GPS accuracy is the Holy Grail

[Mauiwind]

As a cinematic tool, I can look past the limitations of sharpness, color bit depth, controllable DOF, and all the other limitations of the Solo/GP platform.
What has me hating life (and its by no means exclusive to Solo) is the accuracy of manifesting GPS coordinates.
OK, Ok.. I get it that civilian satellite GPS has a purposeful limitation. But I hate it.

In my mind, what keeps a UAVs from replacing actual cranes, dollies, etc is accurate and repeatable camera movement. Sure, they work great for sloppy shots where there is no needle to thread but that wow wa woo tracking is for the birds. ( more like lame ducks )

Give me me a camera platform that will allow me to program a shot with reliable, drop dead accurate, repeatable motion control. Yeah, lets make it tricky, lets autonamously fly through holes the size of tires, lets program in easing in and out of waypoints and camera movements. Lets use the inherent flexibility of the UAVs to go whete no crane or steadicam has gone before.
Can't do it. Crap, call in the cranes, burly grips and ground based motion control bots.

Until we can make a drone track a high resolution, pre-recorded path, the platform will never be the cinematic tool of choice. (and budget) that it could be.

Note: The first person to figure this out without illegaly hacking military GPS will get a grammy and change filmaking forever.

 

[Ian [P13]]

You can get this from RTK GPS, the Pixhawk 2 is alpha testing this, this very week. This provides centimetre level accuracy.

The project for Solo to be able to run Arducopter Master is a prerequisite to be able to run RTK though as the GPS corrections are sent via MavLink.

I was just this morning saying to @Proficnc (Philip Rowse) how cool it would be to do an RTK smart shot.

 

[monterx7]

Michael Bay did use our same setup(solo/gp4) to film shots in 13 hours.

 

[Ian [P13]]

Michael Bay did use our same setup(solo/gp4) to film shots in 13 hours.

That was just a normal smart shot though so if multiple takes were done it would have been close but not exact.

 

[Mauiwind]

Well. Im not savy to ardu or other nomenclature but tight tracking is a major game changer. If sure like to participate in the development. Its a big deal

 

[Ian [P13]]

Well. Im not savy to ardu or other nomenclature but tight tracking is a major game changer. If sure like to participate in the development. Its a big deal

If you keep an eye on this Facebook page you'll see information on the progress of the RTK GPS which is a Ublox M8P.

需要安全验证

The Solo's progress on Arducopter Master will likely be mentioned here and on the main Solo Facebook page too.

 

[franknitty69]

it would be nice, but for comparison a Milo motion control rig, costs $500,000 and thats without a camera. i'm grateful that i can get cable cam on repeat for $500 even if its off by a foot or two between takes. its not an issue for vfx tools like Mocha Pro or Nuke.

 

[stampbags13]

As a cinematic tool, I can look past the limitations of sharpness, color bit depth, controllable DOF, and all the other limitations of the Solo/GP platform.
What has me hating life (and its by no means exclusive to Solo) is the accuracy of manifesting GPS coordinates.
OK, Ok.. I get it that civilian satellite GPS has a purposeful limitation. But I hate it.

In my mind, what keeps a UAVs from replacing actual cranes, dollies, etc is accurate and repeatable camera movement. Sure, they work great for sloppy shots where there is no needle to thread but that wow wa woo tracking is for the birds. ( more like lame ducks )

Give me me a camera platform that will allow me to program a shot with reliable, drop dead accurate, repeatable motion control. Yeah, lets make it tricky, lets autonamously fly through holes the size of tires, lets program in easing in and out of waypoints and camera movements. Lets use the inherent flexibility of the UAVs to go whete no crane or steadicam has gone before.
Can't do it. Crap, call in the cranes, burly grips and ground based motion control bots.

Until we can make a drone track a high resolution, pre-recorded path, the platform will never be the cinematic tool of choice. (and budget) that it could be.

Note: The first person to figure this out without illegaly hacking military GPS will get a grammy and change filmaking forever.

 

[stampbags13]

I got my solo & gimbal as soon as it came out all I did was use the 3dr iris+ gps shield over my solo's gps and my solo lands maybe 3inches off from takeoff point everytime! and honestly if it is windy as hell it seems to fight and land exactly from where it took off from!if you have a solo you need to get polarprofilters head&tail light set, propguards,leg ext and of course there gopro filters and solex app is a must, Solex is what makes solo great and Solex keeps adding new feathat always work! and they have a lot of new stuff coming that keep making other quads like the p4 toys!!!

