Hi guys,
I've read several threads here in "my project" section and just wanted to share my project of preventing flyaways of the drone with you.
I was happy owner of DJI Phantom 2 for two years till the time when it tried to flyaway to China for some reason. Hopefully, the God was supportive to me that day and my drone crashed not so far, so I could find it quickly. But it was totally broken: motors, gimbal, arms and so on.
After that crash I thought that there's not much control over the copter and started to lookup for safety systems. There were only two ways to save the copter - install radio beacon and install the parachute.
But the problem that no one knows how much time the copter will fly after disconnect from RC (it may be 5-10 km) and radio beacon's distance commonly is around 200-500 meters from the ground. So even it is cheap way, but it doesn't work as well as I wanted and the chance to lose he drone is still very high.
With parachute system I could stop the drone and eject parachute, to prevent falling down, but the problem that I need my RC to be connected to stop props and eject the chute. In case of flyaway I can't do this, the drone does not respont to sticks.
After doing some research I found that there are no other way to stop flyaway, but to have separate rc with alternative working link and ability to control the electronics even it doesn't answer. Also there were other things I wanted to solve, such as light weight, avoid problems to navigation/gyro/accelerometer, and so on.
Since all popular frequencies were busy (433, 468, 1.2, 2.4, 5.8) I have chose 868MHz which is available in most countries and still have long range with rather small antenna.
Discussed with my friend technician and after some time he did the job, I've got two prototypes of first Stop Flyaway system. Transmitter uses MRF as radio modem, and receiver has MRF and high-load keys, which can manage peak loads upto 190A.
The main advantages which we have achieved are the following:
- Works up to 3 km distance, so I could stop the model even it flown away out of direct view.
- I can switch off the controller and activate parachute system, emergency locator and other recovery systems.
- Receiver can manage huge load upto 190A in peak.
- Receiver is light (16g only) and I can use it in most aircraft models, it doesn't affect the whole weight of the drone.
- Independent connection channel lets me to control my model even the main controller is damaged, frozen, doesn't recognize original transmitter's commands.
- Receiver is absolutely safe for GPS and other navigation modules, it doesn't transmit anything.
- Very low power consumption do not affect the whole flight time.
- 868MHz frequency is free to use and does not cross any of drone's usual communication frequencies.
Looks perfect, uh? :)
Finally I've wrote software for these devices (forgot to say that I am software developer in the past) and project became alive.
It works the following way: in normal mode (from the start) the power from the battery flows through the receiver's key and continue flowing to the drone's electronics. When I press button on my transmitter, the receiver disconnects the power from one output (which goes to all drone electronics) and switches on another output which I can use for my own needs. So with one click I do three jobs: power off the failed electronics, switch off props, and eject the parachute, which must be connected to second output.
I have installed this device to my drone, my friend's drones and produced several devices to share with my other friends copter owners and other drone owners who want to have same alternate safety button.
It was not Kickstarter or whatever else project, I did this for my own money as a device for myself, but people liked it.
Hope to upgrade the firmware soon since I have some new ideas how to improve the functionality and usability, will post later here with the news :-)
Any questions and advices are welcome!
Eugene
Replies
added video with install to Phantom 2-3
As you may know, ArduCopter and maybe some other serious controllers already have geofencing with option for force land, or parachute.
Even better, Pixhawk has dual sensors, and run EKF that can detect and avoid using bad data, or handle situations like that.
Unless user/builder have made huge mistakes, a flyaway is usually a toys and DJI experience.
Yes, DJI has much more issues than other controllers have. I actually did Stop Flyaway for DJI drones and thought that after releasing PH3 and Inspire they fixed everything, but as youtube shows their copters still fly away.
Same with other controllers. I see that it may be also not controller problem but receiver or transmitter issue. Just found new post here, will be looking forward what could happen with his drone.
At least, in any such cases separate button and ability to do something is better than nothing. There's special website lostdrones.ru, they collect posts about flown away copters, planes, wings and so on. And I see that not only DJI are lost. So I think Stop Flyaway is still usable thing.
the issue you linked to seems like a classic bad-build issue , bad mounting, too much Z vibrations.
this is what i was talking about. not only controller issue can cause uncontrolled actions in the air and fly away.
the more posts i read the more i see that stop flyaway module can help users to save their copters.