Pick-and-Place System using 4DOF Manipulator
Using FK, IK and computer vision to sort objects based on color
Project Overview
The objective of this project was to understand the forward and inverse kinematics of a serial manipulator, learn to execute trajectories in joint and task spaces, and grasp various aspects of computer vision processing. Our task involved integrating these techniques into a functional pick-and-place system. In this system, colored objects placed in the robot's workspace would be sorted into different areas based on their color, with the robot determining and executing the necessary trajectories.
Technical Implementation
For this project, we were provided with:
- A 4-degrees-of-freedom robotic arm
- A camera
- A vision calibration board within the robot's workspace
We employed MATLAB and its Computer Vision Toolbox to complete the project. Our robot successfully manipulated and sorted the provided objects across multiple trials.
Future Directions
In the future, I aim to:
- Apply similar techniques using Python and ROS
- Explore further optimizations in path planning and object manipulation, particularly in the context of this project's pick-and-place sorter scenario
This project was completed for WPI's RBE3001: Unified Robotics III Course.