Can/will DataPilot be able to upload waypoints from an external file like a shapefile or CSV file? Screen shots suggest that waypoints are entered manually.
H520 DataPilot (TM) Pics
- Thread starter Rayray
- Start date
-
- Tags
- h520 datapilot
On Page 4 Q#33 "Jason Hagen" had asked if it was at all possible to load other data sources such as SRTM data so that the H520 would be able to "follow" the terrain as to keep a consistent GSD? So is it? I realize that "following" the exact terrain elevation difference could pose an issue for ex. If flying over an vegetation that data wouldn't show an elevation changes if simply some vegetation as it where flying where much taller then the rest cuz it'd end up crashing into it! But WTH was the $60 million investment from Intel for if they aren't going to use it as much or at all anymore, ie if Yuneec truly took advantage of the technology they have been able to come up with ONLY because of the investment, they would be installing Realsense on every one of their drones & expand on just adding 2 additional sonar sensors by maybe adding on an camera that faced down as well, that way the H520 would automatically stay an exact height above the terrain even if it where going uphill with occasional dips or whatever by using sonar & the down facing camera to track it's elevation! No I'm not talking about LIDAR where it can penetrate vegetation because like I mentioned above that may pose issues if there where varying height of vegetation (well I guess you could combine the front facing Realsense in that situation)!
Also a rather cheap additional add-on that they could've potentionally put on the H520 is simply a vertical gimbal mount or top side gimbal mount that way you wouldnt need to be so far away to capture detailed pics of the underside of bridges or wind farm turbines, currently ive thought of a mod for the H480 by only finding the wiring that goes to the gimbal inside of the H then you could splice it (in a professional way to prevent noise in pics) & put wires to the top of H then adapt a stock Gimbal mount to the top & pretty soon your in business by having a top mounted cam that'd have the ability to view straight up vertically from the H so that a closer inspection of bridges etc could be possible!!! Yes this would mean you'd only be able to run a single cam at anytime so NO it wouldn't give you any ability to run a CGO3+ on top & a CGO-ET on the bottom at the same time cuz you'd have a really messed up video stream to your ST-16 (but I wonder being as each cam records directly to their own onboard SD card if it may give you an output recording onto each card showing you different views once you put them onto your PC??? Still don't know how exactly you'd be able to know what they are capturing at any specific point other then your controls for pan tilt would most likely still work fine just in Reverse for the Top cam & both bottom & top would move in reaction to the pan & tilt at the same time!!! Idk I'm thinking of going with a different drone manufactured cuz there are way to many out there that already have full 360 degree object avoidance as well as full up & down. Batteries for 40min run times are approx $180 which only gives me the impression that Yuneec either is holding quite a few innovations back & only raising their prices or simply just raising their prices & giving you all innovations that they have! Either way they are not only behind the times of far to many others but do not project any kind of image showing that they may have a chance of catching up with them while also keeping their pricing competitive!
Also a rather cheap additional add-on that they could've potentionally put on the H520 is simply a vertical gimbal mount or top side gimbal mount that way you wouldnt need to be so far away to capture detailed pics of the underside of bridges or wind farm turbines, currently ive thought of a mod for the H480 by only finding the wiring that goes to the gimbal inside of the H then you could splice it (in a professional way to prevent noise in pics) & put wires to the top of H then adapt a stock Gimbal mount to the top & pretty soon your in business by having a top mounted cam that'd have the ability to view straight up vertically from the H so that a closer inspection of bridges etc could be possible!!! Yes this would mean you'd only be able to run a single cam at anytime so NO it wouldn't give you any ability to run a CGO3+ on top & a CGO-ET on the bottom at the same time cuz you'd have a really messed up video stream to your ST-16 (but I wonder being as each cam records directly to their own onboard SD card if it may give you an output recording onto each card showing you different views once you put them onto your PC??? Still don't know how exactly you'd be able to know what they are capturing at any specific point other then your controls for pan tilt would most likely still work fine just in Reverse for the Top cam & both bottom & top would move in reaction to the pan & tilt at the same time!!! Idk I'm thinking of going with a different drone manufactured cuz there are way to many out there that already have full 360 degree object avoidance as well as full up & down. Batteries for 40min run times are approx $180 which only gives me the impression that Yuneec either is holding quite a few innovations back & only raising their prices or simply just raising their prices & giving you all innovations that they have! Either way they are not only behind the times of far to many others but do not project any kind of image showing that they may have a chance of catching up with them while also keeping their pricing competitive!
arruntus
Premium Pilot
- Joined
- Aug 19, 2017
- Messages
- 1,851
- Reaction score
- 730
If you'll allow me Tom, I'm going to give my opinion.
The drone's ability to use terrain data such as SRTM would be a very interesting option. Normally and looking a little bit at the accuracy SRTM gives (from other sources the same thing happens) they themselves tell you the average error it can give and I think I remember that it is over 20m if I am not confused. With this information and to avoid problems it is recommended to fly over 60 AGL as a safety measure. This recommendation is made when using the Px4 past flight scheduling systems that already have this option.
