[OpenRelief Developer] Airframe design (was: Re: introduction post)

[Ankur Kumar]

If you can get two (or three) cameras on the plane, it will make a lot of
computer vision tasks A LOT easier. We will be able to build point clouds
out of 2 cameras in real-time which is much faster/seamless than using
Structure From Motion methods.

On Sun, Jun 17, 2012 at 2:34 AM, Karl Lattimer <karl at qdh.org.uk> wrote:

>
> On 15 Jun 2012, at 10:13, Shane Coughlan wrote:
>
> Hi Mohammad
>
> On Jun 10, 2012, at 10:34 AM, Mohammad Athar wrote:
>
> Why was Blender chosen for airframe design?
>
>
> http://openrelief.org/pipermail/developer_openrelief.org/2012-April/000048.html
>
> It doesn't seem like there are many analysis tools in blender (FEA, fluid
> dynamics, etc).
>
> What advantages does it offer over solidworks?
>
> Any input on this would be appreciated.
>
>
> First of all, our prototypes are using off-the-shelf airframes, and we are
> learning from that. Step two is to create a open hardware airframe, and
> Karl has begun the brainstorming process on that.
>
> As part of this Karl is using Blender to put together thoughts regarding
> the shape of the airframe. I am sure he would appreciate your contribution
> if you have experience in fluid dynamics (etc.) and access to other tools
> that are more targeted towards detailed models.
>
>
> At the moment the airframe model is simply a shape design which takes some
> aspects of our existing airframe and extends and improves on it. We're
> using blender because I don't have any decent CAD software and its been a
> long time since i used CAD.
>
> The idea is to effectively build a scale model of the airframe for testing
> - we will hopefully be able to print it shortly and test it with some basic
> materials.
>
> I've tried to model the shape on a dragon fly as it would give us a unique
> appearance and a very slippery shape. The wings are a slightly improved
> version of the FPV 168 in that they have a slightly sharper rise to the
> ends and longer and wider elevators/ailerons - we also opted for a V-Shaped
> tail to reduce complexity and simplify the build.
>
> What I'm missing :-
>
> * The capability to slice and section the airframe accurately
> * The capability to produce high quality graphics of the airframe
> * Vorlax or similar tool for testing the pressure zones around the
> airframe
> * A laser cutter for cutting plywood
>
> What needs to be done :-
>
> * Design of wing attachment parts - a piece which attaches the wing to the
> fuselage, it must be strong enough to hold the weight of the wing
> * Design of tail attachment parts - similar deal with holding the two tail
> wings to the central column
> * Design of the overall fuselage internal structure
> * Design of the internal ribbing and structure of the wing
> * Detailing the assembly process from the above parts.
> * Design of camera mounts (two or three cameras possibly with IR/RGB
> Fisheyes underneath and a nose mounted first person view)
> * Design of servo mounts, preferably OpenSCAD printable parts -
> constructed from two pieces, one is the holder the second is the cover,
> these two parts screw together and into the wing.
>
> Specifications :-
>
> * Aircraft must fit into a box no larger that 50cm x 30cm x 80cm
> * Aircraft should be able to carry some weight in the tail section
> * Aircraft should be able to carry some weight under the fuselage
> * Total maximum load to carry should be no more than 2kg - excluding what
> is needed to make it fly (computers etc...)
>
>
> Variations on this basic design, re-using the wing sections are also in
> the works, I haven't managed to settled on those airframe configurations as
> yet.
>
>
> It seems the UAV is in production
>
>
> http://openrelief.org/pipermail/developer_openrelief.org/2012-April/000047.html
>
>
> http://openrelief.org/pipermail/developer_openrelief.org/2012-April/000025.html
>
> Where can I get specs on this model?  (I.E. maximum payload, maximum
> speed, estimated endurance, etc.)
>
>
> http://openrelief.org/pipermail/developer_openrelief.org/2012-May/000146.html
>
> is a design report available?
>
>
> The first two prototypes have been made. I believe Karl has documented the
> build process and intends to publish that documentation. The lessons
> learned can probably be usefully shared, so everyone can know what went
> wrong, what went right, and why that was.
>
>
> Assembly of the FPV 168 will be documented when I find time.
>
> We don't have specifics for lifting capacity or endurance yet, we hope to
> find that out in testing.
>
>
>
> in regards to:
> http://openrelief.org/pipermail/developer_openrelief.org/2012-April/000040.html
>
> Could prototyping-level processes be employed?
>
> If so, we could provide instructions for people to go build their own UAV
> at local shops/hackerpaces/garages.
>
> The other parts seem easy enough to source, and being able to build it
> would make for a less expensive entry cost.
>
>
> OpenRelief is a design project, so what we intend to do is provide the
> instructions for people or companies to build the various parts to make
> some or all of a drone.
>
> Plasti-dip isn't right.
>
>
> http://openrelief.org/pipermail/developer_openrelief.org/2012-May/000152.html
>
> We need something lightweight.  Decent results can be achieved by carbon
> fiber.  The best way to manufacture that is in an autoclave, but you can
> also use a negative cast to create the needed pressure.
>
>
> http://openrelief.org/pipermail/developer_openrelief.org/2012-May/000126.html
>
>
> http://openrelief.org/pipermail/developer_openrelief.org/2012-May/000114.html
>
>
> At the moment we are looking at airframe prototyping from fibreglass. It's
> not an ideal material but it can be sourced around the world, and it is
> pretty tough.
>
>
> BR,
> Karl
>
> _______________________________________________
> Developer mailing list
> Developer at openrelief.org
> http://openrelief.org/mailman/listinfo/developer_openrelief.org
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://openrelief.org/pipermail/developer_openrelief.org/attachments/20120617/e4643231/attachment-0001.html>