Expedition drones

Hi,

 

After having flown the Parrot AR Drone, which cost less than 300 USD, and being interested and followed the development of drones for some years, an intriguing idea came to my mind.

 

I am leaving for several expeditions the coming years, and have been looking at a new business segment using drones for aerial photography during these - preferably finding someone that is interesting to join or take the lead in this development.

A little about the business idea:

The sport enthusiasts that are going skiing, kiting, climbing and so on, are typically people with good economy. They purchase and own typically one or more helmet cameras, posts their movie / experience to Facebook, Youtube and Vimeo, with the aim of sharing an unique day and get feedback. A new way of filming these actions, to stand out from the rest, would typically interest these people, if the product can be sold at a reasonable price, accepting lower than military standard on the equipment. Cheaper equipment can typically take a crash, change some "crash" parts, and be up and running again, in favour of the good shoots from air in 80% of the times (depending on operational criteria).

 

This could also be used by journalists as an add-on in dangerous areas without being afraid of losing the drone because of its low price, and one could easily bring spares if one is lost as well as interchangeable equipment if something is damaged. It can be competitive in the market for people that need helicopters for aerial photos. It can be used on ski arenas or resorts by reporters during competitions for aerial photography, either following a specific athlete (carryin a recco / rfid style identifying brick) - and so the ideas goes on. For a lot of these markets, the requirements could be lower than mine I presume. However, I plan to bring it for climbing above 8000m (if possible), and at the poles.

 

Some initial requirements:

 

- Must be light and small so it can be carried in a small box stored either a rucksack or a pulk (sled pulled behind the skier).

- For expeditions it should be able to fly for 30-60 minutes and be recharged using a 20-30 W solar array on the ground afterwards within 5-10 hours (solar array attached to the pulk).

- Must operate round the athlete in a fixed / predetermined pattern. If the athlete moves, the drone must move accordingly.

- Must be able to withstand a crash without expensive damage and be traceable

- Must be able to operate on its own from "start" is pressed (on e.g. a remote control), to autonomously land next to the athlete when "stop" is pressed.

- Emergency system or low battery bringing the drones autonomously back to the athlete, or landing safely as close as possible and sending out a "find me here" position.

- Must carry a camera for pictures and movie (can be the same). (does not need to be live streaming. Data can be stored onboard).

- Must be able to operate in wind (ideally up to 30 knots, but 20 knots and down to 15 knots is acceptable)

- Must be able to fly in a few different patterns so the video don't get "monotone" being only from one fixed angle following the athlete. Can typically be circular pattern within X radius, Figure 8 pattern within X radius, change of altitude in a fixed pattern within X radius of the athlete, and so on. There is only a need of some predefined patterns that can be chosen using e.g. a switch before it takes of.

- The drone must have a very intuitive and have a easy user level, and preferably be "launch and forget" style, before you bring it home.

 

I am open to discuss the requirements and make changes to them, if I can fit it into my plans. Anyone that can give some feedback on this of what to expect and a guesstimate on price vs reliability on this? How far are the drone industry from achieving something like this, and if something still remains to be solved to do this, how far is the industry from it? What typical components needs more development, which requirements do I have to make a compromise on?

 

Looking forward to a fruitful discussion about a new potential market and its requirements.

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

Join diydrones

Email me when people reply –

Replies

  • Nice to see you on the forum Olav.

    • Light is down to airframe choice. There are plenty to choose from. Raven military drone would probably suit.
    • 30-60min flight time is possible with big battery 5000mAH+. Charging would be a feat but doable in good conditions.
    • Loiter around a position. This is already implemented but having it moving is a bit more complex.
    • Most airframes are cheap and withstand crashes but make sure to get one that you know you can get spares for not just entirely new planes. EasyStar would be my choice as my friendly model place sells just parts. Tracking is made harder by your previous request of following your target as communication will be used for that. I'd suggest a GSM eval board just stuck inside but if you are out of service you are then out of luck.
    • Auto land and take off. Both are already implemented but few success stories have been published. Take off could simply be full throttle and stabilise. Landing is much more risky thanks to speed and solid interaction :D
    • Low battery failsafe is already in the coding but I am unsure of its status and use. If not what you want you could code it yourself to RTL.
    • Cameras add mass. More mass = more power needed which in turn means more mass. Its definitely possible but make sure to set boundaries so you don't get bulky. If you only want 480i or 720p resolution tiny camera off eBay work very well (I should know I use them on mine exactly like you want).
    • Wind means more power to cancel out its effects as well as instability. Glider airframes and trainers will help this part.
    • Pattern will be set but due to movement of the focus will never be exactly the same. Conceivably you could create a few "shots" for the setup though.
    • To be "muppet ready" is a BIG leap. APM would suit everything you say except this as its a developing platform. You could put your own section of coding in to push the ease boundary but inherently whatever you do will be broke by someone. This is the hard one.

     

     

    For your tracking of subject there are a few things to overcome. Tracking requires working out where the target is, this can be done visual (camera) or with GPS.

    Visual tracking is can be easy or can be complicated dependent on the technique used but is always processing intensive. The Blackfin camera could help you here as a separate system for your guidance.

    GPS tracking however is much more easy, you need the GPS co-ordinates, the cameras orientation in relation to the craft and a pan and tilt system for the camera. I have already started work on this and in home tests seem ok so far. Your problem is your target will be updated (sometimes VERY quickly) so you will not be able to focus it manually from a GCS you need a cheat a quick way to add it. My advice would be to use the telemetry port so instead of beaming data to the GCS so you can find it (as I said earlier) you could update the craft via a target mounted "GPS-Xbee box" and that way you could use the existing loiter code to your idea.

This reply was deleted.

Activity