 

[Mike Boland]

Until we can make a drone track a high resolution, pre-recorded path, the platform will never be the cinematic tool of choice. (and budget) that it could be.

Note: The first person to figure this out without illegaly hacking military GPS will get a grammy and change filmaking forever.

OK, wheres my Grammy.
We are using RTK at present and have a measured GPS point spread of +/-6mm on the copter.
That accurate enough?

 

[quick_dry]

OK, wheres my Grammy.
We are using RTK at present and have a measured GPS point spread of +/-6mm on the copter.
That accurate enough?

Does the RTK correction also give you that accuracy in altitude, or is that still referencing from the baro?

 

[Mike Boland]

I'm just building this machine, and there is some questions about height accuracy.
from the client "Just acquired a 5 metre pixel resolution DEM for Qld (mostly coastal areas and inland town areas) it is vertically accurate at 0.8 metre and horizontally accurate at 0.3 metres." which is to be used in the GCS.
As far as real height goes this is where I get lost in the jargon of surveying and this is a question we are asking developers :
"The Septentrio stores coordinates in the coordinate system we are using. (MGA94 - Zone nn). MGA94 used GDA94 datum which is slightly different to the WGS84 datum that GPS use. While they use the same spheroid (GRS80) they use slightly different transformation parameters, scales etc which equates to about a 20cm difference horizontally. In Australia we also have a specific height datum tied to port datums and approximated using a gravimetric grid (Ausgeoid09 being the latest accepted model). The GPS uses an ellipsoid height based on the spherical model used to approximate the earths surface. The Septentrio send coordinate to the pixhawk in wgs84.

So….. in short. It would be good if all systems (Septentrio RTK GNSS and Pixhawk and Mission Planner) all used the same datums and coordinate systems. The simplest solution would be to have Mission Planner allow you to select the coordinate system (other than WGS84) that you want to use for your project including implementation of Ausgeoid09 to calculate height in AHD.

MGA94 is ITRF compliant and as such accounts for the 7cm per year continental drift. Also very important if you want to work in cm accuracies over extended periods in same location ie landfill airspace monitoring."

So I hope that answers your question, cause I'm none the wiser.

 

[Mauiwind]

I'm just building this machine, and there is some questions about height accuracy.
from the client "Just acquired a 5 metre pixel resolution DEM for Qld (mostly coastal areas and inland town areas) it is vertically accurate at 0.8 metre and horizontally accurate at 0.3 metres." which is to be used in the GCS.
As far as real height goes this is where I get lost in the jargon of surveying and this is a question we are asking developers :
"The Septentrio stores coordinates in the coordinate system we are using. (MGA94 - Zone nn). MGA94 used GDA94 datum which is slightly different to the WGS84 datum that GPS use. While they use the same spheroid (GRS80) they use slightly different transformation parameters, scales etc which equates to about a 20cm difference horizontally. In Australia we also have a specific height datum tied to port datums and approximated using a gravimetric grid (Ausgeoid09 being the latest accepted model). The GPS uses an ellipsoid height based on the spherical model used to approximate the earths surface. The Septentrio send coordinate to the pixhawk in wgs84.

So….. in short. It would be good if all systems (Septentrio RTK GNSS and Pixhawk and Mission Planner) all used the same datums and coordinate systems. The simplest solution would be to have Mission Planner allow you to select the coordinate system (other than WGS84) that you want to use for your project including implementation of Ausgeoid09 to calculate height in AHD.

MGA94 is ITRF compliant and as such accounts for the 7cm per year continental drift. Also very important if you want to work in cm accuracies over extended periods in same location ie landfill airspace monitoring."

So I hope that answers your question, cause I'm none the wiser.

 

[Mauiwind]

OK, your commincating with your own kind. You and your buds are way way advanced. Could you lower your IQ for a sec and explain in terms us Neanderthals can grasp, what RTK is and how it would be implemnted into something like Solo? Is it going to be really expensive to implement? is it compli ated? Why hasn't it been used on octo copters with expensive cameras, etc?
Thanks from the cheap seats.

 

[Mauiwind]

OK, wheres my Grammy.
We are using RTK at present and have a measured GPS point spread of +/-6mm on the copter.
That accurate enough?

