Solo Parachute System

[Pedals2Paddles]

No, I don't think so. Sorry but no human can recognize the motor failure happened, process it in your brain, switch flight modes, and take "evasive actions" to stabilize it in the fraction of a second it takes for a quad to tumble on motor failure. In order for the pilot to even recognize it, the thing is already keeling over. That's absurd. The flight controller doing it without human input, MAYBE, depending on a lot of factors. Most of those factors are only perfect in lab or demo, like being perfectly balanced with no payload. But more than likely, it's not going to work using today's tech in a real word practical scenario. The only really stable in flight motor failures I've ever seen require 6+ motors.

A gust of wind is mostly lateral force, applied evenly across one side of the aircraft frame. Not a vertical force pulling one corner down (gravity) and another vertical force pulling the opposite corner up (thrust), all from the far end of a lever (the arm) giving that force mechanical advantage. Wind is not even remotely close to the forces and responses that are in play with a motor/prop failure. That is not an assumption, that is just physics.

 

[Jubalr]

Sure, in a small nicely balanced quad with no paylod, taking off from a the ground, in a controlled environment. Attach a 4S battery and gimbal to that thing and see how well it works. Kill a motor or break a propeller in flight and see what happens, because that's the circumstance that needs to be addressed. The moment a motor or prop fails in flight on a quad, it will tumble over. I don't see there being time for a human or flight controller to react to that.

The principle of sacrificing yaw control to maintain rough attitude control is great. It would need to switch to headless control (simple mode in Arducopter). A motor failure on a octo, hex or X8 or Y6 can result in this, and maintain control in flight. I just don't think the FC or human pilot can recognize and react before it is unrecoverable in a quad..

This is a video of my old quad losing a propeller in flight. I think the hub failed. It went from stable to tumbling over instantly.

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.

Accept third party cookies

The video I was looking for, but didn't have the time, was where they disable 1, then 2 motors while in flight. I agree that adding weight and a gimbal would make it work harder. Maybe to the point where it's more a controlled 'crash' than controlled flight. But my reason to look for the video was strictly in response to your statement:
"How do you propose some code changes make a quad remain in controlled flight with one or two motors failed. Remaining in control with two out is a fantasy. Remaining in control with one out is not just a matter of adding some fancy code."

 

[RichWest]

And then you have an issue of thrust, could solo stay aloft on only three motors?

I think the truth in the magic is with the magicians.

"Any sufficiently advanced technology is indistinguishable from magic." - Arthur C. Clarke

 

[PdxSteve]

The goal (in my eyes) would be to bring it down ASAP for a soft crash landing/minimal damage as opposed to just free-falling out of the sky with no control due to a MP/ESC failure. The processor in the Solo is perfectly capable of making these realtime calculations and determining if it there is a MP failure and compensating and landing asap. It's not really different than detecting if the battery voltage is getting low and kicking it into RTH or any of the other myriad of processes that the solo is constantly calculating, assessing and acting upon. No it wouldn't work for all scenarios/payloads, but they could do a model of just the solo, and a model of the solo+gimbal since those are known and the solo can detect the gimbal and act appropriately.