Robotic arm, table, and camera represented in 3D visualization
College of Engineering Unit: 
Electrical Engineering and Computer Science
Project Team Member(s): 
Keegan Nave, Garrett Scott, Cyrus Swihart and Adam Kerr
Physical Location at Expo: 
Kelley Engineering Center
Project ID: 
Project Description: 

Our project partner is working on research for robotic manipulation and the benchmarking of grasping tasks. To accomplish this, vast amounts of real-world data need to be collected. Performing these experiments requires the robotic arm and camera to be calibrated to the environment.​ Prior to approaching the CS Capstone program, our partner was relying on unrefined command-line tools to achieve workspace calibration. Visualization and simulation were done separately from the calibration process without a unified system. With more research being done, our project partner wished to partner with us to create a more fleshed-out software system to accomplish these tasks. ​

Our Calibration Process

  1. Launch GUI calibration interface and select desired settings.​

  2. Calibrate the main camera using the GUI walkthrough, using OpenCV and ROS interface to find camera parameters.​

  3. Drive robot arm to touchpoint locations and record joint angles using the GUI. (Location of the arm is calculated using IK and ROS relative to the table frame). ​

  4. Choose to save calibration information or visualize the calibrated system in Rviz.