Merken

Okinesio – Development of an Open Hardware Activity Tracker

Zitierlink des Filmsegments
Embed Code

Automatisierte Medienanalyse

Beta
Erkannte Entitäten
Sprachtranskript
home I thank my mom and only have chosen E
B OK and thanks for coming to my
talk about our research project needs your uh which
we are working on and off University in an image
that's in southern Germany in north and the barrier for about 2 years we're funded by the and the variance states and we have a lot of fun developing an open-source hardware for an activity tracking we had team off and 6 people right now and I'm teaching injection assigned at the wolf university and I have tools and research assistants with me and I'm about to win Sri students which are working on the project and bring different tasks like electronics and information the information the solicitation and so on at 1st we
ask the question why what do we really have to and develop another tracker another and fitness tracker because they're not all of them on the market 2014 there were only if it didn't 90 but in 2015 there were not of brands on the market which you can buy for about 50 euros to 100 euros and there were quite good they're looking good have precise and what's the reason
why and we are developing another activity check an open hardware the activity tracker and because we were searching for a very long for a tracker and where we are holding our own data where we can work with our data download our data and visualize what data and we did not find anyone on the market which happens sufficient API to work with the data and and and there are a lot of privacy issues was the conductivity trackers and you don't own debated you don't know what happens with the data and once they are uploaded to the cloud and we tested a lot of them and we did some research were as proxies and so on and had a look at what's happening with this data and we whereas the bit afraid and about what we saw on and that's the reason why we thought we we should develop another 1
1 uh where your own in your data where you decide as the user what happens with your data and once you upload all which advised and regarding the data we have a lot of time on people who want this activity data and it's not only how many steps you're walking each day but it's a lot of information about how we sleeping uh end up automatically conditions on so insurers and in Germany for example are very keen on getting this data and we think a lot of people you wouldn't want to have their has data have noted and analyzed by different companies so we decided we produced our own a teacher and and in the beginning we didn't exactly know how to do it and that if we can do it because we're sciences from uh we pretty coders but we are not years electrical engineers and so we are based on the Arduino platform on which were with which you're working for several years and as open hardware platform gives it a kind of
democracy in democratization of fiber production I everyone designers artists and so on are able to build up on what other and designers created was we and create remix of newer hardware platform called off existing out that forms this democratization and what we have
seen em everywhere we're looking around especially here and in the past and uh we we see that I'm only a few people could use plastics and um model is out of all of the 6 grenades uh for example with this extruder and
nowadays we and we all have 3 D printers in office and uh at our universities where we can produce ideas within an hour or 2 all of which would have taken the months and a lot of money in the past the same
and the same situation we have an electron it's which is the uh and new development of where in the past only if you could create a new piece of the uh um uh dates for new electronics and nowadays with for example the other we
have that font we can uh creating new interfaces new devices with in and some minutes or and we're is platforms like fruits and from put some uh we we we order it also able to create an fixed new piece of these firms from our sketches and creates monitored devices of our own this is from the area you know platform and so it takes only a 2 weeks to produce them such a piece of in in your Institute they're on radio and some boards which we could have beat up on on the new odd
we 1 for example it has some 3 DOF as exterior I'm Stenzel's actor meters periscopes for tracking steps and motion but it's quite large it's a very affordable from about so the of what Europe but it's quite large so it's too large to fit it for example in the clock or in a know the same you can put on and the other 1 is from and then has also very good tracker was due to the but there we sorted submitted too complicated to work with this that too much features that work which features we do not really need so we have created a
