Existing air quality sensing projects

123 views
Skip to first unread message

Ariel Levi Simons

unread,
Apr 1, 2012, 1:42:01 PM4/1/12
to Safecast Air Quality Discussion
There are some new projects out there to capture air quality data,
which would be useful to try to plug into an update Safecast API:

Existing air quality sensor project: http://www.citizensensor.cc/
Another multisensor platform:
http://www.kickstarter.com/projects/108684420/node-a-modular-handheld-powerhouse-of-sensors?ref=card
Tricorder project: http://www.tricorderproject.org/about.html

Janos Gyerik

unread,
Apr 9, 2012, 6:50:19 AM4/9/12
to safeca...@googlegroups.com, in...@ecomobilecitizen.com
In the EcoMobileCitizen project we have a component that you could start using right away:

This is an open-source Android app to bridge the gap between sensors and the internet: download data from sensors and upload to a map server. It is flexible to work with any sensors and upload data in real time to your server of choice. So for example if you build your custom air pollution sensor and make it broadcast data with bluetooth or other wireless protocols, the Android app will give you an easy to use interface to collect the data from the sensor and upload to safecast. (We have our own map, but you don't have to use it. http://blog.ecomobilecitizen.com/map/ )

The vision of the EcoMobileCitizen project is to bring participatory sensing to the masses. Our goal is to make it easy for citizens to measure their environments and collectively build a public pollution map for the benefit of everyone. It's a work in progress. 

We are not a company, the project is not tied to any businesses and organizations, just a bunch of people who like the environment. We hope you will find the Android app useful and we'll be happy to assist or collaborate.

Ariel Levi Simons

unread,
Apr 15, 2012, 6:31:02 PM4/15/12
to Safecast Air Quality Discussion
Janos,

With Safecast we looking to just start aggregating data from a variety
of sources and put them on a single platform. Basically a Google maps
and analysis tool for as much environmental data as we can throw in
there. Thank you for being amenable to sharing data. Hopefully we
can get more folks on board.

Janos Gyerik

unread,
Apr 18, 2012, 2:56:48 PM4/18/12
to safeca...@googlegroups.com
> With Safecast we looking to just start aggregating data from a variety
> of sources and put them on a single platform.  Basically a Google maps
> and analysis tool for as much environmental data as we can throw in
> there.  Thank you for being amenable to sharing data.

That's exactly what our Android app is for: collect data from any
source (= sensor), and upload it to anywhere you want.

On this mailing list I see a lot of discussions about sensors. Then
there is Safecast, which is basically a map website. But how do you
get the data off your sensors and onto the Safecast website? That's
what our Android app can do for you.

This Android app was designed with these main goals:
1. Make it possible to gather data from as many sources (= sensors) as possible
2. Make it really easy to use (and preferably stylish) to entice as
many citizens as possible to collect and contribute data

It should be really useful to bridge the gap between offline sensors
and the Safecast website.

Cheers,
Janos
--
Janos Gyerik
http://blog.ecomobilecitizen.com/android/
http://blog.ecomobilecitizen.com/project/

Sean Bonner

unread,
Apr 18, 2012, 3:19:07 PM4/18/12
to safeca...@googlegroups.com
Hey Janos, I think there's some confusion...

On Wed, Apr 18, 2012 at 11:56 AM, Janos Gyerik <janos....@gmail.com> wrote:
On this mailing list I see a lot of discussions about sensors. Then
there is Safecast, which is basically a map website.

That's incorrect. Safecast is a large volunteer driven organization that collects and publishes data. The map on the website is just one small piece of that visualizing the data we're providing.

 
But how do you
get the data off your sensors and onto the Safecast website?  That's
what our Android app can do for you.

We have a number of databases and custom applications (and APIs) that do just that, many of our sensors are web connected and publish the data directly to us, others are handheld and people upload the data to us directly when they get a connection. We've been very direct about not putting a 3rd party between us and our data, so we manage our data ourselves.

We also publish our data out to the world so other people can use it.

Hope that makes more sense.
-s

 

This Android app was designed with these main goals:
1. Make it possible to gather data from as many sources (= sensors) as possible
2. Make it really easy to use (and preferably stylish) to entice as
many citizens as possible to collect and contribute data

It should be really useful to bridge the gap between offline sensors
and the Safecast website.

Cheers,
Janos
--
Janos Gyerik
http://blog.ecomobilecitizen.com/android/
http://blog.ecomobilecitizen.com/project/

--
You received this message because you are subscribed to the Google Groups "Safecast Air Quality Discussion" group.
To post to this group, send email to safeca...@googlegroups.com.
To unsubscribe from this group, send email to safecast-air...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/safecast-air?hl=en.




--
Sean Bonner

* email is painful so I'm trying to keep messages short and direct. Please excuse my brevity. Here's 6 tips to make email suck less for everyone.

