-2

I have been coding embedded systems for (*cough*) decades now. Mainly telecomms & satcomm, with some telemetry & SCADA. I can also produce Windows, Linux & browser based apps and have good database knowledge.

Sound like prime IoT developer material? I also have some free time, so, to keep me out of the pub, I would like to start a project.

I would prefer something which grows in phases. Maybe develop the server, then the clients, then some browser-based reporting, maybe some test tools. Perhaps a phase one with minimal functionality, then phase two adding more features, etc

I would like something to occupy my evenings and weekends for months, maybe years. I am undecided as whether to develop something open source, or something where I stand a slim chance of turning a shilling.

The one Teensy flaw in my grandiose scheme is that I haven’t go the faintest glimmer of an inkling of the beginning of the kernel of a close as what to develop.

Does anyone have a suggestion for me (preferably not involving hairbrushes)?

  • 5
    I have voted to close your question with a reason borrowed from Robotics.SE: "Life Questions are off-topic. Questions about choosing how to spend your time (what book to read, which class to take, what robotics project to construct, what career to pursue, etc.) may be about difficult decisions, and they are often important, but they are too specific to your own situation and are unlikely to help future visitors to the site. They would be better off asked in [...] Chat [of Things]." We do not have discussion about it on Meta, I will open one today. – Bence Kaulics Feb 23 '17 at 11:30
  • 1
    Do you have any suggestions on how this might be reworded to become an acceptable question? – Mawg says reinstate Monica Feb 23 '17 at 15:54
  • 2
    @Mawg: I think this question naturally doesn't suit the Stack Exchange format very well, since it's more about discussion and advice than objective answers. I wonder if a question along the lines of "How can I determine whether device _____ can be automated to do ____?" might be near what you want. – Aurora0001 Feb 23 '17 at 17:09
  • 3
    @Mawg Both answers so far demonstrate this question being difficult, but might help to re-shape it by directing away from those. It seems you're looking for an existing hardware platform, as open and extensible as possible, where you can develop some re-usable software infrastructure. I feel the open side of the market is not mature, despite things like openHAB being well established. No doubt companies are working on the things you describe, but except for niche applications (I heard cranes on building sites) these are early-adopter toys. Pitch a place in the stack, ask specific questions. – Sean Houlihane Feb 23 '17 at 17:41
  • 3
    You should be specific in some aspects, I think. You have to come with some basic idea yourself to narrow things down. It is not that your question cannot be answered, I want to do like a dozen stuff myself only I do not have enough time and I could suggest you all of them. I could suggest you to design a smart pet collar but maybe you do not have pets or an automatic plant watering system but you do not have a garden. There surely are things for example in your home you want to improve, pick one, think about it, come up with a prototype design and share that, so we can suggest improvements. – Bence Kaulics Feb 23 '17 at 17:43
5

This is a very wide question, but here is something I plan to do in my future house: make it smarter. Bonus for you, this may be a never-ending process:

  • what about a smart temperature management? You can set up some heat sensor in various places, then connect them to a central server. Then, find a way to know when your heating device(s) is(are) on, and make your system learn how the temperature where sensors are reacts to heater on/off. Now, you can save your habits into your server (wake-up at 7am, leaving for work at 8:30, going back at 5pm etc...) and set each temperature you want when present/absent. The ultimate goal is to make your system start heating just when needed, to wake-up in a comfy environment at 7am and let the house getting cold when absent, etc... You may need to find a way for your server to activate each heating device
  • in addition to smart heating, you can set up automatic shutter opening and closing. Winter time? Open the shutter when sun is up to heat your house for free, and close them when sunset to not let any degree get away via your windows. Summer time? Reverse the principe: protect from external heat at day, ...
  • you can set up smart access to your home too, like an automatic garage door
  • and so on.

To do so, you may have to:

  1. set up a server
  2. create a link between it and every sensor/actuator in your house
  3. create a client to monitor your house (temperature efficiency, time with shutters closed/open, etc...)
  4. imagine algorithms for the smart heating part and for shutter commands (and internet requests for sunrise/sunset time)
  5. make statistics and every data you want
  6. take a look on this question for informations about how to store your data on the cloud

What do you think about that? Let me know!

  • Persoanlly, I am dubious, as I don't think that smart home technology is advanced enough yet. However, as you describe it, it does sound like a good project. On balance, I think that I will wish you look & keep looking, though – Mawg says reinstate Monica Feb 23 '17 at 10:21
  • 3
    You can find here a already working product on smart heating. In french, sorry about that. I have to admit the idea of smart heating wasn't from me. But I find it more interesting to do it on my own. About the "home technology is not advanced enough yet" part, this is true if you talk about what you can find in your electronic store. But doing your own system, you can make it more advanced. No one need a beautiful automatic thermometer if you can't connect it to anything. This is what my idea is about: connect everything togother. – le_daim Feb 23 '17 at 10:38
  • I think the point is that I want to develop somethign that will be used by others. Not many will conicidentally have the same hardware as I. – Mawg says reinstate Monica Feb 23 '17 at 11:10
  • 1
    Aaaaaand you pointed out the reason why everyone talks about IoT, but without any product being as of today in every single house. Too much hardware, software, needs, without any international standard (except communications (wi-fi, bluetooth,...)) to allow wide spread of any product. – le_daim Feb 23 '17 at 13:43
4

Another approach is to buy the components and make them run,and then you decide what do to with the them. Surely you will come up with ideas while developing

You can start with one of the adfruits kits.

https://www.adafruit.com/

Adfruits for raspberry pi

Once you can control the I/O then the path to the remote control from the server is quite straightforward. You start turning On/Off LEDs then you can add relays and power something bigger, a Fan, Lights, motors, water pump... etc

Not the answer you're looking for? Browse other questions tagged or ask your own question.