Project information

• Category: Research Project
• Location: Champaign, IL
• Project date: Dec 2024
• Project URL: View Report

The development of an autonomous drawing system that utilizes a UR3 robotic arm integrated with ROS to transform digital images into high-fidelity physical sketches. Leveraging advanced image processing techniques and real-time robot control, the project focused on precision path planning, seamless hardware-software integration, and adaptive algorithms. The system was rigorously tested to ensure accurate execution of intricate artistic renderings, demonstrating its reliability and technical robustness.

• Designed and implemented an end-to-end pipeline for converting digital images into physical drawings using Python, OpenCV, and ROS, featuring grayscale conversion, Gaussian blurring, edge detection, and contour extraction.
• Engineered custom robotic control algorithms using inverse kinematics for precise (x, y, z) mapping, ensuring accurate movement sequences and real-time error handling within ROS.
• Optimized keypoint extraction through adaptive preprocessing and skeletonization techniques, achieving smooth and detailed path trajectories for the robotic arm.
• Developed and validated visualization tools with Matplotlib and OpenCV for user verification of image processing outputs prior to robotic execution.
• Performed extensive testing on complex images to fine-tune system performance, ensuring high fidelity in reproducing intricate sketches with minimal path deviations. Produced a research report of the findings.