Discover the IoT App-Store and Connect your devices to the world of IoT
IoT and data science consulting: visualizing optical sensor data

Use Case: Reading and Visualizing IoT Sensor Data

The Record Evolution Platform provides a general foundation for different IoT application scenarios. In this article, we present the interaction between optical high-precision sensors developed by Precitec and the platform. We connect a Precitec device to the platform and install an IoT app on it to read out and visualize IoT sensor data.

Precitec Optronik is a worldwide leader in the manufacturing and development of solutions in the field of laser technology and optical metrology. Precitec’s CHRocodile sensors are the industry standard for wall thickness measurement of glasses used in a wide range of applications such as glass bottle production, CERAN glasses, medical glass containers, smartphone inspection, and many more.

Using the Record Evolution platform, we connect an edge device (Raspberry Pi 4) to a CHRocodile Compact device. We collect real time sensor data and build a dashboard to monitor the measured glass thickness. The dashboard is hosted on the IoT platform. Authorized users can inspect it from anywhere directly from the web browser. 

The User Story

As a user, I want to analyze data measured by the CHRocodile device used in the production line. On the one hand, I want to know more about the current production status and want to receive notifications if something goes wrong. On the other hand, I want to be able to analyze the historical data to gain further insights into production failures. 

Also, I might be interested in including the data of our other IoT sensors (like temperature or pressure) into the analysis in order to discover typical causes for errors. Further, I want to share my analysis with selected users and present the findings on a dashboard. The implementation should be simple, ideally without coding, since I have no experience in building IoT pipelines and setting up databases.

Architecture Overview

A CHRocodile optical sensor device working together with the IoT platform - a schematic representation
Image 1. An architecture overview

In our case, the edge device is a Raspberry Pi running ReswarmOS, which is a Unix-based operating system specifically designed for IoT scenarios. Furthermore, ReswarmOS allows the IoT device to be remotely controlled by the Record Evolution IoT Development Studio. The sensor data gathered by the CHRocodile device is transferred over an ethernet cable to the Raspberry Pi. From there, it is sent (over Websocket) to the Record Evolution Data Science Studio, where the data stream goes into a data warehouse and becomes available for dashboard visualizations or data analyses. 

The IoT Development Studio allows for the remote control of the edge device. This involves, for example, changing device settings (Ethernet, WiFi) and installing, starting & stopping IoT apps. For a comprehensive analysis of the gathered data, the Data Science Studio provides built-in data science workbooks allowing you to run explorative SQL queries and perform machine learning analyses using Python. 

Additionally, you can create custom dashboards and visualizations (infographics) that are hosted on the platform and can be shared with authorized users. Note that these infographics are also connected over Websocket, which means that the sensor reading from the CHRocodile device can be inspected in real-time in the dashboards.

The Setup

the initial setup for extracting IoT sensor data
Image 2. The setup for extracting and reading IoT sensor data from the device

Building a solid IoT pipeline can be time-consuming and complicated. Using the Record Evolution platform can make this process significantly easier.

We follow these steps:

  1. Create an account for the Record Evolution Platform
  2. Download the REflasher app to flash the SD card of the Raspberry Pi
  3. Install the CHRocodile Sender app on your device
  4. Configure the ethernet interface of your device
  5. Create a data pod
  6. Create an IoT source in the Data Science Studio
  7. Build a data pipe for the sensor data

Note that only the last step involves writing an SQL query for data transformation. All remaining steps require no code at all. 

The CHRocodile Sender App

the app for receiving IoT sensor data on the Record Evolution app store
Image 3. The app for receiving IoT sensor data as shown on the platform’s App Store

Similar to mobile platforms like Android or IOS, the IoT Development Studio has its own app store, allowing the distribution and convenient installation of apps on connected devices.

The CHRocodile Sender App (currently not publicly available in the store) fetches data packets from the CHRocodile device, parses them to JSON, and transfers them to the Data Studio (or any WAMP router). 

Device Configuration

the IoT app for collecting IoT sensor data as installed on the device
Image 4. The IoT app as installed on the device

As soon as the edge device establishes an internet connection, it can be controlled remotely. With a few clicks, you can install, start and stop apps, alter user privileges, change ethernet settings, install updates, or apply other settings. In our use case, it is crucial to adjust the IP address of the ethernet interface to the address range of the CHRocodile device. Then, we can start the CHRocodile Sender App and are ready to import the data.

Data Import

setting up the data import environment
Image 5. Setting up the data import from the device

For this use case, we will create a new data pod in the Data Science Studio. In order to import the IoT data into the data pod, we can define several different data sources. We use the IoT source, which connects to the packets published by the  Chrocodile Sender App. Then we will create a “raw table”, which will receive all data from the source.

So far, not a single line of code had to be written but the base of our pipeline is already finished.

To improve on our pipeline, we can add a transformation pipe, which transforms the JSON packets to an ordinary column structure. On top of that, we can start creating reactive dashboards and analyses.

transforming the IoT sensor data in pipes
Image 6. Once the data is loaded into the platform, the next step is transforming the IoT sensor data in pipes

The Dashboard

visualizing the IoT sensor data in dashboards on the platform
Image 7. Visualizing the IoT sensor data in a dashboard on the platform

Using the Record Evolution platform, we can create reactive dashboards that consume the data from our IoT pipeline. In general, the dashboards are created using Javascript. Instead of building everything from scratch, we can leverage existing Javascript Libraries like Chart.js or Plotly. These provide a wide range of different mobile-ready responsive chart types.

For demonstration purposes, we have built a dashboard that monitors the IoT sensor value for the wall thickness in a glass bottle production line. 

The monitor shows the current production quality and the signal data of the last minute. It also shows statistics on the historical data. The dashboard updates are in near real-time and can be shared with other authorized users.

Conclusion

Creating IoT pipelines from scratch is no trivial task and often requires profound knowledge about IoT protocols, device connectivity, cloud computing, and databases, among others. 

The Record Evolution Platform enables you to create carefully designed and robust IoT pipelines in just a few steps. There is no need to implement your own infrastructure. Additionally, the integration of the Data Science Studio provides a scalable data warehouse for storing historical data. The Data Science Studio offers an environment for extensive data analyses and interactive real-time dashboards.

Using the Record Evolution Platform in combination with Precitec’s optical sensors can provide the foundation for a wide range of use cases. This involves production line monitoring, remote maintenance, or comprehensive data analytics on historical data.

Record Evolution Logo

About Record Evolution

We are a data science and IoT team based in Frankfurt, Germany, that helps companies of all sizes innovate at scale. That’s why we’ve developed an easy-to-use industrial IoT platform that enables fast development cycles and allows everyone to benefit from the possibilities of IoT and AI.