Lego Stacker

Class: Intro to Robotic

Semester: 2019 Fall

Programming language: Python

Software: ROS environment

Hardware: Sawyer arm from Rethink Robotics

Team member: Jiabin Lin, Boljot Kaur, Angus MacDonald, Muyu Deng

Demo Video Link: https://www.youtube.com/watch?v=NB3cVBDb964&feature=youtu.be

Objective:

The goal of this project is to control the Sawyer arm to stack the legos.

Introduction:

We operate Sawyer arm in the ROS environment, using topic in order to read the arm's coordinates, which provides us with significant information for navigating our target position and determine the force for stacking the lego.

Figure 1: Image taken from Sawyer arm (Black pieces are lego position)

Figure 2: Image Processing result

We also implemented robotic vision by using the camera from the sawyer's arm in order to locate the lego's position and target position. We move the sawyer right hand camera to a specific location in order to get a clear view of the legos. Since in real life, the picture has too much noise, we blur the picture to get rid of the noise and some white reflection, then pass the image to the darkness filter so that we can filter out lighter color. Find the blobs by expanding on the output of the filter, and then find the centroid of the blobs in term of the picture.

However, due to time limitation, we are unable to integrate the vision and sawyer arm movement code together. The video above is the result of hardcoding lego position.