Hi,
I’ve been working on a home automation project for few months now and made some progress towards building a prototype.
it involved stuff like Arduino/AVR, 3d design, some wireless devices, some pcb design etc.
I am looking for a possible co-founder who would love to join and help taking this to the next stage. There is not really an NDA or anything involved and I will love to discuss in detail if someone is interested in it.
-Arif
What do you hope your co-founder would do? Fund-raise? Manage? Design? Look good for the camera?
Let me think out loud a little bit.
The next few technical milestones i can think of right now will be these:
I am finding the AVR to be inadequate for the codes and port the whole thing to an ARM based system. (not a lot of work, especially if I stay within ATMEL SAM processors). However, all the codes need to be tested and so a little legwork required.
The radio modules I started with, were easy to use but expensive. I’ve zeroed down to a few which I can probably use to get the cost down.
The 3d desings / printing I have need to be tweaked into something more presentable.
Need to work on the app and the server side. I have some basic codes running and working. definitely not the most efficient.
Need to some stress test as well.
Based on the current situation of the project, fund raising or getting some traction is important before jumping into doing all the improvements. I do not have much experience in this so can’t really tell what exactly I must do and thats what i am reading on. Having someone with some experience in this will probably go a long way. Currently am working on to get into / find an Accelerator program. Anyways this part is sketchy at this point and I definitely need help.
I hope that gives you an idea of what expectations are. Any ideas are welcome.
Try the site that was developed by two DMS members; ventrifuge.com It allows you to post your project and look for co-founders and partners for projects, it would help them out to have more projects posted.
Thanks Mitch. I will definitely try that out.
Home automation is my current focus as well. What sort of stuff are you making? A bit of my progress is documented at http://openlabs.co but I left out the temporary light switches and outlets which would never pass and weren’t meant to pass any applicable regulatory testing, and I’ve yet to blog about some recent changes in hardware and OS direction. Here’s the light switch: (the outlet is just a light switch rigged to an outlet; it’s not pictured)
I’ve settled on 6LoWPAN over 802.15.4 running on EFM32. My prototypes all run Contiki on LPC11 but I just discovered RIOT OS and am about to port it to EFM32 hoping that it’s ready for use. I use a Raspberry Pi for the 802.11<>802.15.4 bridge.
What radios were/are you using? I’m confident that 6LoWPAN over 802.15.4 is the way to go. I had to wait a few years for standards to lead to implementations and whatnot, but finally there are several OSs with usable 6LoWPAN stacks and it was totally worth the wait. Pinging the IPv6 address of my lab lightswitch from my laptop while sitting in Starbucks is a sweet thing indeed. 
Software is basically the one thing I’ve given no consideration to yet, and I’m really dreading it. I mostly do hardware. I am many things but I am not a software designer. Currently I control my lights from my phone with a dead simple RPi-served web page consisting of a few ugly buttons and no thought whatsoever given to UX. It’s bad.
Oh, and I don’t know what the hell to do about all the criminally expensive regulatory certification. Until now I’ve been content to use the stuff I make in my own home at my own peril, but joining DMS has made me realize that I’m going to want to outfit the space with this stuff, and it would be inappropriate to do that without proper testing even if it were legal, which I’m sure it isn’t in a public structure.
Are you at the space often? I usually have my prototypes with me there. I don’t normally team up but we should still talk anyway.
I am working on a new way to control the HVAC. Trying to stay within modules (both 802.11 and 802.15.4) to avoid the certification costs . 6lowpan, although would be perfect, is not quite there yet as 802.15.4 IMHO (enlighten me if I am wrong). Also I am yet to test 6lowpan vs 802.15.4 for battery operated remote sensors and actuators.
I dont either, have much idea about software, but I am trying to get hold of a programmer. I have an ugly android app running with amazon ec2 server. But again no idea how to scale this up.
I am not regular at the space. But do let me know when you’ll be there and I can go over. I would love to talk to you. you can also call/text me at 817-323-7975.
-Arif
Thomas, will you be at dms anytime soon? let me know if i can meet you.
I do try to go most nights but lately I’ve needed to catch up on work. I think I’ll be there tonight but I’m still waiting for a mouser order to show up. Usually I stay until midnight or much later.
I’m not familiar with every 6LoWPAN stack but for the most part I think they’re being developed alongside an 802.15.4 stack. You can run 6LoWPAN over bluetooth too but most I think are running it over 802.15.4. Contiki’s stack is full featured as far as I know. Linux’s is usable but missing some features like beacon support. I don’t know about RIOT but I’m hoping to find out shortly.
Battery operation does strongly indicate using 802.15.4 beacon mode though and I haven’t tried 6LoWPAN like that yet, but I do intend to do it at some point. One of my next projects is to be a CR2032 powered climate sensor and since a 12 hour battery life obviously isn’t acceptable it will need to use beacons.
I will catch up with you tomorrow evening. Thanks for the input.
In my project, using 6lowpan will simplify lots of stuff but I am not well learnt yet about using that for sleepy devices. I can discuss that when i see you.
thanks for replying
-arif
For a programmer, I might chime in and see if this is helpful. I have integrated lots of custom control interfaces using National Instruments/LabVIEW toolset. I have never implemented the same thing twice. The power of the toolset lies in the ability to leverage existing pieces of low-level code written, debugged and tested already, and then stitch them together at a high level with prewritten graphs and controls and what not.
I could envision a control interface with options for viewing multiple cameras simultaneously, taking data from sensors, and sending signals to controllers, all managed by a user interface with some custom code behind it. Logging to database is straightforward, and providing control interface for android can also be done.
If you are looking for a programmer, and you know the starting set of hardware you want to interface to and over what interfaces, maybe its worth considering using this powerful toolset to get it done…
-James
Thomas, Are you at the space now ?
Yes I usually hang out in electronics.
Sorry was talking to someone else. I am here now in the electronics room.