Janos Gyerik

unread,
Apr 18, 2012, 3:58:43 PM4/18/12
to safeca...@googlegroups.com
> That's incorrect. Safecast is a large volunteer driven organization that
> collects and publishes data. The map on the website is just one small piece
> of that visualizing the data we're providing.

You're right, I phrased incorrectly. I meant about the technicality of
getting data from offline sensors to a web service.

> We have a number of databases and custom applications (and APIs) that do
> just that, many of our sensors are web connected and publish the data
> directly to us, others are handheld and people upload the data to us
> directly when they get a connection.

Where are these APIs? I looked for them several times in the past (2
months ago) and couldn't find them. I am completely unaware of your
APIs and documentation, and would be happy to read up if you give me
the links.

> We also publish our data out to the world so other people can use it.

The app just uploads to Safecast. Yes, as we all know already very
well, Safecast publishes out to the world so other people can use it,
I don't know you felt compelled to stress that.

I don't know why are so strongly against a 3rd app in the middle.
Without an app in the middle, all sensors are required to be not only
internet enabled, but to implement your APIs. In our project we're
targeting environmental and physiological sensors, and most sensors
I've seen and work with are not internet enabled. And why should they
be! I think a sensor should be a sensor, and you should not have to
re-implement protocols per device type, it's much better to have a
smart glue in the middle to bridge the gap. Without the Android app
the sensors I work with are very inconvenient to use, and to get the
data from them on the internet you have to have something in between.
A smart phone app is perfect to fill that gap.

Let's say you make a home made sensor at your local hacker space. Will
that come with internet and ready to upload to Safecast? If you can
focus on just making a sensor that works well and can transmit raw
data wirelessly and pass it on to our Android app to pass it on to
Safecast that would be nice don't you think? Don't you want to attract
the widest possible audience? The app I'm trying to introduce here is
made for this very purpose. And it's open-source. It does something
that you would have to do anyway to enable sensors that are not
internet enabled.

Hope that makes more sense.

Janos

Sean Bonner

unread,
Apr 18, 2012, 5:33:37 PM4/18/12
to safeca...@googlegroups.com


On Wednesday, April 18, 2012, Janos Gyerik wrote:
> That's incorrect. Safecast is a large volunteer driven organization that
> collects and publishes data. The map on the website is just one small piece
> of that visualizing the data we're providing.

You're right, I phrased incorrectly. I meant about the technicality of
getting data from offline sensors to a web service.

> We have a number of databases and custom applications (and APIs) that do
> just that, many of our sensors are web connected and publish the data
> directly to us, others are handheld and people upload the data to us
> directly when they get a connection.

Where are these APIs? I looked for them several times in the past (2
months ago) and couldn't find them. I am completely unaware of your
APIs and documentation, and would be happy to read up if you give me
the links.

They aren't public yet but we hope to release them before the end of the month. We are using them internally and with a few testers and finalizing documentation now. 




 

> We also publish our data out to the world so other people can use it.

The app just uploads to Safecast. Yes, as we all know already very
well, Safecast publishes out to the world so other people can use it,
I don't know you felt compelled to stress that.

I don't know why are so strongly against a 3rd app in the middle.

Because we don't want someone else to be a bottleneck. 

 
Without an app in the middle, all sensors are required to be not only
internet enabled, but to implement your APIs. 

No, data can be uploaded any number of ways. From the sensors if they are connected, from the website directly, via email or via any number of apps that will the API



 
In our project we're
targeting environmental and physiological sensors, and most sensors
I've seen and work with are not internet enabled. And why should they
be! I think a sensor should be a sensor, and you should not have to
re-implement protocols per device type, it's much better to have a
smart glue in the middle to bridge the gap. Without the Android app
the sensors I work with are very inconvenient to use, and to get the
data from them on the internet you have to have something in between.
A smart phone app is perfect to fill that gap.

We agree, which is why the cat majority of the devices we've made are not connected as well as the new design we've just released that is going into production is not connected and relies on a smartphone, feature phone or hand entering data onthe website. 


 

Let's say you make a home made sensor at your local hacker space. Will
that come with internet and ready to upload to Safecast?


I don't know of many hackerspaces without Internet. 



 
 If you can
focus on just making a sensor that works well and can transmit raw
data wirelessly and pass it on to our Android app to pass it on to
Safecast that would be nice don't you think? 

You can. Our API allows data submission and a number of apps will be using it publicly shortly. 


 
Don't you want to attract
the widest possible audience? The app I'm trying to introduce here is
made for this very purpose. And it's open-source. It does something
that you would have to do anyway to enable sensors that are not
internet enabled.


I think you are askin us if you can do something that we've already set out to allow people to do. We don't need AN app, all apps will be able to do this. 

-s
 

Hope that makes more sense.