lot on with a focus on simplicity and we have only 1 up the border not my process or and and in my 160 sense or for the motion tracking which is a great little device the smaller 1 in the back and it's only 3 by 2 millimeters and it's very intelligent and attracts steps and saves a lot of energy so it was a perfect and sends or for um for all session before we started
from before we started the development we did the evaluation of current hot then we took more popular on the market in 2015 last year there's Fitbit Flex miss the Chinese job on Oct 24 with things activity pauper garment is B was smart and Sony as w art 10 so these were the most popular trekkers in 2015 we took them and several it would
relations with 1 another we we had the a runner which 2 different of exercises running 100 steps walking 100 steps running 600 steps walking 600 steps and become pay out there on the steps we were actually running and as the results the trek out so as to divide us and I on the other hand if we had a small because we put a small robot which was simulated the movement of the hand in order to reproduce and so on to reproduce the running tasks and to eliminate the arrows which could have been a made by the runner for example so we have these 2 approaches this looked at them like this we had really what fracas on the 1 hand we did and other exercises with only 1 tracker and and then we compare out of the results the
results were quite interesting and most of the echo the track did really find on you see it here the you can read it from the the tool to our tracker and there's the misfit shine words should didn't whereas in the 100 meter I walking on tests body in the running tests and they ordered to quite well because there's much more energy and more weight on the sensor or and that's quite common so more and more accurate yeah electrical didn't work work it
would so that's certain no or trigger
did and not very well in the walking on test that's why we that's because we didn't calibrated the to the weight and the size of the run in the 1st and we have this uh um enabled now on all the accuracy is much better and in the running task so you result calibration we where deliver the same quality as commercial trackers and on the market that was a proof that the whole hardware especially with the course the my 167 little works as a great as commercial trackers and so uh we knew we could go on and produce our own hardware so
it and since we have designed as an
hour and electric engineers we and did and an approach we also approaching in other projects we are we are doing as a trial and error and the learning by doing approaches by time trunk you unduly
and but that's about it the more complicated when you're working with hardware than when you're working with software there's no come
on the floor and going a step back and there's no button for recompiling hardware when you did some something wrong you have to create new the ports and wait 2 weeks on the they of produced in order to eliminate your error that was something we have to learn and it was an interesting learning curve we did that last month and on the left side using our 1st prototype that's a standard on we know micro and on top of this is our and our go or for the view my 160 and we use to commercial and we know what and tried if I was in the works of OK with this board we did it looked a bit sketchy but in the end it worked very well and since this work we could go on and and
such an open hardware platform we have on the so we didn't have to and created much the complete UPC he bought from scratch that wouldn't have been possible for us and but we build upon the works of other designers and here I believe that use the hardly we design which is very great and fit our purposes so 1st more board with since reboard around and 3 process or uh which is working on top of recording said we want to get work was printing platform have from and if a pot some and to create a new to remixing new board of the pad border to fit all a size and where we could put in our emotions and so on which is now on board of the other we know this look like uh the 1 on the right side and then we produce the