Seriously, if you make it real and reliable you will change cinema forever. It would be a really big step forward for cinematography. GIANT
It would replace millions of dollars of equipment in a heartbeat.
Honestly, its akin to Something like a Free and Continuous energy source. (Eh... not quite) Lots of people would lose Their jobs (jib jobs) and mfg. might want to put a contract on your head. You cant imagine the expensive and complcated instuments that such a capability would make obsolete.We're talking really accurate motion control systems with no boundaries in 3d space. Just have it available and the movie industry will hire you as a consultant instantly.

 

[Mike Boland]

@Mauiwind Simply put RTK is a GPS correction system to get sub cm accuracy.
You use a ground station (although SBAS uses a satellite to improve accuracy in a similar way) over a know survey point which in essence gets the same GPS fix as your mobile unit but it KNOWS the GPS is wrong and relays that correction to the mobile unit. That's it, but VERY simply put.
We are using a Septentrio multi constellation, multi frequency GPS that accepts RTK correction, has it's own on board logging, and a host of other features, but that little credit card size board is apparently $12K. :-(
UBlox is coming out with a similar thing, RTK correction, but I am not sure how that will work without an accurate base station at a known survey point.
For example: The test survey point in my back paddock was established by setting the base station over it for 6hrs, collecting all the data and sending it off to a company that returns a 10 page report on the point to give a known location accurate to 1.2mm. This stuff is way above my head.
So you need some known accurate location for the RTK base for it to work.
I will be watching with interest to see how UBlox are going to implement this in real life, considering what we have to do to get this accuracy.
And the Spatial Analysts still insist that you cannot get sub cm accuracy with a single frequency GPS system, but that's another whole discussion.

 

[SARDG]

OK, your commincating with your own kind. You and your buds are way way advanced. Could you lower your IQ for a sec and explain in terms us Neanderthals can grasp, what RTK is and how it would be implemnted into something like Solo? Is it going to be really expensive to implement? is it compli ated? Why hasn't it been used on octo copters with expensive cameras, etc?
Thanks from the cheap seats.

LMAO... your post cracked me up!

 

[Mauiwind]

@Mauiwind Simply put RTK is a GPS correction system to get sub cm accuracy.
You use a ground station (although SBAS uses a satellite to improve accuracy in a similar way) over a know survey point which in essence gets the same GPS fix as your mobile unit but it KNOWS the GPS is wrong and relays that correction to the mobile unit. That's it, but VERY simply put.
We are using a Septentrio multi constellation, multi frequency GPS that accepts RTK correction, has it's own on board logging, and a host of other features, but that little credit card size board is apparently $12K. :-(
UBlox is coming out with a similar thing, RTK correction, but I am not sure how that will work without an accurate base station at a known survey point.
For example: The test survey point in my back paddock was established by setting the base station over it for 6hrs, collecting all the data and sending it off to a company that returns a 10 page report on the point to give a known location accurate to 1.2mm. This stuff is way above my head.
So you need some known accurate location for the RTK base for it to work.
I will be watching with interest to see how UBlox are going to implement this in real life, considering what we have to do to get this accuracy.
And the Spatial Analysts still insist that you cannot get sub cm accuracy with a single frequency GPS system, but that's another whole discussion.

So let me see if I got this right.... The RTK can sort of be looked at as a relay point with high accuracy because it knows ecactly where it is. (Way better than me cause I have no idea where I'm at)
, What sort of range from the RTK can be had. And... its curious. How can one station orient a client in 3d space. ((I must not be getting something here. ))
I do want to say that to get your Grammy, you need to actually shoot some footage thats position critcal. When you can orbit right around your head and fly back and forth between your legs at speed. Youll not only have a Grammy, youll be instantly bankable. If your demo fails, youll be in the hospital but a huge hit on youtube.

 

[cynfab]

If you are interested in some of the hardware that u-blox is coming out with and some of the testing they have done with their L1 RTK systems
(they use the M8P-0 and M8P-2) see this article in GPS World
Centimeter positioning for UAVs and mass-market applications
The first page or so is a good overview. The rest can get quite dense.

 

[Mike Boland]

When you can orbit right around your head and fly back and forth between your legs at speed.

That might be a stretch as the Octo is 1.4m tip to tip.
The stills we do are over commercial areas and are commercial in Confidence. As well, I do not do the image processing, that I leave to the experts.
I just build em, tune em, and fly em.