Overview / Usage

Most universities (Kenyan) do not have sufficient equipment to help equip students for the 4th Industrial Revolution. This project looks into building a Predictive Maintenance and Condition Monitoring Learning Kit. Leveraging Robotics, IoT and Machine Learning technologies the platform will provide an open-source, low cost and easy to build learning kit to better equip learning laboratories and help prepare students for the 4th Industrial Revolution workforce.

Methodology / Approach

The system entails a 6-dof Robotic Arm that acts as a model industrial machine. A custom control circuit board with a STM32F4 Discovery board is used to control the robot movement ans well as collect sensor data that relate to the machine's "health". In this case tilt in 4 axes, vibration and Noise. The sensor data is then serially transmitted to an IoT gateway. The STM32 micro-controller runs a Real Time Operating System (FreeRTOS) to enable both control and sensing to take place concurrently. This setup is housed in a 3D-Printed enclosure and mounted onto the industrial machine.

The IoT gateway contains both WiFi and 2G transmission capabilities to send the sensor data to a cloud platform for storage. This data can then be used to train a machine learning model that is able to predict whether the machine is working nominally or approaching failure as it operates in real time.

Technologies Used

• Robotics
• Internet of Things
• Machine Learning
• Embedded systems