About this Abstract |
| Meeting |
2025 Annual International Solid Freeform Fabrication Symposium (SFF Symp 2025)
|
| Symposium
|
2025 Annual International Solid Freeform Fabrication Symposium (SFF Symp 2025)
|
| Presentation Title |
Digital Twin-Enabled Intelligent Adaptive Control and Defect Mitigation for Robotic Additive Manufacturing Process |
| Author(s) |
Sen Liu |
| On-Site Speaker (Planned) |
Sen Liu |
| Abstract Scope |
This paper presents a digital twin synchronization framework to enable real-time communication between virtual environment and physical robot for robotic additive manufacturing processes. A novel Unity-ROS2 data synchronization connection was developed to send the virtual robot’s joint data via a TCP server and instantly move the physical robot with a delay of 20 ms. The reinforcement learning-based smart agent was included in the framework that enables the virtual robot task training (20× faster than real robot training) and adaptive control of the physical robot arm for new tasks. We have developed our methodology using a ViperX 300s robot arm, implementing two distinct environments: static target reaching and dynamic target following to mimic the defect correction in real practice. Experimental results show rapid policy convergence and robust task execution in both simulated and physical environments, with performance metrics including cumulative reward, value loss, and entropy demonstrating the effectiveness of the approach. |
| Proceedings Inclusion? |
Planned: Post-meeting proceedings |