Skip to content

In this project an algorithm was developed to autonomously detect 3D weld seams on workpiece of any geometrical shape using Point Cloud Library. Further we used Neural Networks to trace the detected weld seam using a 5 DOF Robotic Manipulator. Other projects like autonomous_grasping using feature matching are also included.

Notifications You must be signed in to change notification settings

PatilVrush/3D_Weld_Seam_Detection_and_Tracking_using_PCL_ROS_ANN

Repository files navigation

3D_Weld_Seam_Detection_and_Tracking_using_PCL_ROS_ANN

In this project an algorithm was developed to autonomously detect 3D weld seams on workpiece of any geometrical shape using Point Cloud Library. Further we used Neural Networks to trace the detected weld seam using a 5 DOF Robotic Manipulator. Other projects like autonomous_grasping using feature matching are also included.

Youtube Link: https://youtu.be/B7zc4auMrHQ

The main code for weld seam detection is: weld_seam.cpp, weld seam tracing: weld_nn.m Research paper: Research_Paper.pdf

Harware Setup FlowChart Result

About

In this project an algorithm was developed to autonomously detect 3D weld seams on workpiece of any geometrical shape using Point Cloud Library. Further we used Neural Networks to trace the detected weld seam using a 5 DOF Robotic Manipulator. Other projects like autonomous_grasping using feature matching are also included.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published