If apart from that we have an altitude detection system, Intel Realsense most likely would greatly improve the accuracy and safety of the flight. When they made the presentation of the H520 in France in an interview, a manager commented that later on other models would come out and would incorporate it. We don't have confirmation yet but I think they will implement it as it happened with the H480.
With regard to incorporating the possibility of an anchorage for the gymbal at the top would be something very nice and would give more versatility. The idea is good but there are many problems with gimbals performance when installed in top position. I think they're products that play in different leagues. Carrying two cameras means increasing the weight, decreasing the flight time and making the aircraft rise above 2 kg so that in many countries and according to regulations BVLOS flights could no longer be made.
Having available batteries of $180 with an autonomy of 40 minutes for a certain aircraft does not mean anything. Each aircraft is focused on something and designed to do that function. I'm not saying anything about the H520 batteries because I think they're expensive too, but let's not confuse 4- or 6-motor multirotors (if I'm not mistaken) because they're different worlds. I use 6s and 12A batteries and get 35 minutes but I don't have any other virtues that the H520 has, all in its right measure. By the way, can you say what battery and drone are you talking about specifically to compare? You've left me with the doubt and having other points of view is always interesting.
At the fair I was talking to some nice guys who had designed a quadcopter with LioN battery and 55 minutes autonomy. The engines and propellers price were of the other league where the H520 dont plays. I didn't even want to ask the price of the set because I'm sure it wouldn't interest me. There's a lot better things out there, but you have to pay for them. I say this because price is often a fundamental factor when it comes to starting to look at drones, among other things of course.
Finally and making it clear that I am not trying to argue or offend anyone. I've been looking for a long time to buy a good drone for making videos and images, specifically for the image theme. I don't think that the H520 is a bad choice considering the path Yuneec has chosen to continue taking into account the advantages of being able to do photogrammetry.
I think that over time the functionalities of the aircraft will increase but we don't know anything for sure yet. This is a problem for people who want or need to buy it now but this will always happen to us. You're gonna buy something, and there's always something better about to come out. In the end you have to make up your mind, otherwise you'll never buy anything.
The drone's ability to use terrain data such as SRTM would be a very interesting option. Normally and looking a little bit at the accuracy SRTM gives (from other sources the same thing happens) they themselves tell you the average error it can give and I think I remember that it is over 20m if I am not confused. With this information and to avoid problems it is recommended to fly over 60 AGL as a safety measure. This recommendation is made when using the Px4 past flight scheduling systems that already have this option.
If apart from that we have an altitude detection system, Intel Realsense most likely would greatly improve the accuracy and safety of the flight. When they made the presentation of the H520 in France in an interview, a manager commented that later on other models would come out and would incorporate it. We don't have confirmation yet but I think they will implement it as it happened with the H480.
With regard to incorporating the possibility of an anchorage for the gymbal at the top would be something very nice and would give more versatility. The idea is good but there are many problems with gimbals performance when installed in top position. I think they're products that play in different leagues. Carrying two cameras means increasing the weight, decreasing the flight time and making the aircraft rise above 2 kg so that in many countries and according to regulations BVLOS flights could no longer be made.
Having available batteries of $180 with an autonomy of 40 minutes for a certain aircraft does not mean anything. Each aircraft is focused on something and designed to do that function. I'm not saying anything about the H520 batteries because I think they're expensive too, but let's not confuse 4- or 6-motor multirotors (if I'm not mistaken) because they're different worlds. I use 6s and 12A batteries and get 35 minutes but I don't have any other virtues that the H520 has, all in its right measure. By the way, can you say what battery and drone are you talking about specifically to compare? You've left me with the doubt and having other points of view is always interesting.
At the fair I was talking to some nice guys who had designed a quadcopter with LioN battery and 55 minutes autonomy. The engines and propellers price were of the other league where the H520 dont plays. I didn't even want to ask the price of the set because I'm sure it wouldn't interest me. There's a lot better things out there, but you have to pay for them. I say this because price is often a fundamental factor when it comes to starting to look at drones, among other things of course.
Finally and making it clear that I am not trying to argue or offend anyone. I've been looking for a long time to buy a good drone for making videos and images, specifically for the image theme. I don't think that the H520 is a bad choice considering the path Yuneec has chosen to continue taking into account the advantages of being able to do photogrammetry.
I think that over time the functionalities of the aircraft will increase but we don't know anything for sure yet. This is a problem for people who want or need to buy it now but this will always happen to us. You're gonna buy something, and there's always something better about to come out. In the end you have to make up your mind, otherwise you'll never buy anything.
PatR
Premium Pilot
Tom,
At our price level what we have for terrain following doesn't work all that well much more than 6-10' off a surface. Sonar and optical flow suffer excessive disbursement and ability to maintain a specific distance from a surface goes out the window. The real good stuff, read that as expensive, uses a radar altimeter or a laser rangefinder, with either slaved to a very fast response auto pilot. We could have the super good stuff but the penalties paid in price and weight are high.