Janos
--
Janos Gyerik
http://blog.ecomobilecitizen.com/android/

--
You received this message because you are subscribed to the Google Groups "Safecast Air Quality Discussion" group.
To post to this group, send email to safeca...@googlegroups.com.
To unsubscribe from this group, send email to safecast-air...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/safecast-air?hl=en.

Janos Gyerik

unread,
Apr 18, 2012, 6:02:21 PM4/18/12
to safeca...@googlegroups.com
> We agree, which is why the cat majority of the devices we've made are not
> connected as well as the new design we've just released that is going into
> production is not connected and relies on a smartphone, feature phone or
> hand entering data onthe website.

Aha! So you already have a phone app in the works.

> I think you are askin us if you can do something that we've already set out
> to allow people to do. We don't need AN app, all apps will be able to do
> this.

I was not *asking* for anything. I was trying to *offer* something you
could use, for free, open-source, something I assumed you didn't have,
having researched the Safecast website not so long ago, several times,
and being a regular follower of the project.

We have developed our app because it did not exist. We developed it in
a way to be as open and extensible as possible, to cover the widest
audience, because we want to enable all citizens to be able to collect
environmental data and build a public pollution map for everyone.
Since our goals obviously have some overlap, and I had no knowledge of
similar work on this at Safecast, I came forward to offer something
you can use.

I get it now, you're already developing your own thing. That's
awesome. I had no idea about that until your last response.

Sean Bonner

unread,
Apr 18, 2012, 7:13:40 PM4/18/12
to safeca...@googlegroups.com
On Wed, Apr 18, 2012 at 3:02 PM, Janos Gyerik <janos....@gmail.com> wrote:
> We agree, which is why the cat majority of the devices we've made are not
> connected as well as the new design we've just released that is going into
> production is not connected and relies on a smartphone, feature phone or
> hand entering data onthe website.

Aha! So you already have a phone app in the works.

Of course, as well as talking with the makers of other popular apps in the app store who be adding our API (some are already testing it) so that data can come from any number of sources and hopes that once the API is public others will include it as well - both to display our data and to submit data to us.
-s

Ariel Levi Simons

unread,
Apr 19, 2012, 11:48:18 PM4/19/12
to Safecast Air Quality Discussion
In general it would be more useful for us, Safecast, to either gain
access to current streams of environmental data or get the sensors out
there on our behalf that could stream the data to our site. An app is
fine for getting data from a device to an online database, but our
current issue is our need acquire a steady stream of reliable
environmental data.
For your app do you currently have sensors linked to any Android
devices? If so could you let us know what the sensor specs are, and
if we could see some example datafiles?

Thanks!

Janos Gyerik

unread,
Apr 20, 2012, 1:39:16 AM4/20/12
to safeca...@googlegroups.com
> In general it would be more useful for us, Safecast, to either gain
> access to current streams of environmental data or get the sensors out
> there on our behalf that could stream the data to our site.  An app is
> fine for getting data from a device to an online database, but our
> current issue is our need acquire a steady stream of reliable
> environmental data.

Our tools (the app and the map) will be finished this year, they are
not ready yet. We do not yet have current streams of data.

> For your app do you currently have sensors linked to any Android
> devices?  If so could you let us know what the sensor specs are, and
> if we could see some example datafiles?

The app is confirmed to work with the environmental sensors on this page:
http://www.sensaris.com/products/senspod/

...and with the health sensors on this page:
http://www.zephyr-technology.com/bioharness-bt

The environmental sensors send data as text "sentences" like this:
$PSEN,CO2,ppm, 0455
$PSEN,Noise,dB,039
$PSEN,NOx,V,2.574
$PSEN,COx,V,1.627
$PSEN,Hum,H,49.48,T,31.22

Cheers,
Janos


--
Janos Gyerik
http://www.janosgyerik.com/

bvirtual

unread,
Apr 21, 2012, 2:39:05 AM4/21/12
to Safecast Air Quality Discussion
Thanks Janos for the lead to this technology. It's impressive.

I read several pages, the 3 sensor pages, including several tabs, and
a downloaded user manual, for CO. I felt it was enough to post my
initial evaluation. Here is the old compare and contrast evaluation.

Compare

These units match exactly what I visualized, in many areas. An
Arduino like chip, connected to wifi, multiple sensors, local storage
for buffering data, for later uploaded to a central web site, or IP
and port number of choice.

These units could easily jump start what I called a 'primary' device
design. We should find funds for a purchase, or ask for a free one,
to begin development, or even one of each, from the firm. Why would
they give us them? The potential of thousands of future sales. We
should find the prices independent of their website web form "Request
for Quote". Maybe 2 of each, to also test the throw on the roof
device.

These units have microphones, temperature, relative humidity, and
other human measurements (I did not read that page closely). All in a
very small and light package.

