5G-Enabled Legacy Machinery for Low-Cost Sensing and Connectivity
Project Lead: Rutgers
Member % Cost Share: 50%
CESMII % Cost Share: 50%
Duration: 12 Months
Problem Statement
Legacy machinery is deeply rooted in the US manufacturing industry. However, the lack of network connectivity has prevented them from turning the real-time manufacturing data into insights and business values. Therefore, it is very challenging to augment the capabilities of legacy machinery by leveraging wireless sensors for online process diagnosis, prognosis, and control with use cases.
Project Goal
To develop a smart milling testbed by integrating a legacy mill with a 5G-enabled accelerometer for real-time chatter diagnosis and prognosis. Develop an adaptive automatic repairing testbed by integrating a legacy lathe with a 5G-enabled laser scanner and multi-tool robot for real-time identification, cleaning, deposition, machining of high-value components.
Technical Approach
Augmenting legacy machinery with wireless sensors is a viable technique route to solve the challenging problem. Integrating legacy manufacturing equipment with new wireless sensors and edge computing would be the most cost-effective way to augment the capabilities of legacy machinery on the shop floor. Two distinct use cases, blade milling and automatic repairing, are proposed to demonstrate the proposed approach. The overall technical approach is to augment a legacy mill and a legacy lathe by integrating wireless sensors (i.e., 5G-enabled accelerometer and 5G-enabled laser scanner, respectively), and streaming process data to the 5G edge for data analytics, machine learning, and edge-based control.
Key Tasks & Milestones
-
Integrating legacy mill and real-time 5G monitoring of milling vibration – A 5G enabled wireless accelerometer to collect vibration signals and milling chatter
-
Chatter diagnosis and prognosis – Real-time learning-based chatter predictive tool
-
Integrating legacy lathe – A 5G edge-based vision sensing system to identify defects of parts with known reference
-
Defects identification and quantification – A reference-free 5G edge integrated vision sensing system to identify defects in real-time for robot motion control
- Automatic defect repairing via edge-based deep learning – A 5G edge based control algorithm implementation method and a system via vision sensing without typography reference
Potential Impact
Live demos of augmented legacy machinery with wireless sensors and 5G edge with ultra-low latency (< 50 ms), high-speed connectivity (up to 10 Gb/s), high flexibility and mobility, and low installation cost and complexity.
Benefits
- Pre-competitive enabling 5G sensing and connecting technologies
- Best practices of 5G smart manufacturing technology with use cases
Project Selection & Announcements
