Research Overview
With the widespread development of Industry 4.0 and smart manufacturing concepts across industries, sensor development, system integration, and data analysis have become important aspects of efficient manufacturing operations. This research presents the development of a sensor system capable of sampling acceleration and rotation data up to 400 Hz and wirelessly transmitting the data over Bluetooth Low Energy (BLE)
Publications
-
[1] Selvaraj, V., Nagaraj, A., Whiffen, B., Gregory, & Min, S. (2024). Development of a wireless smart sensor system and case study on lifting risk assessment. Procedia Manufacturing (Accepted)
Why was this study conducted?
- An effort to design and open-source hardware and software architecture for data collection applications in manufacturing scenarios.
- Enabled modularity in the hardware designs.
- Synchronized DAQ across multiple sensors.
- To understand the limitations of wireless communication protocols for DAQ.
System's Current Capabilities
- Can sample IMU, and orientation at 400Hz.
- The open-sourced hardware and modular design enable the addition of multiple sensors - temperature, pressure, etc., for different applications.
- Signal processing and Machine Learning at edge (tinyML)
High level overview of the hardware and software for the wireless sensor system.
Software Development
Wireless Sensor System Features
| Feature | Details |
|---|---|
| RTOS | Zephyr |
| Protocols | Bluetooth Low Energy (BLE) |
| Firmware Updates | OTA (Over-The-Air) |
| Power | Battery Operated with continuous acquisition time of 30hrs |
| Network Topology | Mesh (To enable synchronous DAQ) |
| Data Collection | Continuous data streaming or event-driven data transmission |
| Sensor Diversity | Ability to add and remove sensors easily |
| AI | Machine learning at edge |
| Low Power Consumption | Energy-efficient design with advanced sleep modes |
| Security | Encrypted data transmission and device authentication |
Supremus - An application to control the sensor systems (Windows, Linux, & macOS)
| Feature | Details |
|---|---|
| Framework | Node.js, Electron |
| Firmware Update | Can update the firmware for individual sensor systems remotely |
| Cloud Connectivity | AWS Cloud for data logging and AWS Sagemaker inference |
| Real-time Monitoring | Live data plotting (Acceleration and Orientation) |
| AI | Machine Learning model deployment at Edge (On the sensor system or in the laptop running supremus) |
| Sensor Configuration | Enable remote updates to sensor's sampling rate, sensitivity, range, etc., |
| Maintenance | Can interface with multiple sensor systems at a time |
Code Repositories
- Worker lifting risk assessment study using the developed sensor systems - https://github.com/vigneshuw/work_risk_assessment
- Firmware adapted to NiclaSenseME from Arduino due to its ease of availability - https://github.com/vigneshuw/NiclaSenseME
Demo
A demo consisting of three wireless sensors systems, that are controlled using the application supremus running on macOS. In the demo the data were collected synchronously from three sensors and logged locally, additionally, live plots were created for real-time monitoring.
Wireless Sensor Systems were tested using a case study can be found here