The Perfect Camera System - Can anyone do this?

I have spent 7 years messing with drones.  I've used them from a 500lb UAV down to 55g of weight.  I know how to design them and conduct operations from mapping to SAR.  I'm more than knee deep in this stuff.  In all this time I've yet to see a camera that does it all.  I also know that there are people out there working on this problem and I also acknowledge whole-heartedly  that I don't know everything. 

In that spirit I'd like to describe what my version of the perfect camera would be and maybe have the community here chime in and make a better description. The ultimate goal here would be for someone to build one of these at a cost that would be reasonable.  I'll be as descriptive as possible.

1.  Onboard GPS - Having the ability to connect your own GPS receiver (like Red Edge has) allows the user to use their own accurate GPS antenna instead of relying on one that is out of date and impossible to remove from the camera.  This also removes manufacturing cost and size of the camera while allowing the user to place the GPS receiver in a much better location for reception.

2.  Size - Obviously the smaller the better, but a form factor of the Parrot Sequoia or GoPro would be ideal, the max size would be around the Canon S100 (roughly 4x2x3 inches).  This allows fixed wing operators to place payloads in an already cramped fuselage.  Multi rotor types have plenty of room due to "external" mounting of the camera. 

3.  Weight - Obviously, the lighter the better, but let's define light.  100g without sensors (GPS etc.) would be ideal.  Given the current Commercial Off the Shelf (COTS) airframes that most of us must use I can't see a camera weighing more than .7 pounds being easily carried by fixed wing and still have a useful flight time.  Multi rotors can carry a heavier payload but pay the same price as far as flight times.  A typical multirotor can only fly for 15-25 minutes so the lighter the payload the better for them as well.

4.  Shape -  This is perhaps the biggest overlooked factor.  Make it a simple shape, for example the A6000 is hard to mount due to all of the curved surfaces while the GoPro is perhaps the easiest to mount.  Most of us have to figure out a way to mount a payload by using tools you would find at home in order to carve fitting foam blocks or hacksaw a screw to "just the right size".  Simple shapes are best.

5.  Mounting options -  What would be counter productive is to have a bunch of mounting options that are excessive in the bulkiness department and even worse they would unnecessarily increase the weight of the payload.  Perhaps a novel way to provide an easy and SECURE way to mount the payload needs to be thought of.

6.  MegaPixels -  If Real Estate is all about Location then cameras are all about RESOLUTION, as in pixels per inch.  Resolution is accomplished mostly by having a high MP rating.  Minimum would be 16mp in order to beat what I consider to be the baseline camera system, the Canon S100.  Preferably it would be 20-25MP.  Higher than this is ok but then you start to run into problems with having to transfer huge amounts of data over narrow bandwidths.  Not too much, not too little.

7.  NO MOVING PARTS - The AW120 is a good example of this.  If your lens moves in and out then it will have a high chance of not surviving a crash.  Shock proof as best as possible is the goal.

8.  Ruggidizing -  Difficult, yes,  but I would want some water resistance and shock ratings if possible.  It's bumpy in the sky and VERY bumpy on landing too!  The casing does not need to be carbon fiber (please don't use carbon fiber) but something that is plastic is fine.  Maybe you can just provide the circuits and leave the body up to those that have rapid prototype machines (PolyPrinters for example).  That way we can share body designs and take the pressure off of the manufacturing process.

9.  Lens must be cleanable - Most are, but to be thorough, the lens must be accessible while the camera is mounted in order to remove debris or water vapor by hand.

10.  Battery Life -  Very surprising here, don't make one with an on board battery!  Make it so that i can use an EXTERNAL power supply!  You will never be able to fit a big enough battery for the flight time of my drone (or yours).  I can, however, tap the main power supply for my motor, step it down to the required voltage and power the camera for as long as the drone will fly.  This way the payload and the airframe are perfectly matched.

11.  Voltage - Put in reverse polarity protection and the ability to plug in up to a 6S LiHV (different than LiPo!).  This means you are providing the BEC (step down voltage regulator).

12.  Heat Sink - Nice to have, especially if you can make it modular and placeable in more than just one location on the camera.  I need that flexibility because I don't know how I'm going to mount it so I need options for cooling the camera.

13.  Shutter Speed - Nominal and useful is 1/2000th of a second, what I want is 1/4000.

14.  ISO - up to 6400 is fine as most of us can not/would not fly in the dark.

15.  WiFi - NOT NEEDED.  It does not have the bandwidth to push the pictures to my laptop fast enough.  It certainly wont beat an accessible SD card.  There might be another purpose for Bluetooth or WiFi data connections to the camera but I can always transfer the data "manually" faster than WiFi ever will, so save the complexity, room, cost and weight for the other requirements please.

