Catch up on what the experimenters have been up to this week.
Children’s artwork was put up at the school, next to their rainwater tanks.
Following a stakeholder meeting with the Alberta Tenants and Residents’ Association (TRA) we agreed not to decorate the Ambergate tank as they preferred it neutral. We did agree to change the tap so they could more easily attach a hose to it.
We have been working on developing our Artificial Intelligence model. The model aims to optimise our tanks’ levels. This is done using several parameters (e.g. site and tank characteristics, environmental factors, etc.) and based on various, sometime conflicting, objectives (e.g. maximising water level in the tank for re-use, etc.). We have been using some historical as well as the data we have been collecting at our sites to develop the first stage of the model. The results from the initial model runs are promising, and we are now moving to next step to enhance the model and calibrate it.
It has been a very complicated week for the ParGarCom project, especially for programmers. When it looked like the app update (with substantial changes allowing uploading of images and comments was complete), we detected problems with the Organicity API. We were able to create assets with comments but when we issued a request to the Organicity servers, it returned them to us without the text and we only had the id and geolocation of the comment. On the other hand, if the query was made by requesting a certain comment, the Organicity servers returned the complete information to us. This happened not only with comments but also with images.
To solve these problems, we sent an email to our mentor. While awaiting response, we have looked for a provisional solution, redirecting the storage of information (comments and images) to ParGarCom's servers, in order to upload the update and allow several Santander users can perform a real test of the app, before officially launching it in the first week of March.
Once problems with the Organicity API are resolved, the information will be redirected again.
Past week there were some issues with the accuracy of the measurements of the fill-level devices as when the battery life deteriorated the measurement deviated. This was due to different voltages of 3.3V and 5V in the circuit that had to be changed. A booster for that was ordering, and after thorough calibration, the issue was resolved.
Meanwhile, the existing devices were brought back and re-deployed. Currently, the rest of the devices is being prepared for final deployment Monday next week. This will be done together with the QR codes and buttons.
Co-creation activities and talking to similar (IoT) experimenters from other contexts has to lead to many ideas and suggestions to be tested soon of engaging the citizens properly. In Herning, the Danish soccer team is currently number one in the league, and the incentive for participation if citizens can win a ticket to a match is high. So by including the local community one can engage citizens in reporting filling levels, this is to be exploited. Citizens will be able to report back fill-level but also see live how sensors are in the bin and the fill level etc. Safety measures has been ensured in Azure Cloud.
As connecting to the OrganiCity API is crucial for the experiment, a lot of effort has been put into these. An experiment, application and assets have been created and the data from the sensros is now being linked and stored. What is remaining is formatting the assets according to the Organicity specifications. It is expected that all the data will be integrated by next week.
The last weeks were spent in programming mode over at Kanda and today the whole team got to download the first test release of the app - yaay! All of us will now review it and pinpoint where it needs finetuning before we start our 2 weeks of user testing from the 5th of March.
Last week, we had the first meeting with the Advisory board that we have put together with the aim to gain feedback for our concept from professionals in the field of citizen safety. The Advisory board has participants from the Police (innovation department), the Fire department (responsible for the regional contingency plan), Aarhus Municipality (responsible for digital communications with citizens, also with a focus on crisis situations), Danish Emergency Management Agency / Mobilvarsling.dk (project manager for an app that pushes notifications about urgent traffic, weather and crisis matters) and Orbicon Informatik (smart city specialist).
We had a fruitful meeting with workshop elements where participants provided feedback and new ideas to further sharpen our concept. Reactions from the participants were positive, they saw the need for more channels for the authorities to reach citizens with important messages. Aarhus municipality's "Citizen Tip" app has proved very successful, which indicates that citizens are happy to contribute to creating a better city together through digital platforms. SafeCity provides a new angle in this context, providing also the opportunity for citizens to interact with and help each other - participants found this an interesting perspective.
