I’m not trying to argue with you. I’m just stating the facts and letting you know what alternatives are already available. I’m glad you were able to make the soft switch work. I have been flying an autonomous drone for $100. Soon everyone will be able to purchase one. Is that what’s coming next? I don’t understand your meaning.
Replacing the MCU in an Imp with a 4G modem, is like replacing an AM radio with an FM radio. The only thing that stays the same is the speaker, since the rest is in the MCU. I wish I could think of a better analogy. It would be like replacing an old AMPS phone with a CDMA main board. There’s nothing left behind. One way to reuse the Imp MCU in your design is to connect the 4G modem via Serial port. Which is exactly the same as what I’ve done. There’s no need to make the Imp MUCH bigger for everyone if only some people need this connectivity. Please come up with an idea that is an improvement on what already exists. I feel my goal here is to let you know what is already possible. Then maybe you will see what it is you want.
Why remove the existing functionality of Wifi from the Imp? It is only the size of a dime or less, taking no power when it is off. If you want to add a 4G modem, then add a 4G modem yourself! It only takes a minute of effort. 2G is good enough for me. I’m not going to argue with you about that.
The biggest part of cell phone technology is the antenna. I assume you mean to develop a separate module from our Imp. Obviously the antenna would be external. How is this so different from what already exists?
I’m curious to see the code you wrote to wake the ZTE. It is not as simple as bringing a pin high and low. Maybe I have already solved this for you last year. You must be able to detect when it is connected.
Imagine Electric Imp making a GSM version instead of wifi. A breakout board with imp and sim slot and antenna connector. No wifi. This could be used in many applications. The only reason to remove the wifi from an imp is just because it wouldn’t be used. This is less cost for manufacture, less BOM, so on. I love the wifi imp, no doubt it is perfect (well for next to all applications)! And yes, a 4G hotspot is just peachy. Works great. I just think there are certain benefits of doing this internally through Electric Imp (cost, FCC certs, support, etc).
The reason the imp system is promising for completely wireless data transmission, is they have already taken care of the issues with wireless communication. No DHCP, no dynamic IP, so on. It simply connects and the same url can talk to it. No doubt the arduino and GSM shield can do this, however the coding can be reduced by a huge factor, simply because Electric Imp has already created a solution for that. Having an imp internally rely on GSM (instead of the wifi now) means you don’t need any special skills to get it working. You insert sim and go to the url and execute HTTP requests. It’s not the hardware, its the firmware that Electric Imp has provided that makes it promising with GSM.
And yes autonomous drones are which I speak of. There are government programs (similar to the x-prize) concerning the development of autonomous drones that can wirelessly communicate data (GSM of course) that are cheap and so on. Electric Imp has a huge potential to fill this gap. We just need a simple breakout board that incorporates everything into one module, leaving only the antenna and sim to the user. And it’s not just the military. It is also the amateur division. Real estate, law enforcement, etc. A drone that has no bounds is desired. And yes a hotspot + Imp can do this now, but why not incorporate the two? People will buy it! I know I would…And i’m sure you would too…
I hope you see what I mean!
And yes the ZTE wake is not as simple as I originally thought (your earlier posts kinda made it seem so). However I was able to realize the proper protocol to make it happen. Still tweaking. Most likely the same path you took.
Considering your first 2 paragraphs of ideas…
Why do you want an Imp to be part of an Autonomous Quadcopter? Would you control the craft over Wifi/4G? The definition of Autonomous is to control oneself. Would you send video back? Not the best choice. Not sure what your goal is. I have combined manual control with waypoint guidance in a seamless way so all you have to do is turn on the transmitter to control it a mile away. I have also implemented auto-north, auto-altitude, position hold better than GPS can provide. You can choose any of these modes independently from waypoint guidance. Most of my clients have been private investigators. Either accident or divorce related.
I’m happy to learn from you and share ideas. What would be the advantage of an Imp-controlled Quadcopter? What kind of data would you want back immediately? What data requires 4G speeds instead of 2G? I have noticed that up here in the North Woods, the 2G area is dominant and 4G is very limited.
Well control isn’t what I am aiming for. As you stated it wouldn’t be autonomous then. However being able to remotely upload new waypoints or missions during flight is the goal. Also relaying the telemetry data back from the uav is a goal of mine. I currently have my APM 2.5+ telemetry out (usually for xbee or 3dr radio) hooked up to the imp. I am working on having the imp relay the telemetry back to the internet (my ground station). This offers real-time tracking, not only for just GPS coordinates, but also altitude, attitude, etc. My goal is basically to use the imp as a Mavlink gateway. I have alrady implemented this kind of wireless transmission using a pandaboard and a huge Java program, along with a 4G hotspot. It works, however the pandaboard is $300, and the Imp is alot cheaper, smaller, efficient. The imp code would be roughly 100-200 lines of code, and my Java is ~12,000 lines so far (this includes the GUI for the ground station - BTW, I chose Java because it is cross-platform [Windows, Linux, etc] independent). Electric Imp has already solved many of the issued I ran into, and it’s cheap!
4G, 3G, 2G, whatever is fine. However the quicker the internet, the faster I can send data back, and hence a higher update rate for the ground station. The goal isn’t to control the quad, simply be able to see the flight dynamics and mission critical information. Not to mention the ability to send auxiliary sensor data (infra-red, sonar, tracking algorithms, etc). Basically any quad with with sensor data could benefit from this. Think about a micro quad that takes Geiger readings. These could be sent into danger zones and relay the data it collects real time.
I am hoping to finish this project in the next few weeks. Even one update a second by the imp would be awesome! The Imp is just a gateway to completely base-independent drone capabilities! I want to take my plane and fly it a while, land on some lone road, charge batteries for a while, then take back off! Cross the country like this! (I would never do this though, as it would violate the current FAA regulations regarding amateur UAV usage!!)
Wow! Interesting. We have much in common. Even radiation detection. I’ve been using this and some old units from the 60’s. I started with RACES in the 80’s, my Dad is the Civil Defense director for our county.
My recent project is tracking birds with GPS and a GPRS board over 1000’s of miles. It’s got a tiny battery and solar cells that are 30% efficient, that’s double the normal cells. Each cell has 4x the voltage enough to charge a battery. Having a finished device even 4x the weight of an Imp would be wonderful! The GPS is already smaller than a dime. Besides weight reliability is important. The signal needs to continue for at least a year to capture the birds, otherwise they will die. It’s literally a matter of life and death. Much effort is being put into the coding. Just found a 50 day bug that would cause an endless loop…
My next project is making the Quadcopter code work with my plane. Great minds think alike! Did you read about an R/C plane in the 70’s flying across the country?
No I haven’t seen that. Will look for it though. Have you heard of the Ardupilot system from DiyDrones? Sounds like you wrote your own code for your quad. Interesting! Your bird tracking/survival project sounds intense. Lots of things to consider. Must be a big bird!?
Yes read about Ardupilot. I use 2 microcontrollers, the KK board, and Arduino between KK and receiver to slowly change PWM positions. KK 2.0 can potentially handle both. It works with the Arduino IDE. The birds are medium to large Raptors, they have no problem flying with it. Your idea could make it even smaller? The code to send data to an FTP server from Uno really isn’t so bad. In fact I wish I could do the same thing with the Imp, but it must be done via HTTP? Any ideas how to code that?
Let’s start a new thread and make it happen!
Certainly it’s possible via HTTP. Is there an alternative to get to FTP?