A while ago I posted about lego mindstorms and Pinky suggested going down the Arduino path instead.
Well, now that I've planted the seed you will see that it is a lot more flexible and researching it and making it work is very rewarding.
It took a little while to come back to and start researching properly but it was good advice. I've been able to recycle old electronics for doing random things with instead of binning. Not to mention the pricing of components.
Example project I did a while ago was a networked garage door opener & status monitor (closed/midway/open). I've started working with many types of sensors and bluetooth for streaming data for heavier processing on pc on other projects.
Anyway, this thread was just to say thumbs up - I took your advice and am enjoying these. For the price of a single lego nxt** I've got a large collection of useful parts and many arduino varieties (pro minis, nanos, unos, megas).
** I've think i've actually sunk about 2 grand in parts so far (this includes testing/soldering/cables/connectors/electrical/pcb equipment + sensors/controllers/motors,etc, etc). But I don't need to worry about waiting for them to arrive if I start something new. I had looked at netduino and fez for a bit before I started purchasing but decided that the arduino ide was easy enough to use and support at the beginning was far easier to find. Also cost of the boards, sizes available, etc meant that I could buy many more cross-compatible varieties for the price of a single fez/netduino.
does it work on the mobile network or can you connect to wifi outside your house.
Because i'm using fairly basic security (the buzzer in the video is used for login success/fail and open/close) on it I currently only have it running on the lan via a wifi bridge (in the video the phone and controller are both connected to the wifi network), but all I need to do is forward a port for it from the router and yes I could use it from anywhere.
Instead if I need to control it remotely (remotely as in outside the wifi range) atm I can remote in to a pc in the house and use it to connect to it or setup a vpn (less worries about security doing it this way). I added in device lockout for a small number of failed logins just in case anyway where it won't work until physically reset.
It is possible to add mobile coverage and control it by sms (but that leads to ongoing costs and account maintenance)
i didnt go the full DIY but I am doing a lot of code myself.
With the new marine tank I bought I am using this www.reefangel.com which is powered by Arduino
nah hadn't come across it. I was thinking of updating for a html5 page to make it work like an app on android/iphone and useable on pc
my one will confirm whether the door is actually fully closed or fully open so that it never gives a false reading (e.g. 'closed' when the door is half open). This is done by the disto confirming when the door is fully open (in it's view at a certain distance +/- small inaccuracy) and a magnetic switch when it fully closes. If the magnetic switch does not pass a signal back up the line and the disto doesn't detect the door (door moving or half open) then the page will report 'midway'. A N-mosfet is used to trigger the connection for the door when requested. A spare router with DD-WRT provides both USB power + ethernet for wifi bridge (I saw a page where a guy has used the same router to do Serial to Ethernet which could be a good idea to reduce cost).
Before deciding to use the disto + magnet switch I had set everything up using a rotary encoder from a mouse scroller attached to the main cog on the motor of the door opener.. however it was too inaccurate inconsistent to be able to determine either a pattern or movement required to confirm closed/open.
I have another prototype built which has halved the cost/size of the project by using a pro mini and ENC28J60 ethernet module but will require some work to get it running properly. It would be good if wifi gear for the arduino wasn't so high (i.e. far cheaper to use wifi bridge and ethernet).
I mean having a small remote that is on the cars' key ring, push button, done.
even the bigger older style remotes, a little velco on the dash, push button, done.
iphone app, log into wifi/make sure it is one, open up app, select open, other than a very good back up to the remote system, it hardly seems a practical replacement
I dont actually use remote controls to get in/out of the garage anymore since installing an RFID reader and keypad to open it if I can't get in (dont need to carry keys); and another RFID reader and keypad to get in normal door.
iphone app, log into wifi/make sure it is one, open up app, select open, other than a very good back up to the remote system, it hardly seems a practical replacement
it was more a learning project (+fun) - but the advantage is not so much for getting the door up/down as a remote replacement. If it was ever going to be considered for commercialisation the focus would be more for people who forget whether they've left the door open when going out (or already in bed :) ) - the ability to monitor (and action if required) is where it holds the advantage. Also you could setup temporary access logins for people if you needed them to come do something at your house while away.
