Needed: LOW ALTITUDE DJI Pilot app software enhancements

David M Millsaps · Jun 13, 2015

Eduardo said:
My opinion is that a lot can be made on the rc controller side before thinking on more complex sensors and software.

The problem you mentioned for example, I think it's an error to have power and rudder controlled by the same stick (where you need sub-milimeter precision).

Hi Eduardo,

Thank you for your thoughts.

You are correct that it is problematic "having power and rudder controlled by the same stick (where you need sub-milimeter precision)" ... but that is standard "Mode 2" setup, and changing these around may cause other problems.

I agree that it is pointless to talk about adding sensors or hardware ... it's simply not going to happen. Modifying the DJI Pilot app with needed enhancements is a much more feasible goal.

To me, it makes more sense to look at data already being streamed out from the existing unmodified Inspire, and seeing how the DJI Pilot app software (and firmware in some cases) could be modified/enhanced to trigger screen-view or audio/voice altitude-related warnings -and- a very basic rough minimum altitude prohibition (except under very specific circumstances) … with options toggles so as to best suit the needs of the particular pilot and the particular flight mission. For details see:

NEW PDF - 3RD REVISION: As of 6/16/2015, see this link instead: http://www.unifourpeds.com/Needed_DJI_Inspire_low-altitude_enhancements_REVISION_3.pdf
Or Word Document: http://unifourpeds.com/Needed_DJI_Inspire_low-altitude_enhancements_REVISION_3.docx
... (Original (old) document: http://www.unifourpeds.com/Neede ... de_enhancements.pdf)

David M Millsaps · Jun 13, 2015

The Editor said:
Hi David - PM now sent, thanks.

Thanks Mark/Editor - Sent information that you requested via email - Thanks for your interest.
Dave

The Editor · Jun 14, 2015

David M Millsaps said:
Thanks Mark/Editor - Sent information that you requested via email - Thanks for your interest.
Dave

David, I haven't given out my email address so I'm not sure how you could have sent me anything.
Could you please reply to my PM with the info?

Thank you.

John Alan · Jun 14, 2015

Seems to me a better way to do search and rescue grid flying would be an autopilot app flying the drone and the operator "flying the camera"...
There are several grid flying autopilot apps soon to be released by several sources...
Set your flying height above any obstacles when you define the search zone... (fixes your biggest concern)

Now then... even this solution may need some helpful additions... geared to the task at hand... SAR

Have one button (A) halt the drone mid air in case you see something...
Another press (A) and the search pattern continues where it left off...

A different button to press (B) if autopilot was halted... could then record your exact position and then let you fly off in full control...
Take a closer look at something... even let you crash if you don't take a look around first... full control is yours...
Think of this as a 2nd temporary midair home point as far as the auto pilot is concerned...
Another press (B) and it flies straight back to this temp spot in space...
Hit (A) and it continues on the pattern...

Of course you set your RTH altitude separately and have the option to do a RTH (C) button press at any time to disengage the autopilot and go back to your original take off homepoint...

I could go on with ideas...
Have the app recalc the grid pattern if you halt (A) and or wander off (B) to shorten the pattern for remaining battery charge...
Have the app break large areas down into smaller areas for you based on estimated flight time per battery...
Define the entire area and your home base on the google map... speed and altitude to stay at (for clearance)... calculate...
Shows you needs X batteries at 100% charge each and asks which subgrid do you want to do first...
Marks grids flown as you go... you pick your order as you go...
Recalculates unflown grids based on actual verses estimated times on flown grids... to take wind gusts, etc out of the figuring.
It might start out with 8 subgrids... but after 2 flown you may have 7 left to fly because of it recalculating...

No... there is a reason airports have ILS glidescopes and airliners have autopilots that will fly the plane right down to the numbers on the end of the runway...
Sometime the safest way to do things... is get a machine to fly the plane so the human is out of the equation....

Your PDF insists the human pilot stay in the equation and put in multiple things (to have to be learned and remembered) to help him...
I say take the SAR piloting away from the pilot and give him a pause (A), return & resume (B), and abort (C) buttons...

Just my opinion on this... for what it's worth...

David M Millsaps · Jun 17, 2015

r

John Alan said:
Seems to me a better way to do search and rescue grid flying would be an autopilot app flying the drone and the operator "flying the camera"...
There are several grid flying autopilot apps soon to be released by several sources...
Set your flying height above any obstacles when you define the search zone... (fixes your biggest concern)

Now then... even this solution may need some helpful additions... geared to the task at hand... SAR

Have one button (A) halt the drone mid air in case you see something...
Another press (A) and the search pattern continues where it left off...

A different button to press (B) if autopilot was halted... could then record your exact position and then let you fly off in full control...
Take a closer look at something... even let you crash if you don't take a look around first... full control is yours...
Think of this as a 2nd temporary midair home point as far as the auto pilot is concerned...
Another press (B) and it flies straight back to this temp spot in space...
Hit (A) and it continues on the pattern...

Of course you set your RTH altitude separately and have the option to do a RTH (C) button press at any time to disengage the autopilot and go back to your original take off homepoint...