PC people what grew from printing where you have to send your EGA fires counterfeiting in 2 weeks later you get for a ball to assist for euros you get 1 PC people what a blank piece of the board and and then you can order you of parts of the process or and the the motions and all and all the other parts of the use the port and so on and place it on the board and burn it in Europe regional or when the price and is very interesting we're on our head around 20 euros the production costs right now and that's an open hardware project and so we on selling them yet and and we have by buying the parts in a very small amount so it's quite OK to get of finished modeling of pork was motion sensor for around 20 euros and How does this production work in the beginning in the beginning we repeat because these parts are very small at the the unwise support tool by 3 millimeters wide depends on the on the board on the a half a millimetre wide so it's very tiny very small and but in the end and it was quite easy to place them them on the board because you don't have to wear like in the past but you placing them on the board like on the left side and then you put it in a regional open which is nothing else on itself over and was the temperature control and and then the so ordering them this certain pays methods and fixes the on the part so would this was quite easy we we have to test to was reports and which didn't work in the beginning and I have to get experience with placing the parts on the board but right now a forward model my resynthesis assistance there and it's a task for our half a day to produce Newport's it's very interesting that we saw it would take much much longer but right now it's really no problem for us to produce new newport as soon as the PCB and it is these right that was approved of it worked on our bought after some learning curves we did a lot of errors like we always do we learn from our errors in the end it worked we are now in the
in the 2nd year we begin with cells in the 15 and we began working on the hardware and some of the cells in 15 and in February and we had our 1st
working on we know more than than it took some more approaches to get the emotions and the working in software working but I'm since tourist 3 weeks we have to find you working for what I do also was read I was a battery order so it's independent and works in stores or motion data so I have 2 thoughts with me if someone would like to beta-test somewhat supportive
tool can make ports was the uh which I can give away and we're happy you if you're using it and what are the next steps what are we doing this the tragic right now we're doing a lot of experiments with those students which is very interesting this semester we had a project was skateboard brand where we're specializing forces of skateboarders here with light for example uh and so we did information was annotations uh uh in 3 D and where we where forces was color and will step focus years and there were very interesting and results in his most about the fun to produce in a state or currently
we are designing of the case in the connector also form we come since we think that privacy is very important we have 3 different approaches we have a an act gets worse and you really connect your and bias to your computer and nothing else so there's no through to see the war and boards to other wireless technology and as the data is stored on your device the 2nd is more of a common approach for more word forms we have the following check life and like some other activity trackers report in your from check entrance a transitive data and so the approach is to propose a new variant of which of more connects to your and phone over the and we're working on the case is a very simple this redhead printed cases cases fit by Rob and so we don't want to go into will review production but producer cases different cases in all that referring to where we are at
most make affairs we were last year we to make of the burden for those unique and a single we will there this year 2 we had should have some of an important paper of the HCI conference where we are and publishing all the results that we made a or the the results were made over the last year in a paper dictionary HCI in Toronto that would be very interesting and we're
publishing all our own efforts and progress on our website can and AUC and we have a
large knowledge base that which is very important for people who want to start creating own devices on PCPs on electronics and all these things we had to learn the 1st year we can put in or knowledge base about electronic Thibault Data API so from the fitness checkers and so on so it's a and growing and knowledge base where we hope that others and profit of the true and was designed and put on the top so if someone wants to remix our board and put some other and on it and creating completely other the virus and it's open for remakes and all the files are certainly source and put them on the top I'm so that at my 20 minutes was thanks for