16.  Multi spectral Ability -  As it turns out you can get more data (including data gained by NIR, G, RG and UV bands) by putting a normal RGB photo through filters.  While this is not as good or detailed as a dedicated camera it does serve the purpose (so far).  I'd make this an option but not put into the first generation as it is too specific of a payload type.  We won't even discuss an Infra Red Camera at all because that's a different beast altogether.

17.  Autopilot Triggerable - BIG one here.  I want my autopilot to be able to communicate to the camera in all respects.  I want my GPS data that I already have the hardware to get to be able to be used by the camera to know when to be triggered.  I expect a cable to run from the camera to the back of my autopilot to accomplish this.  This would also eliminate a need for the camera to have its own GPS unit but having the option would be HUGE.  It would also need a way to be triggered via IR or Hotshoe for those of us that don't want to use the autopilot to trigger it.  The more options the better and remember, a companion computer is going to be the standard for autopilots very very soon.

18.  1080 at 30-60fps is fine.  Most computers can't run 4K and anything above 2.7K tends to incur bandwidth issues.  Also a Video Out so I can hook up an On Screen Display is also a must.  You can also consider putting in your own OSD on the motherboard of the camera, I could live with that!

19.  MAKE THE OPERATING SYSTEM OPEN SOURCE!!!!  I want to be able to pluck from the world of programmers, not just from the one guy you hired at your company.  He won't be able to keep up.  Think of it like this, I want "sub programs" to be loadable, for example KAP UAV 3.6.

20.  Mechanical links must be secure!  Screwable connections (latching, etc) would be nice for cables to and from the camera.

21.  The ability to take a picture WHILE recording video with no visible disruption of the video.  Picture must be trigerable via an intervelometer system OR by the user remotely.

22. Field of View -  I don't want to have to add an extra step in processing to get the fish-eye effect out of my image.  Time is money.  90deg is probably good enough, 60deg is probably too low, 120deg or more is stupid.  Variable would be impossible I think, but important if possible.

23.  Format - RAW is needed by the few, most just need JPEG, please don't use anything anticipated to be "the next best thing coming up" as I need the operating system I have on my desktop to be able to use the photos immediately.

24.  Ability to use high capacity SD cards - Not lower than 32Gigs for sure, prefer around 64-128gigs currently for a camera that can do all of these things

25.  Lens quality - I know next to nothing about this aspect but I know that a cheap lens will negate all electronic efforts to get a good picture

26.  Magnetometer - nice to have and I'd like to pull that data from my autopilot so I am not having to do a separate calibration on my camera after I JUST DID ONE FOR MY AIRCRAFT!

27.  The little things:  I want the ability to remotely turn the camera off and automatically collapse the lens if it is a moving lens type.  This can help improve survivability of the camera if I know the bird is going to crash.  It can also save precious power needed to keep the aircraft in the air.

28.  Camera speed - HUGE HUGE HUGE.  I need to be able to take a pic once every second or better so I can fly my missions at a safe speed and/or full speed.  I need a full pic, not an unfocused "burst" that we have with normal cameras now.  Fully balanced and focused please.

Things absolutely not needed:

1.  Macros, for example: red eye reduction, panoramic and landscape macros.  I can't think of any I would need to run while the camera was going 40mph in the air.

2.  A lot of the control of standard camera settings in automatic modes tend to take good enough pics

3.  view finder - I am not going to be using the camera in a manner that would need one, besides everyone uses the 3" LCD to frame pics now.

4.  3" LCD, don't need the weight or cost increase on the ones that swivel out for the user.  Make this as big as possible but not mechanically complicated.

5.  Zoom - electronic zoom is ok, but manual zoom is useless for mapping and requires additional servos or control inputs.

6.  Flash - just flat out not needed for a lot of reasons

7.  Manual focus - I can't ever use this feature as I can't fly up to my drone myself

I'll add things that are mentioned as this thread progresses.  I'd love to know if anyone else wants something like this.

You need to be a member of diydrones to add comments!

Join diydrones

Email me when people reply –

Replies

  • Hi Gustavo,

    Any update on the result using RX0 ?

    Very much appreciate if you could share some of the images taken using the RX0.

     

    Gustavo Guevara said:

    I just ordered one and maybe next week will make a test and will post the results

    Waladi said:
    RX0 looks very promising for aerial mapping. Anyone tried it ?
  • Great Gustavo, waiting for your result. thanks.

