Idea & Project Partners

The CityLAB invited four universities and other cooperation partners to explore the potential of data-driven innovation together. The goals were developing new insights and tools as well as a better understanding of how our cities need to change and adapt in the face of modern challenges.

The first part of the summerschool consisted of several expert lectures from these institutions. Then, interdisciplinary teams experimented with the practical implementation of the challenges.

Challenge

Our challenge partner was the Stadtmanufaktur Berlin. One of the many projects of the Stadtmanufaktur are hydroponics systems. In hydroponics, plants are grown in water. The plants are kept in a granulate or nutrient solution and do completely without soil. This procedure saves space, money and reduces the amount of work required for plant care.

The Stadtmanufaktur is faced with the challenge of monitoring existing and new hydroponics systems and displaying the results in an app. The plants are to be monitored in their growth and wellbeing.

The aim was to determine the necessary parameters, record them sensory and visualize the data obtained in this way. The aim was to process and visualize the data in a way so that even non-experts could interact with it in a meaningful way. Within the framework of the Summerschool 2020, this challenge was to be examined and worked on practically by an interdisciplinary team.

Team

Our team consisted of students from three universities – HTW, FHP and Charité. We contributed our expertise in business informatics, interface design and medicine. 

HTW Berlin – University of Applied Sciences for Engineering and Economics
Kapitalina Haensel | Daniel Hiller | Chin Cheung | Christoph Henning

FH Potsdam – University of Applied Sciences Potsdam
Anna Meide

Charité University Hospital
Marc Higgens

Data pipeline & prototyping

The prototype we worked on was the so-called Dutch Bucket, which we subsequently christened the DIY (do it yourself) Dutch Bucket.

As a first step we had to face the hardware and software requirements. As hardware components we used several Arduino Nano 33 BLE boards and an Octopus ESP-12F. The sensor technology – consisting of multiple different sensors – was either installed directly on the development boards or extended via plug-in boards.

In the course of the project, different software was used. To access the functions of the Arduino boards we used the Arduino IDE. IDE stands for Integrated Development Environment, meaning that it is a simple development environment, that can be accessed via the Arduino editor on a computer.

Once the data was read using the sensors and the Arduino IDE, we used ThingSpeak, an online platform that allows data to be transferred to a cloud via an API and the HTTP or MQTT protocol. In addition, ThingSpeak allowed us to analyze and roughly visualize the data with the help of MATLAB.

We used the online tool Figma to design the app prototype. The software is free of charge and enables simultaneous, collaborative work for students. The app prototype is a digital click prototype that could not be connected to real data due to the tight time frame.

Sensors

At the beginning of the challenge, the participants took part in a hands-on workshop at the Stadtmanufaktur. The workshop served to get to know the challenge using practical examples. The subject of the workshop was the manufacturing of a so-called Dutch Bucket. Roughly speaking, the construction consists of two plastic buckets, a piece of cloth, a hose, plant granules and seedlings. In the application case, the aim was to collect and evaluate the parameters of the plant. The parameters to be collected were left open and were developed within the challenge.

There was a subdivision into spatial and plant-related sensors. Temperature and humidity were measured primarily in the room, since in theory, plant areas (like green houses) were assumed. All other sensors used were plant-related and located directly on the plant or the Dutch Bucket.

The following sensors were used:

- Temperature sensor
- Air humidity sensor
- Ultrasonic sensor (to measure the water level)
- Color sensor
- Photo resistor
- Air quality measurement (MQ-135 sensor)
- Electrical conductivity of water
To measure the electrical conductivity of the water, a connecting cable was connected to the A0 input and the 3.3V output. Furthermore a distributor from the output to the A0 input was installed. To avoid short circuits, a separate 10 Ohm resistor was installed, which diverts the overvoltage into the GND. The results were verified in rinse water, salt water and sugar water.

source code: The used code can be found on GitHub.

Data visualization

Not all sensors could be controlled properly, so that the collected data sometimes made no sense in the visualization with MATLAB. In the end, we reduced the number of sensors to those that could be used reliably.

The data visualization primarily served the understanding of data. Which data can be collected and which can be combined in a meaningful way to make relevant statements.

App Development

The next step within the challenge was the development of an app prototype. The app should serve as an interface between data and data non-expert users.

Due to the very limited time we developed only three screens. These show how a potential user looks at the condition of a basil plant. 1. She selects the plant from among my plants 2. then comes to the overview screen and recognizes at first glance what is problematic with the plant by the traffic light symbology. 3. As a final step the user can look at a detailed data visualization and see in the use case e.g. that the ph value was stable over the week and increased strongly towards the weekend.

The suggestion points out that due to the water level, which is already shown in yellow on the overview site, it is strongly possible that the plant should simply be watered.

Visual language

The application's visual language was based on a sentence from the challenge: „Mobile blue green infrastructure. Monitoring of building blocks of a climate resilient city.“ Different shades of blue and the color green were defined as the main colors. The signal color red was adopted as the classic warning color and then the color palette was extended to a yellow as a traffic light warning system. Furthermore, the Apple Human Interface Guidelines were used as a visual bracket in the design.

Future Scenarios

With the so-called future scenarios we wanted to show the potential of hydroponics systems. We showed photos of already existing projects in the field of vertical urban landscaping as well as school projects that deal with vegetation because of the DIY character of our prototype.

We also enriched the presentations with process graphics to show how sensors, data collection, data visualization and front-end interfaces interact with each other.

Reflexion

I am extremely glad to have participated in the Summerschool 2020, despite the risks of Corona.

This time I was able to gain the valuable experience of working closely with business informatics specialists. The approach of this discipline was noticeably different from my own. A circumstance that once again showed me that interdisciplinary teams with clear expertise can work together very fruitfully.

This was the third course in my study path that dealt with data visualization. And it was only now that I got the feeling to understand what it is all about and how exactly the process of such a project is. A great experience! Because the first indispensable steps in implementing a project with social impact seem to me personally to be a well thought-out data pipeline and project plan.

Regarding my visualization skills as a future interface designer I had to realize that I still have a lot of catching up to do. Both the level of visual expressiveness and the speed at which I am able to implement it urgently need to be improved.

As an outlook for me personally, I would very much like to participate in the development of an app that covers the entire installations of the Stadtmanufaktur. Preferably under the lead of an experienced interface designer.

Many thanks at this point to everyone involved in the Summer School 2020!