For the SafeCity platform to gain trust with citizens, the authorities should be the owner of it. Our intention is also to use the Advisory board for creating awareness for the concept in the respective organisations, with the aim to find the right organisational setting for it to "land" and be further developed into a full-fledged app that can be used in Aarhus as well as in other cities around the world.
It has been an intense week for the TalkingCity team! While the first version of the chatbot was getting tested in Aarhus, with the support of our local Anorak partner, the dev team in Trento was working hard on improving the usability and features of the prototype. And, in the meantime, our colleague Rossana was busy managing the official Organicity Instagram account 🙂
The next version of the chatbot will be released on Monday 26/2, and will go through another round of testing in Aarhus. Also, next week the experiment lead, Dr. Miorandi, will be in Aarhus to meet with the local stakeholders and work on the post-experiment exploitation of the service. In case you want to meet him, drop a line at email@example.com, the coffee is on him!
This week we are hosting our first Tranquil City exploration events, one in each of our experimentation areas, London Bridge & Bankside, City of London and Deptford.
The purpose of the explorations is to encourage people to think about and better understand what their impression of tranquillity is in the urban context, as well as to identify which spaces and features may help to invoke that feeling. This thinking will help encourage more people to explore tranquillity in their own time, make them aware of the Tranquil City project and our objectives, as well as hopefully contribute spaces (subjective data) to the Tranquil Pavement London map.
Our first exploration was held in London Bridge & Bankside and was during Wednesday 12:30pm to 13:30pm, to explore how tranquillity can exist even in the time you have for your lunch break. The exploration was really successful, and we had a good, mixed group of people thinking about what tranquillity meant for them.
When discussing the exploration with one of our stakeholders, Team London Bridge, they suggested that our exploration deviated away from their focus area a little too much, and we sat down following the event to plan how our other walks could be planned specifically for their area of interest. Team London Bridge’s interest in the event is to identify which spaces people feel are tranquil already, as well as defining what type of urban features can help people feel more tranquil. Team London Bridge have plans to work with Better Bankside (another one of our stakeholders) to re-invigorate an area between London Bridge and Bankside, inviting people off busy streets onto quieter, cleaner and pedestrianised pathways, promoting better health and wellbeing. We will be analysing our participant’s responses next week, understanding which features we could help introduce into the areas planned for improvement, increasing the area’s ability to invoke the feeling tranquillity in such a vibrant, busy and urban environment.
This week our collaborators Outlandish Co-operative posted a blog about their experience with creating the Tranquil Pavement London with us, from a more technical, big data perspective. Really helpful tips for anyone looking to display 5.85 million data points in a web-app! Outlandish
We are also now planning for our second development stint with Outlandish, to refine the web-app following our user feedback, as well as hopefully implementing an additional feature.
If you’re in London this weekend, come along to one of our Tranquil City explorations!
Saturday 24th February @ 12:00pm – 14:00pm – City of London
Sign up here.
Sunday 25th February @ 12:00pm – 14:00pm – Deptford
Sign up here.
Two weeks out from our first workshops and things are getting serious in the WearAQ camp. This week has been about making the final connections between all the different pieces of the platform. Primarily, we’ve been making sure that the server works consistently and remains stable across 20 devices each pushing and retrieving data from the backend.
The wearables are all built on microcontrollers that are linked to a phone and interface with an app which pings the server with data from perception and measured AQ. We worked on making sure that the apps are synchronised and uploading data to the server with no errors. The week began with the apps flip flopping between errors across the board to full synchronisation. This was one of the issues. Thankfully, we’ve cleared that up and now everything wearable->app->server seems to be working consistently.
The priority next week is to make sure that the model to predict and recommend areas to collect data is linked up to the platform. Since it’s the final step, we’ll be moving on to testing and making sure that everything goes smoothly for the first workshop. Look forward to next weeks update where we’ll go into more detail about the model and how we plan to use the data we’re collecting. As always, get in touch if you have any questions!