    Gustavo Guevara said:

    I just ordered one and maybe next week will make a test and will post the results

    Waladi said:
    RX0 looks very promising for aerial mapping. Anyone tried it ?
  • I just ordered one and maybe next week will make a test and will post the results

    Waladi said:
    RX0 looks very promising for aerial mapping. Anyone tried it ?
  • Just notice the RX0 lens focal length that equivalent with 35mm, it mean that more strips on mapping mission compare with RX100
  • RX0 looks very promising for aerial mapping. Anyone tried it ?
  • Yes "Multi/Micro USB Terminal" as stated in the spec sheet. Can be triggered by relay.

  • Good post to read!

    Thanks.

  • I saw the new Sony RX0 (https://www.sony.com/electronics/cyber-shot-compact-cameras/dsc-rx0) and look great the Anti-distortion shutter of up to 1/32000 sec and 1" sensor, anyone know of its possible to shutter it from Pixhawk?

    Anti-distortion shutter

    Anti-distortion shutter of up to 1/32000 sec.

    Super-speedy shutter reduces distortion in shots of fast-moving subjects.
    [1] Conventional rolling shutter [2] RX0 anti-distortion shutter

  • If you have one, would you like to share your UAV pictures on here?

  • For what it is worth someone has made a third party controller for DJI X5/X3 gimbal.
This reply was deleted.

Activity

DIY Robocars via Twitter
RT @SmallpixelCar: Wrote a program to find the light positions at @circuitlaunch. Here is the hypothesis of the light locations updating ba…
16 hours ago
DIY Robocars via Twitter
RT @SmallpixelCar: Broke my @HokuyoUsa Lidar today. Luckily the non-cone localization, based on @a1k0n LightSLAM idea, works. It will help…
Thursday
DIY Robocars via Twitter
@gclue_akira CC @NVIDIAEmbedded
Wednesday
DIY Robocars via Twitter
RT @luxonis: OAK-D PoE Autonomous Vehicle (Courtesy of zonyl in our Discord: https://discord.gg/EPsZHkg9Nx) https://t.co/PNDewvJdrb
Wednesday
DIY Robocars via Twitter
RT @f1tenth: It is getting dark and rainy on the F1TENTH racetrack in the @LGSVLSimulator. Testing out the new flood lights for the racetra…
Wednesday
DIY Robocars via Twitter
RT @JoeSpeeds: Live Now! Alex of @IndyAChallenge winning @TU_Muenchen team talking about their racing strategy and open source @OpenRobotic…
Nov 20
DIY Robocars via Twitter
RT @DAVGtech: Live NOW! Alexander Wischnewski of Indy Autonomous Challenge winning TUM team talking racing @diyrobocars @Heavy02011 @Ottawa…
Nov 20
DIY Robocars via Twitter
Incredible training performance with Donkeycar https://www.youtube.com/watch?v=9yy7ASttw04
Nov 9
DIY Robocars via Twitter
RT @JoeSpeeds: Sat Nov 6 Virtual DonkeyCar (and other cars, too) Race. So bring any car? @diyrobocars @IndyAChallenge https://t.co/nZQTff5…
Oct 31
DIY Robocars via Twitter
RT @JoeSpeeds: @chr1sa awesomely scary to see in person as our $1M robot almost clipped the walls as it spun at 140mph. But it was also awe…
Oct 29
DIY Robocars via Twitter
RT @chr1sa: Hey, @a1k0n's amazing "localize by the ceiling lights" @diyrobocars made @hackaday! It's consistently been the fastest in our…
Oct 25
DIY Robocars via Twitter
RT @IMS: It’s only fitting that @BostonDynamics Spot is waving the green flag for today’s @IndyAChallenge! Watch LIVE 👉 https://t.co/NtKnO…
Oct 23
DIY Robocars via Twitter
RT @IndyAChallenge: Congratulations to @TU_Muenchen the winners of the historic @IndyAChallenge and $1M. The first autonomous racecar comp…
Oct 23
DIY Robocars via Twitter
RT @JoeSpeeds: 🏎@TU_Muenchen #ROS 2 @EclipseCyclone #DDS #Zenoh 137mph. Saturday 10am EDT @IndyAChallenge @Twitch http://indyautonomouschallenge.com/stream
Oct 23
DIY Robocars via Twitter
RT @DAVGtech: Another incident: https://t.co/G1pTxQug6B
Oct 23
DIY Robocars via Twitter
RT @DAVGtech: What a great way to connect why @diyrobocars community is so valuable and important! Have to start somewhere @IndyAChallenge…
Oct 23
More…