Introduction
In the last 3 months I have done a bit of thinking about an as-simple-as-possible and as-small-as-possible "follow-me"-box. The design goal is to have something so small that it can be e.g. attached to clothing by velcro and so simple that it's just switch on and go. After some googling, I stumbled upon the Arduino Fio which seems ideal for this kind of thing. I am primarily maintaining the information on this project on my personal website.
Unfortunately, I haven't had time to build one yet plus I'm lacking a functional APM-based drone at the moment, but here is my preliminary design:
Synopsis
ArduBeacon is a simple GPS beacon that locates it's current position via a GPS module and sends MAVlink waypoint messages to an ArduPilot-equipped UAV.
Status
Draft
Components
- Arduino Fio
- Mediatek GPS module
- XBee pro module
- LiPo battery
Comments
Interested in doing this as well. Wondering if anyone has working hardware suggestions.
Hi, i'm interested in this project, currently i'm working on it for a school project, i have some problem on how to start, for now i have a quadcopter with crius board from hobbyking, with MegapirateNG 2.8 R3, i did the tuning part and it is ready to fly, the next step is to make a ground station with an arduino borad to send waypoints to the quadcopter...
after some research i discovered MAVlink witch is a message that APM can inderstand.
i'll appreciate any help :)
Any news about this?
Pat, I just realized that we could add a FollowMe mode to mavelous very easily.
Take a battery-powered wifi AP like the 3.4 ounceTL-MR3040, put DD-WRT or OpenWrt on it and then use your iPhone or Android phone as the GPS source using the HTML5 geolocation API.
It might be tight getting Python running, but if that was too much of an obstacle I think we could do a C/C++ version of the mavelous server in about a week (I've written web apps in C++ before).
I would have to run some calculations on the actual circuit... I have followed several forums that show wildly different output based on the design quality. Ground plane, good vias and placement of components will affect final range. I am shooting for 300 M LOS. I agree with using a good antenna.. My original intent was to track desert racing trucks so power would come from that source. If this is going to be adapted for small package, turn key beacon it should be packaged with a charger and rechargeable battery.
As a failsafe I think last position loiter, or even return to home could be added to Ardupilot. Especially with the crowded nature of 2.4GHz signal loss is nearly unavoidable.
It doesn't say the RF output power. There should be a good power reserve to overcome potential interference (powerlines...) or unintended "losing" of the UAV, e.g. if for some reason the beacon moves faster than the UAV. 100m LoS range is not awfully much for an object that's moving in the air.
For the same reason I would definitely forget the chip antenna because if you put the board into a box and velcro it to your sleeve, your body will already absorb a good part of the - already weak - RF output.
I am personally more a fan of a reasonably universal solution as opposed to modular, at least in these areas we are talking about at the moment. Having one more or less universal board keeps production and development costs low and is easier for the user.
Here are a few specs...
128K Bytes of In-System Self-Programmable Flash (bootloader will take up a little of this)
- 4K Bytes EEPROM
- 16K Bytes Internal SRAM
• Ultra Low Power consumption (1.8 to 3.6V) for Rx/Tx & AVR: <18.6 mA
- CPU Active Mode (16MHz): 4.1 mA
- 2.4GHz Transceiver: RX_ON 12.5 mA / TX 14.5 mA (maximum TX output power)
- 38 Programmable I/O Lines
- Master/Slave SPI Serial Interface
- Two Programmable Serial USART
- Byte Oriented 2-wire Serial Interface
Maybe not the most powerful RF output, but I intend to use this for a close following quad, maybe a fixed wing with camera tracking and updated loiter. I have a similar circuit to this for another project and could probably integrate and nominal cost... https://www.sparkfun.com/products/10217 - the IC would add about $.60 to the final cost (not counting resistors, led charge indicator, and caps) My first reaction would be to keep things modular and cheap to manufacture.
I personally would want to add a RP-SMA connector (its about same price as chip antenna) but size constraints may turn people off... feedback?
While $20 to $25 is a little low, especially when doing small runs, I could probably manufacture, these for about $35 assembled.
If you add the connector for the 1S LiPo and the USB-charging circuit plus another connector for a (3dr?) GPS-modules and stay in the $20-25 range for the ready board, that would be some serious followme-box!
I would also consider an U.Fl connector or even RP-SMA instead of the chip-antenna.
The XBee-socket of the Fio has so far the advantage that it could also take WiFly or BlueBee modules, which gives more options in regard to programming. Another advantage is the possibility to use 63mW Xbees with RP-SMA connector which give some serious range with a nice antenna.
How much RF output power does the ATMega128RFA1 produce?
Gentlemen,
I have been working on a similar design, but decided to build mine around the ATMega128RFA1 chip. This has a few advantages (mostly price) but would mean a whole new hardware design. The ATMega128RFA1 has built in mesh networking, (similar range to XBEE) lower power consumption, and in quantity can be purchased for less than $7. All of the development boards have a very heavy markup:
http://www.karlssonrobotics.com/shop/atmega128rfa1-development-boar... - $55
https://www.sparkfun.com/products/9734 - $55
I have started working on the basic circuit and libraries to make it 100% Arduino compatible. (adding FTDI, voltage regulation and chip antenna) and could share designs if it means project moves further. Since its a Mega, it has plenty of horsepower to run this code and pins for sensors....
The footprint is pretty small, so a separate module can be designed to be XBEE socket compatible.
I will share what I have, and also have manufacturing capabilities if anyone is interested in getting a few prototypes made up.
@sergei
by now i try to simply use MP (only supports NMEA) to send the waypoint to my rover or quad but i have to dig in little deeper i think to get a idea about other possible options.
got an old apm laying around and was already thinking what to do with it :)
-
1
-
2
of 2 Next