? Is there an API I can use?

#2

Iโ€™m not familiar with ROS but I do know OpenCV can be installed on Raspbian for Robots although you will need a 16 Gig card. And OpenCV installation takes a while.

You can find a write-up here on PythonProgramming.net


#3

This topic was automatically closed after 2 days. New replies are no longer allowed.