The extendability also would make it possible to use a different wireless system that enabled client devices installed in cars to monitor (say with leds or lcd) and control up to a few km (or open the door before you get in the driveway).
Though this form of home automation (and a lot that i've seen) without sinking heaps of dollars in is always going to get 'cool, but isn't it easier to get up and flick a switch then to use your phone to turn off a light' (quote from Koshi when a guy was showing his home automation on sunrise.. his response was 'I could say the same about you and the tv remote?').
What I would like to work on is a security system that talks between components. Using those alarms that go off when a wire is disconnected on all major electrical appliances and having them talk wirelessly to a main host which distributes security actions (i.e. email or sms advising what device is being tampered with) + external sirens say if two+ were disconnected. Similarly with vibration sensors on windows to alert of possible breakages. Monitoring hard cabled devices that thieves may target should work nicely for homes that can't use motion sensors (indoor pets,etc).
i like the idea of a alert that something is out of wack
so, would that extend to, say a pattern learning program (that learns you habits,/routine) to send an sms if something out of order happens?
ie, wednesday at 1pm, you are at work, but there is a light turned on,
I mean the code to sync with your habits, and not to send false alerts at say christmas might take some time, but that would work well with the connected home
I mean having a small remote that is on the cars' key ring, push button, done.
even the bigger older style remotes, a little velco on the dash, push button, done.
iphone app, log into wifi/make sure it is one, open up app, select open, other than a very good back up to the remote system, it hardly seems a practical replacement
i like the idea of a alert that something is out of wack
so, would that extend to, say a pattern learning program (that learns you habits,/routine) to send an sms if something out of order happens?
i guess it could be possible to determine and improve over time - esp. if you have a perfect routine. Working out a sick day or changing annual leave times probably wouldn't work and require some degree of human interaction. You'd be better off with a manual way of activating when you leave as with an alarm (this could be automated using some sort of wireless sensors or awesome RFID readers).
I read recently too about Microsoft HomeOS prototype 'bringing smart homes to the masses' http://research.microsoft.com/en-us/projects/homeos/ which is a bit of security, light control, temp, media, electrical control - one of the features they mention is sensors going off triggering cameras to take photos and alert.
Though this form of home automation (and a lot that i've seen) without sinking heaps of dollars in is always going to get 'cool
if it helps to know - my work's target demographic got in hard with home auto.. and many have been left high and dry as the smaller home auto startups went broke.
Sadly.. I have nfi what to recommend as a carry-on from there.
Jerry your project is exactly the reason why I went and brought an uno and shield in the first place. The issue is practicality, its fucking awesome don't get me wrong and I am building nearly the same thing as you.
But I can simply order an RF Momentary/Toggle Switch from fleabay for all of $15 and re-purpose my my uno+ethernet shield
just saw this, looks like it could provide some fun for $35(+$10 post):
on another note, has anyone seen this kickstarter.com (linked above) site to get funding for projects like this (they are at $160k of their $25k goal)?
In other words, a lot of stuff on the Arduino board itself is not required for this to project to work.
After that you could even design and manufacture your own PCB ( http://dorkbotpdx.org/wiki/pcb_order ) and then you no longer have a prototype - package it and you have the real deal!
Raspberry pi or gtfo
Ridiculous overkill for this kind of project.
I'm very impressed, it's awesome work. So many of my projects don't get this far. I'll start a thread on one I did complete recently for a mate.
My latest project was a paintball competition flag capture timer.
Two teams, one red, one blue. 15 minute game. Points determined on how long each point has been captured. The device I built is weather sealed, rechargable (without being opened), and has a 12V output to power two strobe lights (red & blue). The LCD screen displays how long each team has held the point for.
We had two running for 10 hours straight, and worked flawlessly.
Next upgrade for it is some radio control to reset them between games. Got the radio transmitter and receiver modules from seeed studio last week.
I'll be looking for more details on this, because I want to write an app for Pebble to use it as a garage door opener. The hardest part will be getting the signal from a server sitting in the garage to the motor device.
PM
WWW