Contrast

1) 5 hour battery time... We likely need 16+, meaning a larger,
heavier device. I know some sensors are wattage hogs, so we need a
way to cycle them, to preserve power, and yet 'alarm' with quicker
cycles. An external power plug would be nice. The manual does not
indicate if the unit continues to operate when under charge.

2) No expansion capability, other than getting different sensors in a
2nd device, duplicating the CPU, battery, wifi, weight, power needs,
etc. Expanding these means multiple boxes, and multiple
configurations, multiple batteries, etc.

3) No threshold alarm level(s), no alarms - wifi signal or beeper or
cell call or beeper or other, that I could tell. Perhaps via their
website it sets that?

4) Sensitivity might be reduced for various reasons, increased profit,
reduced air flow due to size and lack of forced air intake


I can see generic 'trouble' with any cell phone device as follows:

1) Air flow to the device while in a pocket or purse.

2) Cost - This firm provides no price, but have been in business at
least from 2009.

3) Weather proofing and increase run time (solar?) if it's to perform
for the throw on the roof device.

Conclusion:

Because these units are made 'smaller' to be more portable, lighter,
these units may not work for 'every' purpose we envision, but they do
get a jump start on many aspects of two of 3 primary device types,
cell and throw on the roof (These are not weather proof - do not get
wet).

Peter

Janos Gyerik

unread,
Apr 21, 2012, 3:15:38 AM4/21/12
to safeca...@googlegroups.com
Hi Peter,

Thanks for the assessment, it's interesting info for us.

Keep in mind that we are only a customer of this company. They are
just an example of what our app can work with. I can share with you
our personal experience with them and probably I can answer some of
your questions, but I prefer to do that in a private mail, not on this
mailing list, and only if you are interested.

Cheers,
Janos

> --
> You received this message because you are subscribed to the Google Groups "Safecast Air Quality Discussion" group.
> To post to this group, send email to safeca...@googlegroups.com.
> To unsubscribe from this group, send email to safecast-air...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/safecast-air?hl=en.
>

--
Janos Gyerik
http://www.janosgyerik.com/

Peter Benjamin

unread,
Apr 21, 2012, 6:15:15 PM4/21/12
to safeca...@googlegroups.com
On Sat, 2012-04-21 at 09:15 +0200, Janos Gyerik wrote:

just an example of what our app can work with.

You're welcome.  I'm wondering by "us" if you mean your endeavor or SafeCastAir or both?


I can share with you our personal experience with them and probably I can answer some of
your questions, but I prefer to do that in a private mail, not on this
mailing list, and only if you are interested.

I highly encourage any and all evaluations of "hands on" with brand name equipment be done in private.  Later, a collection of brand  evaluations can be made anonymous and aggregated into an enhancement list.  This method preserves the bridge with the brand's firm.

Yes, let's have a private thread, following the expectations below.

My evaluation was not "hands on."  My evaluation was made from a few web assets, with limited published information.  I hope the firm might see my evaluations, and address one or more of the issues.  Even give us a time frame for addressing all the issues, even if in terms of years.  I do realized some issues will be negatives for the firm, reducing their profitability, and they may want to keep their reasons to themselves.  That all said, I'm all ears.

For example, making the battery last longer, bigger, heavier, bulkier might impact sales, if not an option.  Having a larger size impacts manufacturing and the price on initial, low volume sales likely is too heavy a burden for a firm entering this new marketplace with unproven demand.

I'm about increasing the "demand" by increasing public awareness of dirty city safety issues, including the dangers of modern home weatherproofing designs, that seals in toxic or poison gas, allowing it to accumulate, at first causing minor health issues, then killing pets, major health issues, hospitalization and death.

Peter

Peter Benjamin

unread,
Apr 21, 2012, 6:25:34 PM4/21/12
to safeca...@googlegroups.com

I've started a table of gases, to have other sensors added, where for each sensor, additional information is listed about other names, and the "application."  The application is a list of purposes, roles, and details.  It's looking like details needs to be divided into "who", and "why."  I envision additional columns of brand, range, price, supplier(s), and more.

I've started designing a group bulk purchase Drupal Feature for my robot club.  The idea is to test it in the robot club, and expand it to other LA robot clubs.  I feel the sensor project falls into this area, as many of the parts we need do match robot needs.  Also, it would be great for Crashspace to have a remote sensor bot.  Each 'finger' specialized for the sensor (gas appliances need a long metal tube to suck in hot gases, reducing the temp), fingers could be "cleaned", and other kewl features.

I'll post the table when it has 20 gases and 10 non gas sensors, likely middle of next week, or later.

The goal is understand, rate, prioritize, how 'big' a plug in bus is needed, to support 'sets' of sensors, and what sensors would be in common in all sets, thus be built into the 'bus'.  How many plugs needs to be in our bus(es)?

Peter

Reply all
Reply to author
Forward
0 new messages