As for the front, upper mounted camera for inspection work, that has already been done with another system, that incorporated very similar mapping features as the 520. It worked very well, used a 24MPXL Sony a6000 for the payload and Pixhawk for the FC. Flight times of 15-22 minutes were common but at a bit over $4k few bought them. It was called the SteadiDrone Mavrik. Why would Yuneec mass produce a model of something that already failed to capture market share? I think Yuneec is changing directions to focus on the commercial market as that's where better margins and more sophisticated customers will be found. The consumer drone market is pretty tough, being driven by having the most features at the lowest price, used by a customer base that manages to send far too many back for "warranty" repairs, which neuters profitability. It's not about making customers happy as it is about making money. As for the Intel investment, there are some that believe the money was not intended for obstacle avoidance in the grand scheme of things. That investment may well have been directed at locking up an offshore manufacturing facility for more sophisticated systems under a different brand name.
At our price level what we have for terrain following doesn't work all that well much more than 6-10' off a surface. Sonar and optical flow suffer excessive disbursement and ability to maintain a specific distance from a surface goes out the window. The real good stuff, read that as expensive, uses a radar altimeter or a laser rangefinder, with either slaved to a very fast response auto pilot. We could have the super good stuff but the penalties paid in price and weight are high.
As for the front, upper mounted camera for inspection work, that has already been done with another system, that incorporated very similar mapping features as the 520. It worked very well, used a 24MPXL Sony a6000 for the payload and Pixhawk for the FC. Flight times of 15-22 minutes were common but at a bit over $4k few bought them. It was called the SteadiDrone Mavrik. Why would Yuneec mass produce a model of something that already failed to capture market share? I think Yuneec is changing directions to focus on the commercial market as that's where better margins and more sophisticated customers will be found. The consumer drone market is pretty tough, being driven by having the most features at the lowest price, used by a customer base that manages to send far too many back for "warranty" repairs, which neuters profitability. It's not about making customers happy as it is about making money. As for the Intel investment, there are some that believe the money was not intended for obstacle avoidance in the grand scheme of things. That investment may well have been directed at locking up an offshore manufacturing facility for more sophisticated systems under a different brand name.
As Pat says, without very much more expensive hardware, terrain following is difficult. Intel dropped the ball on Realsense, but even if they had developed more software for it, it's limited to relatively slow flight speeds and is designed for collision avoidance rather than maintaining a specific height.
The SDK will make it possible to plan missions using high quality terrain datasets. *However* before you get too excited, I have to point out that most terrain datasets are far from robust. Often it's a good starting point to talk about what you want to achieve (ie. capturing images of scrub land for agricultural surveys) and work out how best to achieve it rather than looking at how you can make a specific technology work. As usual, if you have a specific technical requirement, it should be possible to solve it - though some solutions require more work (and therefore more investment - time and money) than others.
The SDK will make it possible to plan missions using high quality terrain datasets. *However* before you get too excited, I have to point out that most terrain datasets are far from robust. Often it's a good starting point to talk about what you want to achieve (ie. capturing images of scrub land for agricultural surveys) and work out how best to achieve it rather than looking at how you can make a specific technology work. As usual, if you have a specific technical requirement, it should be possible to solve it - though some solutions require more work (and therefore more investment - time and money) than others.
arruntus
Premium Pilot
- Joined
- Aug 19, 2017
- Messages
- 1,851
- Reaction score
- 730
Totally agree, to monitor the terrain that is "useful" at low altitude requires an investment in hardware.
For following the terrain when you want to make photogrammetric flights, a main function of the H520, there is no need to invest in hardware since the flight height is usually around 100m. What is needed is a software investment, adapt the DataPilot so that it can do this monitoring by taking the data for example from the SRTM database. Px4 already has this operational function.
It would be nice if DataPilot had it too
For following the terrain when you want to make photogrammetric flights, a main function of the H520, there is no need to invest in hardware since the flight height is usually around 100m. What is needed is a software investment, adapt the DataPilot so that it can do this monitoring by taking the data for example from the SRTM database. Px4 already has this operational function.
It would be nice if DataPilot had it too
hey tom that interesting come to think about terrestrial imaging did something like that for the h920+ double carmera mount for fixed forward cgo3+ and fully movement bottom camera of you choice. but requires 2 st16+ controllers one for the pilot for fix cam and second for camera man for the fully functional bottom camera. Your application require bottom custom gimbal mount (while leaving the bottom mounting plate for the drone unmodified) to be wired for forward fixed facing only and the top mount to be custom wired harness that runs from the custom gimbal to a top mount plate for a fully functional gimbal. note must be done in manor to prevent compass and gps issues. and since the gimbal mount is flipped the picture will be upside down (unless you flip the camera head 180 degrees on the camera side mount which will be hard to do) on a right side up controller viewing the camera also upside down when recording requiring software to flip the image over.
Similar threads