listening the internet interest and Internet addresses open easy we don't fit the i'm undernourished opinions and that and I'd stiII minus all the use and our project website was the project so our students so if you have any questions about their production and so on I'm happy to answer them yet why is it important to have an to the task and connected to the mobile
devices because of Decatur novels race and not and so why is it important to have an activity detector and then connected to the followed by a and the because of the GPS
signal that sound science sending 1 as a tool to CPU chips which are in there they tractable everywhere so reascending knowledge your device ID and every shop and so on and track you although they always city in the world and so it has some unique ID as it's a problem with that and so we decided we have a form of American version where you can put it
into the phone and transmit data directly we are way on the phone and really work well since so my
amendment
Offene Menge
Hypermedia
Gewicht <Mathematik>
Computeranimation
Hardware
Subtraktion
Hardware
Besprechung/Interview
t-Test
Computeranimation
Task
Weg <Topologie>
Datenverarbeitungssystem
Injektivität
Pay-TV
Notepad-Computer
Vorlesung/Konferenz
Projektive Ebene
Information
Grundraum
Varianz
Bildgebendes Verfahren
Aggregatzustand
Proxy Server
Datenmissbrauch
Hardware
Datennetz
Besprechung/Interview
Wärmeleitfähigkeit
Streuungsdiagramm
Computeranimation
Fitnessfunktion
Metropolitan area network
Bildschirmmaske
Hardware
Offene Menge
Konditionszahl
Biprodukt
Information
Urbild <Mathematik>
Biprodukt
Systemplattform
Computeranimation
Hardware
Informationsmodellierung
Besprechung/Interview
Grundraum
Office-Paket
Offene Menge
Flächeninhalt
Font
Datenverarbeitungssystem
Besprechung/Interview
Softwareentwickler
Systemplattform
Whiteboard
Hardware
Schnittstelle
Energiedichte
Weg <Topologie>
Meter
Fokalpunkt
Computeranimation
Task
Resultante
Prozess <Informatik>
Relativitätstheorie
Besprechung/Interview
Zeitrichtung
Strömungsrichtung
Oktaeder
Ordnung <Mathematik>
Softwareentwickler
Roboter
Resultante
Softwaretest
Energiedichte
Weg <Topologie>
Gewicht <Mathematik>
Besprechung/Interview
Meter
Wort <Informatik>
Computeranimation
Task
Softwaretest
Hardware
Gewicht <Mathematik>
Beweistheorie
Computeranimation
Metropolitan area network
Hardware
Software
Verweildauer
Vorlesung/Konferenz
Projektive Ebene
Mini-Disc
Fehlermeldung
Hardware
Sichtenkonzept
Prozess <Physik>
sinc-Funktion
Systemplattform
Maskierung <Informatik>
Whiteboard
Computeranimation
Datensatz
Dämpfung
Offene Menge
Ordnung <Mathematik>
Kurvenanpassung
Graphiktablett
Standardabweichung
Prototyping
Fehlermeldung
Managementinformationssystem
Hardware
Zellularer Automat
Biprodukt
Whiteboard
Dialekt
Task
Informationsmodellierung
Offene Menge
Rechter Winkel
Mereologie
Meter
Gamecontroller
Notepad-Computer
Vorlesung/Konferenz
Projektive Ebene
Verkehrsinformation
Fehlermeldung
Schreib-Lese-Kopf
Resultante
Managementinformationssystem
sinc-Funktion
t-Test
Fokalpunkt
Computeranimation
Forcing
Software
Rechter Winkel
Vorlesung/Konferenz
Projektive Ebene
Information
Kantenfärbung
Speicher <Informatik>
Ordnung <Mathematik>
Aggregatzustand
Resultante
Videospiel
Datenmissbrauch
Subtraktion
Bildschirmmaske
Datenverarbeitungssystem
Besprechung/Interview
Wort <Informatik>
Biprodukt
Whiteboard
Data Dictionary
Offene Menge
Computervirus
Web Site
Arithmetische Folge
Offene Menge
Datenverarbeitungssystem
Quellcode
Wissensbasis
Elektronische Publikation
Whiteboard
Computeranimation
Fitnessfunktion
Task
Web Site
Offene Menge
Besprechung/Interview
t-Test
Projektive Ebene
Reelle Zahl
Biprodukt
Netzadresse
Computeranimation
Internetworking
Metropolitan area network
Bildschirmmaske
Sender
Eindeutigkeit
Besprechung/Interview
Versionsverwaltung
Vorlesung/Konferenz
Hypermedia

Metadaten

Formale Metadaten

Titel Okinesio – Development of an Open Hardware Activity Tracker
Serientitel re:publica 2016
Teil 16
Anzahl der Teile 188
Autor Zöllner, Michael
Lizenz CC-Namensnennung - Weitergabe unter gleichen Bedingungen 3.0 Deutschland:
Sie dürfen das Werk bzw. den Inhalt zu jedem legalen Zweck nutzen, verändern und in unveränderter oder veränderter Form vervielfältigen, verbreiten und öffentlich zugänglich machen, sofern Sie den Namen des Autors/Rechteinhabers in der von ihm festgelegten Weise nennen und das Werk bzw. diesen Inhalt auch in veränderter Form nur unter den Bedingungen dieser Lizenz weitergeben.
DOI 10.5446/20669
Herausgeber re:publica
Erscheinungsjahr 2016
Sprache Englisch

Inhaltliche Metadaten

Fachgebiet Informatik
Abstract The main concern about commercial activity trackers from companies like Fitbit, Sony, Polar, Garmin and others is that users don't own their data and they don't have control about how their data is used by the companies. That's why we started developing an open hardware and open source alternative: okinesio.

Ähnliche Filme

Loading...