Use cases for disarm in flight for arducopter?

[witness...@gmail.com]

Hi All,
   
      the case was proposed for a disarm in flight for arducopter /APM  when sent from CGS/Tower instead of present behavior.

I guess my question is with the new flight termination system Rob is designing,

what is the use case for disarm while in flight  on arducopter given a   new flight termination system  and the fact a disarm in flight  for arducopter effectively is a crash without  inflight recovery as opposed to a motor stop without disarm(which could be quickly reversed before impact)?


      interested parties want to know

       hzl

[David Pawlak]

Personally I think Robert's system is pretty good. Motors on/off and if they are off for more than 5 sec then disarm.

Although it makes me nervous, I think the best case for it is: that once the 5 second point has been reached, further intents to turn motors on would likely produce disaster. Obviously it's falling in that time, therefore closer to ground. Besides, we can assume that the vehicle is disoriented or unstable (the reason for selecting kill in flight in the first place), so turning on again, closer to the ground with a vehicle that could do just about anything, is asking for trouble.

[Robert Lefebvre]

David, so I think your discussion here is related to the manual kill switch discussion. This thread is for discussing GCS operations. ;)

HZL, as it stands now, there is no plan to offer flight termination via this new E-stop switch from the GCS.  Part of the reason for that is it would require a new MAVlink message I think.  At the very least, it would require deciding to utilize some existing message which might already exist.  In either case, it requires communications between these two groups.  At present, I don't believe we've heard anything from any of the GCS guys about this topic.

The only thing we can do in Arducopter, is make the decision to allow disarming in-flight, or disallow disarming in flight.  We are planning to allow in-flight disarming of copters. As scary as that is, the prospect of never allowing disarm in flight is even worse.  Right now, there's a huge number of people using the system, who have no idea how to stop a run-away copter quickly.  They don't understand that you would switch to Stabilize, and they're unlikely to learn that (and Stabilize mode might not be permitted from the GCS anyway).  The only thing they might do, is madly poke at the disarm button.  If the system simply refused to disarm, there's nothing else they can do other than grab their copter that is probably stuck in a bush, props spinning, and try and pull the battery.

We can only hope that the GCS maintainers come up with a safe and robust way to manage armed/disarmed state.

The question about whether or not to allow to re-start flying after a potential accidental flight termination, is it's own separate issue.  I think most of us feel that, the type of flying that is typically done from the tablet, is not likely to be a good fit with a kind of rapid on/off/on without a serious risk of an undesirable outcome.  If you decided to terminate flight from the GCS, you should be really sure that's what you want to do, and it should really not be possible to do it accidentally.

--
You received this message because you are subscribed to the Google Groups "drones-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to drones-discus...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Al B]

Hi, Robert, David,

We are developing an automatic failsafe algorithm that allows a quadcopter to gracefully cope with the loss of two(2) motors by regaining stability and control so it can execute an emergency soft lading.

Will this use-case still trigger a disarm if only 2 motor are off or will this only apply if all 4 motors are shutdown when either an E-stop switch from the GCS is pressed or a specific MAVlink message is sent?

[Robert Lefebvre]

It should only disarm if the motors have been stopped by the user (zero throttle, E-stop, or motor interlock disabled). So you should be good.   The only thing might be if the automatic land detector is somehow fooled, but I don't think that is likely to happen.

[Randy Mackay]

Cagiva851,

 

     Such a  recovery feature would be incredible and if I can answer any question to help you on your way, please feel free to ping me.

     Many have probably seen it but here’s a video of our friends at ETH’s success in this area.    https://www.youtube.com/watch?v=bsHryqnvyYA

 

-Randy

[Bill Bonney]

> We can only hope that the GCS maintainers come up with a safe and robust way to manage armed/disarmed state.

We don’t manage state in the GCS, but report it. The Hearbeat message lets us know if the APM is armed, or disarmed. Heartbeat is sent every second. We don’t manage state, as we don’t try and guess the sate of the autopilot, as that can get out-of-sync easily.

Currently to disarm via AP2 you need to click disarm and then OK that action.

Proposed other options are a Disarm button that needs to be pressed for a short period.

Please advise what you think is the safest way to send a disarm command. I personally like the hold and wait for 2 seconds as its easy for touch based control. And avoids an erroneous quick tap.

Comments welcome?

Bill

[Robert Lefebvre]

I prefer a slide-lock type action, is that possible in AP2?  I find it's easiest to do intentionally, and least likely to do unintentionally, which is why it's used to avoid pocket dialing?

[David Pawlak]

"It should only disarm if the motors have been stopped by the user (zero throttle, E-stop, or motor interlock disabled)."

After 5 seconds right? Or at least some albeit short time. I think an oops feature although short, may be VERY useful.

[Robert Lefebvre]

Yes, that's right.  Though I'm questioning if 5 seconds really is too short.  It's not actually that much time. It's OK for a quick oops.  But if somebody were trying to do a trick, shutting off the motors in flight, you could exceed it.

And for Helis I definitely need to do something different with the logic.

[Josh Welsh]

Maybe param it and set the default to 5, letting the user choose higher or lower values themselves?

 

From: drones-...@googlegroups.com [mailto:drones-...@googlegroups.com] On Behalf Of Robert Lefebvre
Sent: Thursday, March 19, 2015 1:30 PM
To: drones-discuss
Subject: Re: [drones-discuss] Re: Use cases for disarm in flight for arducopter?

 

Yes, that's right.  Though I'm questioning if 5 seconds really is too short.  It's not actually that much time. It's OK for a quick oops.  But if somebody were trying to do a trick, shutting off the motors in flight, you could exceed it.