Collecting data using inexpensive computers and bucket covers

Text: Sari Putkonen

The idea of Helsinki-based VekotinVerstas (Gadget Workshop) is to share data about sensors and sensor platforms and to build a community around the Internet of Things (IoT) using open source code.

“This idea came from my interest in sensors and that of Aapo Rista, one of the organisers of the workshop. We have for long collected data, for example, about water pipes that freeze easily. We have seen how technologies have advanced and how measuring has become relatively easy. In addition, the prices of sensors and sensor platforms, computers the size of a matchbox, have dropped to a fraction of what they were ten years ago”, says Henri Kotkanen, who is involved in developing the activities of VekotinVerstas and who works in the IoT team of Forum Virium.
 
VekotinVerstas is only getting really started. So far, it has already installed sensors in an offi- cial’s workspace to collect data about the quality of air for building management. “One of our most fun projects has been an environmental artwork that would be built in Kalasatama to measure moisture in plants. It would then indicate when it’s time to water the plants”, Kotkanen says.

Data about the quality of air through crowdsourcing

The idea of crowdsourcing is to encourage other people to collect data. Its results are based on the success of campaign messages and the identification of the correct target group. When VekotinVerstas held a sensor workshop at the end of October, it marketed it to everyone who is interested in the IoT.

Building boxes to measure air quality raised interest, and all eight positions were filled. During the workshop, the participants built boxes that they should carry around town to measure the quality of air. The results of the project will be available later. “Official air quality measurements are carried out using equipment that costs hundreds of thousands of euros. In the VekotinVerstas project, the same measurements are conducted using simple inexpensive components”, Kotkanen says.

The project studies what kind of data can be collected using inexpensive equipment and what kinds of measurement errors there are. Kotkanen hopes that data collected in this way could also be used in research in the future.

Collecting data about observations made in nature

The Finnish Environment Institute (SYKE) has a three-year Envibase project in progress. In one of its sub-projects, regular people are encour- aged to make different observations in nature. In addition, a service to save and distribute data about observations made has been developed during the project. “People are very active, as long as the correct target group can be found”, says Timo Pyhälahti, senior expert at SYKE. “Birdwatchers make useful observations of birds. However, they may not be able to report, for example, observations of bats via their own services.”

Crowdsourcing helps to collect data from where people are. This also means that there can be too much or too little data: no observations are made in some areas, while in some areas there are more observations than what are needed. Pyhälahti says that it would be useful if a service that brings all observations together encouraged people to make observations in new areas.

Crowdsourcing always involves the question whether the data material is reliable. JärviWiki and MeriWiki are the most long-standing observation services of SYKE. These are built and published in cooperation between the authorities and regular people. The data collected by regular people in JärviWiki has been studied, and its quality has been found to be as high as that of data collected by researchers.

Bucket covers and other measuring devices

People have used simple measuring devices to collect data in SYKE’s project. One of these is the Secchi disk that can be made, for example, from a white bucket cover. “The test identifies the depth at which the disk can no longer be seen. This gives us data about water clarity, and longer time series indi- cate any changes over time”, Pyhälahti says. Peo- ple were also able to test the iQwtr water quality indicator based on the Secchi3000 technology. Its price was less than EUR 20.

Catching data

When collecting data from regular people, the technologies used have room for improvement. People catch data, and it cannot be used easily.
“There is a broad range of devices to collect data, but not many cooperation platforms. Some services are open, while some are closed. Often, there are no ways to share data between systems”, says Pyhälahti. There are also questions related to rights to use different material. In the SYKE project, data material is shared as open access data using the Creative Commons 0 (CC0) licence.

This means that the licensed data can be freely used in every possible way. Data is anonymised and all personal data is removed. Kotkanen dreams of a data warehouse of the future where people could save their data and define rights to access this data, so that public organisations would be able to use data for research purposes, while companies would only have access to this data in exchange of a fee.

Encourage and reward others

SYKE and VekotinVerstas return data collected by means of crowdsourcing to people as open access data. Results are also demonstrated using visual map material. More innovative ways of rewarding data collectors have been considered. Kotkanen proposes that, for example, someone who gives the City of Helsinki access to their data for research purposes could be rewarded by using Helsinki coins that can be used to purchase, say, public transport tickets.

A 70 euro air quality meter

VekotinVerstas wanted to keep the costs of its air quality meters low. The simple box consists of air sensors (EUR 40), a Raspberry Pi 0 W computer (EUR 10), a power source and a box. A smartphone app is also needed to read the data sent from the box and to add information about the time and place to this data.

A dumb sensor measures data. A smart sensor platform sends data via a Bluetooth connection to a smartphone, from which data is sent to a cloud. Voilà! One air quality meter has been built. It can be built by anyone. More detailed instructions are available from GitHub.

 

Article in pdf >>
Positio 2/2019 table of contents >>