I could go on with ideas...
Have the app recalc the grid pattern if you halt (A) and or wander off (B) to shorten the pattern for remaining battery charge...
Have the app break large areas down into smaller areas for you based on estimated flight time per battery...
Define the entire area and your home base on the google map... speed and altitude to stay at (for clearance)... calculate...
Shows you needs X batteries at 100% charge each and asks which subgrid do you want to do first...
Marks grids flown as you go... you pick your order as you go...
Recalculates unflown grids based on actual verses estimated times on flown grids... to take wind gusts, etc out of the figuring.
It might start out with 8 subgrids... but after 2 flown you may have 7 left to fly because of it recalculating...

No... there is a reason airports have ILS glidescopes and airliners have autopilots that will fly the plane right down to the numbers on the end of the runway...
Sometime the safest way to do things... is get a machine to fly the plane so the human is out of the equation....

Your PDF insists the human pilot stay in the equation and put in multiple things (to have to be learned and remembered) to help him...
I say take the SAR piloting away from the pilot and give him a pause (A), return & resume (B), and abort (C) buttons...

Just my opinion on this... for what it's worth...

Excellent points made. I would hope that some of the concepts that you have laid out will be incorporated into the autopilots to be released for the Inspire. I absolutely will purchase the autopilot when it becomes available, and I hope that this will meets your expectations. Later, please advise me on it's availability by PM or email or text message..

It's not easy to find a second skilled (camera) controller operator when and where I fly, although that would certainly be preferable. I find it to be increasingly difficult to improve my videography skills while maintaining my piloting skills at the same time and with the same controller. Especially at low altitude flight, it is simply too distracting and prone to errors (and crashes) for the pilot to do all of this. .As you say, sometimes it's better to have the computer do at least part of the work for you.

Since the original post of my document, there have been significant revisions (Now on revision 3). This have received a multitude of recommendations from MANY interested individuals in many fields (by email, calls, meetings, and FAX primarily, but very little via Inspire Pilots or DJI forum). I was pleasantly surprised at the high level of interest shown in a "crash prevention topic". Clearly, I am not the only one to crash an Inspire.Folks have put lots of time and thought into this document, so please help us out. This now completely revised document, which, I hope will be more helpful to all.

NEW PDF - 3RD REVISION: As of 6/17/2015, see this link instead: http://www.unifourpeds.com/Needed_DJI_Inspire_low-altitude_enhancements_REVISION_3.pdf
Or Word Document: http://unifourpeds.com/Needed_DJI_Inspire_low-altitude_enhancements_REVISION_3.docx

I have included a Word version to make it easier to mark up changes, edits, and new ideas (please put changes in some other color to make them easier to locate) and email of FAX back to me. I would appreciate your input, and would appreciate anyone else who wants to contribute to this topic. I would probably easier to email or FAX directly to me rather than going them forums.

Thanks,

David M. Millsaps
Unifour Pediatrics PA
Hickory NC
[email protected]
828-320-5152 cellular
828-327-9988 FAX

The Editor · Jun 18, 2015

@David M Millsaps - David, you were going to send me details of who you have spoken to at DJI? Please send them through via PM.

Thank You.

I believe the reason you have not garnered much interest via Inspire Pilots or DJI forum is because we appreciate that the things you are asking are just not going to happen via the DJI Pilot App - EVER!

Additionally, pretty much all of your wish list is already commercially available on other, better suited platforms. Whilst some third party apps will probably come to market as time goes on (there is at least one in beta currently) I believe you are merely flogging a dead horse if you think for one minute DJI are going to be interested in taking up your (9 pages) of ideas.

You say you have had considerable interest in your document but from whom? Not Inspire Pilots members or DJI Inspire Forum members - so who and how many?

Starship · Jun 30, 2015

Eduardo said:
My opinion is that a lot can be made on the rc controller side before thinking on more complex sensors and software.

The problem you mentioned for example, I think it's an error to have power and rudder controlled by the same stick (where you need sub-milimeter precision).

In standard mode, RC airplanes and Helicopters, power and rudder are on the same stick. Mode 2 is the most common configuration. If you need more precision, modify your gains.

Eduardo · Jul 1, 2015

Starship said:
In standard mode, RC airplanes and Helicopters, power and rudder are on the same stick. Mode 2 is the most common configuration. If you need more precision, modify your gains.

I was trying to mean that companies like DJI and 3DR could try and spend more time rethinking the controller itself. Maybe an easier controlller for ATTI mode could reduce fly aways, GPS dependency, etc.

And regarding rudder control, I would prefer it associated with aileron, like this mod here (3:10):

Shazbot4 · Jul 9, 2015

david you got some mad typing skillz. most of it went over my head but i think i got the gist, correct me if im wrong. want you want is the inspire not to drop altitude unless you tell it to. believe me we all want that and 5 firmware updates ago it wasn't an issue like it is now. the inspire use to keep altitude pretty darn well no matter how hard you flung it around and actually use to gain it instead of slowly loose it like it does now.

you want object avoidance. who ever said it's not going to happen or it's not possible is blowing smoke. technically the inspire already has it with the ground sensor sonar and lazier thingy. all you need to do is point that sensor forward and your there.

better yet you can just buy the radar, yes i said radar. a company DIY here in san diego makes almost every conceivable sensor for model r/c aircraft you can think of that would probably do the job your asking for. your inspire is going to look like johny number 5 when your finished but your inspire would have real time active object avoidance in all 6 directions and everything in between.

also DJI is in the firm position of dragging things out and milking consumers for maximum amounts of money for 1 product. all the stuff the general majority of us who bought the I1 who asked why doesn't it have this or that will almost certainly be in the inspire 2.