Home Automation - Suggestions, Guides, etc

[coolVariable]

I have started looking into Home Automation and the possibilities seem very cool.
Sadly, the state of home automation appears to be all over the map without any (big) open standards. Resources on the internet are also less than helpful as far as I have been able to find.
In an ideal world, I would like to put together something like the below draft project outline.

Home Automation:
for 4 bedrooms, 1 living/dining room, 3 ½ bathrooms, 1 kitchen and 1 garage)

Control
• 1 windows (or Ubuntu linux?) server for home automation, ideally running a (programmable) website that provides all the options, so that allowed devices on the LAN can access the website and control the home (and I can design and redo the website as I see fit)
• Various tablets and mobile phones accessing the above website, might even install some (android) tablets into walls as control panels

Automation
• Outlets/lights (probably 1 per bathroom, 2-3 each in bedrooms and kitchen, 5+ in living/dining room)
• Audio zones (4 bedrooms, 1 hallway, 2 baths, kitchen, living/dining room, garage - all should support standard play-to for android, windows and iphone)
• Jarvis (microphones in each room to allow for Siri-like digital assistant or something like it)
• HVAC (just basic control for now; maybe closing/opening of vents per room later)
• Floor heating (4 bedrooms, 1 living/dining room, 3 ½ bathrooms, 1 kitchen)
• Security (TBD – assume it will monitor doors, windows, motion sensors outside, infrared inside?, cameras inside and outside, audible alarm (loud via audio zones), and external notification (sms/email for now, probably central monitoring via internet at some later point)

Outdoor
• Garage door
• Motion sensor lights
• Motion sensor security cameras (webcams?)
• Sprinkler system

Video system
• I am a long term HTPC (Windows Media Center) user. The HTPC does not need to tie into the home automation system with the exception of the audio zones. The HTPC in the living/dining room should utilize the “audio zone” speakers there (or vice-versa)


I want to keep my options open as wide as possible, especially as far as control server and control panels are concerned. As I explain below, I would love for the control server to be a windows or linux (ubuntu?) box running a local website via which it is possible to turn of lights, etc. That doesn't box me into some stupid app that will only work via iphone, or android, or something proprietary. Hell, for the price of most proprietary control panels, you can get an android tablet or an ipod. A website, that you can skin yourself, will probably also look much, much nicer.

I guess that means I will need
• 20+ "controllable" outlets for lights, etc.
• 20+ intelligent light switches
• ~10 audio zone controllers
• a Siri-like software for Win7 or linux
• 6+ microphones to hear commands
• an intelligent controller for the HVAC that connects to the system
• 10 intelligent floor heating controllers that connect to the system
• controller for the garage door
• security - TBD
• 8+ outdoor lights/motion sensors

any recommendations?
I have read about x10, z-wave and a ton of proprietary systems but nothing seemed to be "just perfect"

 

[coolVariable]

Guess AT is not into this ...

 

[Puppies04]

I'm sure if you were asking a single question you would get replies, as it stands the pure amounts of variables in your proposed build is slightly daunting for anyone not willing to spend the same amount of time thinking about this that you have. You already state that currently existing market products are not quite what you are looking for so this sounds like a custom "ground up" project. Maybe you should start with the control server and work your way from there.

 

[coolVariable]

I'm sure if you were asking a single question you would get replies, as it stands the pure amounts of variables in your proposed build is slightly daunting for anyone not willing to spend the same amount of time thinking about this that you have. You already state that currently existing market products are not quite what you are looking for so this sounds like a custom "ground up" project. Maybe you should start with the control server and work your way from there.

Ugh? I thought my question was quite straightforward: for a project such as mine, what is the best tech?

x10? (old, gets interference if too many devices, pretty open, so can be controlled with a custom solution?)
x-wave (proprietary but pretty widespread use and maybe can be controlled with a custom solution??)
insteon (proprietary, bad QC?)

 

[Knavish]

Well, if you want to go for the really high end, you can buy Crestron. I've used their hardware for video / audio / lighting control in corporate settings and it's flawless. I think it's also priced accordingly and may only be available from approved installers.

What you're asking for is not unreasonable, but it will be expensive even if you do it all yourself. For example: would you trust a wireless network to control power to lights / outlets / HVAC? If so, you probably have to spend somewhere between $40-$100 per connection to buy some kind of wireless controlled switch or outlet. (I don't know if these even exist!) If you do not trust wireless, then you could install a relay at every outlet or light switch, but then you need to run control lines to all the relays! Talk about a massive re-wiring job!

---edit----

Have you read about ZigBee? It's the most convincing home automation stuff I've seen recently.

 

[MrTeal]

I've done custom embedded stuff with X-10 devices, and they really are crap. When I last checked into it a couple years ago Control4 seemed to have the best overall solution.

If you're planning on implementing it yourself, good luck to you. Obviously it's not impossible, but it will be a massive undertaking of time. I developed an extended range RFID-like system that allowed me to automatically unlock a door when I was within range and lock it when I was out of range, and even that was a big project. Doing a whole how without just buying an off-the-shelf controller is definitely possible if you leverage existing hardware, but it's still going to be a lot of work.

 

[Coltaine]

Take this with a grain of salt, as I have only done research into some of what you're looking for, not actually implemented anything...

If you're planning on doing it yourself, here are the major roadblocks I found when researching:

You mentioned no tech is a perfect fit, well no tech is going to have the inventory of sensors you're looking for either (except maybe X-10 but thats decades old), so your hardware control will need to connect to multiple different tech interfaces so you can hand pick the sensors you want regardless of underlying tech.

The software running on your home server will have to deal with that as well, OTS software from HomeSeer or other competitors can handle nodes of different tech and manage events across all of them. You can buy software that gives you a website interface and SW with android apps, but I haven't seen anything nearly as sophisticated as to be ready to run via voice activation with multiple mics throughout the house.

My advice would be to get your home server with room to stick multiple USB or PCI devices in there later, run some off the shelf automation software that handles multiple sensor technologies, get three or four sensors and get to work on your crazy front end website/startrek computer. Dont sweat the sensors, you can cherry pick them later, the hard stuff is going to be in the interface.

Personally I liked HomeSense for controlling software, no one for controlling hardware, and Z-Wave for sensor technology, but Z-Wave doesnt have nearly the inventory of sensors you would need for serious automation, and HS's front end is going to be clunkier than what youre looking for...

Good luck to you, and post your progress when you get around to it!

 

[sm625]

That stuff is still incredibly expensive, and will remain so for some time. Mainly because there is no point in spend 40+ hours of effort just to save yourself maybe 1 hour. Let's face it, it just doesnt cost that much in terms of energy, time, or effort to close or open a vent here or there. And... worst of all is even if you had this stuff automated, you're still going to take just as much time figuring out how to use the controller as you would actually physically walking to each vent and opening it yourself. Especially once you include all the time you spend downloading app updates, and of course checking for compatibility with your new phone/tablet once you upgrade that. It is a neverending vicious cycle, and ultimately futile.

Until the average DIY person can see a value in this, it will go nowehere and be nothing but a toy for the stupidly rich.