Is it ok to cancel drone initiated low battery RTH

[Mr. Salty]

As requested by a poster in this thread return home mode., I'll ask the question that's being debated: Is it OK to cancel drone-initiated low battery RTH and bring your aircraft back to land it manually?

Please explain your reasoning either way.

 

[old man mavic]

there could be a situation where because of the preset RTH height, the drone was struggling to return home into the wind, in that case it would be better to reduce altitude if possible, and try to get back manually, or if very close to home but in danger of landing in an unsafe place such as water or into people below the drone , of course it would be better to keep an eye on remaining battery power, and not put yourself in such a position in the first place so yes it is ok

 

[Cyber3xpert]

I would say, yes, it's ok, but don't ever get in a position that you cannot RTH; this is a nightmare to experience. My advice to canceling RTH is the following:

Watch your battery percent levels - You do not want to have you drone die and try to land a mile away.

Know your drone's limits - If you know how much battery it takes to get back to your landing place, then keep that number in mind and don't let the battery get below that level.

 

[RWD]

there could be a situation where because of the preset RTH height, the drone was struggling to return home into the wind, in that case it would be better to reduce altitude if possible, and try to get back manually, or if very close to home but in danger of landing in an unsafe place such as water or into people below the drone , of course it would be better to keep an eye on remaining battery power, and not put yourself in such a position in the first place so yes it is ok

I think this is an excellent example of cancelling a low battery RTH, resulting in the bird getting safely home with more battery reserve than just leaving the RTH active.

 

[Phantomrain.org]

As a general rule if I I have VLOS and the Battery is low , will cancel RTH and fly back, however if VLOS is sketchy will allow RTH to do its magic until clear VLOS has returned and than cancel RTH.

Low battery for beginners is scary: RTH gives you the confidence that your Drone can even make it back .

In order to cancel the RTH with low battery and you have to be confident that your Orientation is correct , you have to be able to see the Screen clearly and if there is Glaring Sun on you screen it can really alter that confidence.

For me RTH is giving me the Quickest Orientation possible and low battery is the signal for me to cancel it and Speed on back in linear fashion.

Phantomrain.org
Cutting edge Gear for your Mavic
Coal

 

[sar104]

As requested by a poster in this thread return home mode., I'll ask the question that's being debated: Is it OK to cancel drone-initiated low battery RTH and bring your aircraft back to land it manually?

Please explain your reasoning either way.

As I posted on that thread:

One certainly should not ignore the smart RTH function but, as @Meta4 pointed out, if the aircraft has enough reserve to return home under FC control then it obviously has enough to return under pilot control. He is also correct in pointing out that flight efficiency peaks at a higher speed than the default RTH speed and so, in general, you can do better than auto RTH, especially if you don't need to ascend to the preset RTH height.

 

[RWD]

As I posted on that thread:

One certainly should not ignore the smart RTH function but, as @Meta4 pointed out, if the aircraft has enough reserve to return home under FC control then it obviously has enough to return under pilot control. He is also correct in pointing out that flight efficiency peaks at a higher speed than the default RTH speed and so, in general, you can do better than auto RTH, especially if you don't need to ascend to the preset RTH height.

I was wondering about that @sar104. I was trying to think how one could generate the data to drone efficiency vs speed, and use that to look for a maximum point like you reference. Should be able to do that, but would want to generate the data on a dead calm day or the results end up confounded by varying wind speeds.

Do you have a dataset on this already?

 

[sar104]

I was wondering about that @sar104. I was trying to think how one could generate the data to drone efficiency vs speed, and use that to look for a maximum point like you reference. Should be able to do that, but would want to generate the data on a dead calm day or the results end up confounded by varying wind speeds.

Do you have a dataset on this already?

I keep meaning to generate one, but I haven't got around to it yet. However, here's an analytic look at the problem based on known drag characteristics:

I thought I had plenty of reserve to return...and the forced landing warning was inducing panic

I think it's an entrance to an underwater tunnel?It is. Out on a bridge then into a tunnel and out to another bridge, I think.

mavicpilots.com

 

[RWD]

Thanks very much for that link ... I am going to have to sit down and digest that one, but looks like a really good analysis. I have a few questions, but need to think a bit first to avoid making a fool of myself with dumb questions....

One of the things I have rattling around in my head is the impact of wind. Without resorting to any detailed analysis, I would expect a purely headwind or tailwind to be easy to account for since the horizontal speed vector in your analysis is airspeed, not groundspeed. So the optimum RTH airspeed is always going to be the same, but groundspeed will vary by + or - the windspeed. I think the problem gets more complicated when the wind is not parallel to RTH heading, since a new component of horizontal velocity would be needed, perpendicular to RTH heading to compensate for the crosswind. Does that sound right to you??

If the weather cooperates around here and we get a calm day, I will try to generate a dataset to see how closely it relates to your analytic view. I think that would be fairly interesting.

 

[Kilrah]

For the Mavic 1 someone had posted data.

 

[RWD]

For the Mavic 1 someone had posted data.

Thanks @Kilrah, I will do some looking for that. If you have anything to help narrow down the search (timeframe, which forum, etc.) that would be appreciated....

BELAY THAT REQUEST @Kilrah ... I think I found it:

Mavic Battery and Speed efficiency chart

My son did a middle school science project using the Mavic. See the attached graph. He (and I) expected a different result. I believe the results very clearly show that at 20 MPH and up to max speed, the battery use per foot traveled is equivalent. Therefore, there is no battery penalty for...

mavicpilots.com

Thanks for the tip !!!

 

[old man mavic]

depending on the wind direction but just for the sake of this thread, then if you have VLOS and are trying to get back directly into the wind
apart from reducing height you could use the same method as a sail boat uses and tack across the wind to make better progress
this would lessen the forces being applied to the drone to some extent, and of course would be dependant on terrain and amount of battery remaining

 

[sar104]

Thanks very much for that link ... I am going to have to sit down and digest that one, but looks like a really good analysis. I have a few questions, but need to think a bit first to avoid making a fool of myself with dumb questions....

One of the things I have rattling around in my head is the impact of wind. Without resorting to any detailed analysis, I would expect a purely headwind or tailwind to be easy to account for since the horizontal speed vector in your analysis is airspeed, not groundspeed. So the optimum RTH airspeed is always going to be the same, but groundspeed will vary by + or - the windspeed. I think the problem gets more complicated when the wind is not parallel to RTH heading, since a new component of horizontal velocity would be needed, perpendicular to RTH heading to compensate for the crosswind. Does that sound right to you??

If the weather cooperates around here and we get a calm day, I will try to generate a dataset to see how closely it relates to your analytic view. I think that would be fairly interesting.

Actually the optimal RTH speed is not always the same, as a simple thought experiment will show. Suppose that in still air the optimal speed (i.e. minimizing energy per unit distance) is 25 mph. Then consider returning into a 25 mph headwind. Energy per unit distance is ∞, and the aircaft will never return. The problem, of course, arises because the integral of airspeed, w with respect to time is not equal to distance travelled, d, (∫ w dt ≠ d ).

 

[RWD]

Actually the optimal RTH speed is not always the same, as a simple thought experiment will show. Suppose that in still air the optimal speed (i.e. minimizing energy per unit distance) is 25 mph. Then consider returning into a 25 mph headwind. Energy per unit distance is ∞, and the aircaft will never return. The problem, of course, arises because the integral of airspeed, w with respect to time is not equal to distance travelled, d, (∫ w dt ≠ d ).

Hmmm. At the risk of wandering down this rabbit hole a bit more (mods, we are getting off topic, please let us know if we need a new, separate thread).

Thanks for the reply. Can you help me understand the flaw in my thinking about the case where head or tail wind is parallel to RTH heading?? In thinking about it more, I suspect it is a faulty treatment of drag and how it increases with square of airspeed. Do I have that right??

And for your thought experiment, groundspeed is zero, that is why energy for unit distance is infinite. And I'd have to integrate groundspeed with respect to time to get distance. Also - checking if I have that right??

PS - thanks for sticking with me on this, I am enjoying the learning...

 

[ZX14-R]

Cancel and return to home in SPORTS MODE if possible.

All depends on conditions and how far away the drone is.

 

[Thomas B]

Cancel and return to home in SPORTS MODE if possible.

All depends on conditions and how far away the drone is.

Agree... when I am hovering directly above the take-off spot for hyperlapse I will commonly cancel RTH until the drone no longer offers the dismiss option. Still gets me down from 200 ft altitude with battery time above zero. Of note, the drone does not, in the events I reference, do any rotation from the direction the camera is pointed.

 

[Meta4]

Cancel and return to home in SPORTS MODE if possible.

Just like driving your car at 150 km/hr gets lousy mileage, flying in sport mode will give less miles/battery .
If your battery level and distance aren't a problem, it's no problem but when things are tight, Sport Mode is not your friend.

 

[sar104]

Hmmm. At the risk of wandering down this rabbit hole a bit more (mods, we are getting off topic, please let us know if we need a new, separate thread).

Thanks for the reply. Can you help me understand the flaw in my thinking about the case where head or tail wind is parallel to RTH heading?? In thinking about it more, I suspect it is a faulty treatment of drag and how it increases with square of airspeed. Do I have that right??

I don't think you have to worry about how drag changes with airspeed - this is just an issue that the aircraft has a maximum airspeed

And for your thought experiment, groundspeed is zero, that is why energy for unit distance is infinite. And I'd have to integrate groundspeed with respect to time to get distance. Also - checking if I have that right??

PS - thanks for sticking with me on this, I am enjoying the learning...

That's correct. Using some simplifying assumptions one can explore that relationship further. Assuming that the power required to stay airborne is fixed (~ 130 W for a Mavic 2), that drag goes with the square of the airspeed, and that the drag coefficient for the M2 is 0.025, the energy usage per unit distance, as a function of windspeed (headwind) and airspeed can be calculated analytically:



The curves show minima, as expected, that increase with increasing windspeed. Differentiating with respect to airspeed to get the location of those minima:



So for zero headwind, that indicates an optimum airspeed of 13.8 m/s (31 mph) to maximize flight distance as a function of energy usage. As the headwind increases the optimum airspeed also increases. 21 m/s (46 mph) is around the maximum airspeed of the M2, and so that suggests that it will struggle into a headwind of greater than 10 m/s (22 mph). We know it can go into greater headwinds than that however, indicating that the assumptions are not completely correct. However, the basic trend is correct.

 

[russellg]

Sure, cancel low battery RTH and then you can post the flight log and have the experts tell you when and why it crashed and help you find it.

 

[Ralph thompson]

As a mathematician & aerodynamicist I love the discussion but I’d caution about being too theoretical. You can reduce drag by flying close to the ground, the way Charles Lindbergh conserved fuel on his epic transatlantic flight but for a small drone that is quite risky. But more importantly, local conditions can vary significantly so knowing the approximate wind speed from local met. services could be hazardous (local downdrafts & circulation from irregular terrain, hills, trees buildings etc. could change wind speed and direction significantly. There’s a saying amongst pilots “a good pilot never needs to use his skills”, good planning, contingency plans & caution will avoid tight situations.
I wonder if spurious RTH signals might be generated by the drone measuring a prolonged wind gust, and deciding RTH is necessay.