Hi guys - I attempted some model assembling including a rough design for the main body and there are several distancing issues - mainly because the feet brackets are super long due to propeller size so the body has to be made either pretty big (we’re talking about 15+ cm) or the legs will have to be shortened, etc…
I have some ideas so I’ll be spending this weekend working on a (hopefully) new and improved, smaller design which should work just as well. :mrgreen:
AKdaBAOUS
Yes, this was my concern. Thats why I asked about the distance between coxa joints.
Look forward to seeing the developments.
Ps, sorry for that extreamly large image attachment. Ill sort that once I get on my pc tomorrow. Currently on TabletPC.
This new design should hopefully look nicer too. 
AKdaBAOUS
Hi guys - I just wanted to check a few things with you before I continued. :mrgreen:
In terms of servos, I’m planning on using HS-225BB’s for the body servos and HS-485HB’s for the legs - or do you think I should go for something like 645’s for the legs?
In terms of batteries, I was going for something along the lines of an 11.1V LiPo which provides a fair amount of current. I haven’t found a battery pack yet though. ^___^
Any advice on choosing an ESC in general? :mrgreen:
Thanks in advance! ^___^
AKdaBAOUS
For esc you need something that is very fast refreshing.
I use mostly esc that are flashed with a custom firmware.
LiPo will be chosen with weight, motors
645’s for the tibia and femur minimum.
May get away with 485’s for the coxas… Minimum.
Again really depends on how you construct it and from what material.
Cool, thanks for the info guys. :mrgreen:
Um, at the moment, material wise, I have an idea to use different materials, etc. for different parts. I was planning on having the foot bracket made out of some lightweight plastic with holes drilled into it to make it lighter (but not soo many that it’ll make it weak and likely to snap); the servo brackets were going to be 3D printed because it would make this new layout idea much easier to assemble. :mrgreen:
The leg brackets were going to be made from Carbon Fiber and I’m still thinking of what the connector (the circular part between both faces of the leg bracket) could be made out of …
As for the main body, I had an idea which involved machining the body shape into some Styrofoam and then perhaps use some kind of coating method to strengthen the object (something like fiberglass coating, etc.).
Finally, the docking system was going to be made out of Forex - that’s very light and it works quite well for the job it’ll play in the docking system. ^___^
AKdaBAOUS
Just a small update - after some consideration, I’ve decided to downgrade the hexapod into a quadruped on the grounds that the removal of the two legs will lighten the robot by about 300g which doesn’t really seem like much but it’s better than nothing. ^___^
I’m still designing the robot to be able to incorporate the extra two legs so if the need arises, they can be added in. :mrgreen:
AKdaBAOUS
Hi everyone! ^___^
So unfortunately, the weekend wasn’t as productive as I had hoped, however, I have managed to sketch an updated design for the leg along with a rough-ish CAD model. :mrgreen:
With this new layout, the foot bracket can keep it’s 30cm length but at the same time, fit into the docking system quite nicely.
The flat part on the top of the foot bracket is where the “key” will go - I haven’t had a chance to add that yet. ^___^
What do you guys think?
AKdaBAOUS
P.S. The leg bracket in the CAD model has just been recycled from the previous models since I didn’t have time to draw up the new one so it doesn’t quite match with the sketch. ^___^
hay nice.
i like the way you have solve the tibia/motor mount.
2DOF…
can you show us a render of how the leg will look when in flight mode.
It may just be me but it doesn’t look like there is clearance for the leg to fold up.
i could be wrong.
Sure - give me a little while so I can draw up the new leg bracket and neaten things up. :mrgreen:
Ok, sorry it took it a while but here it is. ^___^
So we have the new leg bracket:
The leg in it’s normal standing position:
And then the leg in it’s raised “flying” position:
AKdaBAOUS
P.S. Yeah, I was going for a 2 D.O.F to keep everything simple … then again, if it’s going to be a 2 D.O.F, I should probably keep it as a hexapod … :mrgreen:
Do you recommend moving up to 3 D.O.F?
Hi AK,
Your last renders are very interesting. However, I must admit that I am now getting rather lost about how the main body can attach to the hip servo with the leg folded so closely over and behind the hip servo and also surrounding it on two sides. I am envisioning a servo bracket of long thin “fingers” capable of reaching inside the “hollow” folded leg to where the “fingers” attach to the hip servo. The “fingers” would run parallel to the main body far enough back to where they could clear the folded leg. Then they could turn and fasten to the body. Similar mounting and clearance considerations would seem to occur at the other end of your new joint where the remaining parts of the leg fasten to that slotted circle.
Does any of what I am envisioning even begin to approach the way you see things?
A fully rendered, animated CAD drawing would be nice. But even a quick hand drawn sketch showing the rough shapes of what you have in mind would help keep your disciples from straying too far afield.
Anyway, keep at it! Your projects are exciting.
Ted
Hi RoboTed,
Um, yeah, my idea is pretty much along those lines but like you said, it’s pretty difficult to explain without any full concept designs - I’ll spend some time over the next few days drawing up a full model of the robot (with some animations too - hopefully). ^___^
AKdaBAOUS
Hi guys! :mrgreen:
So for some random reason I woke up really early this morning so I decided to start working on the animation, etc. and everything is more or less finished now.
Just a note before I move on - I haven’t yet gotten a design for the main body of the robot (although it’s shape will be based on something similar to the Docking System) so the body servos are kinda just hovering there at the moment so you’ll just have to assume they’re being held in place by the body - apart from that, everything else is cool. :mrgreen:
Ok then, here we go - oh, and enjoy! :mrgreen:
Here’s a (yet again ) updated leg bracket design followed by the foot-rotor bracket:
I think it would be better if I posted a video of the leg over lots of pictures - it’ll be nicer.
Here we have the Docking System - pretty much the same layout as the other ones ^___^:
Um … yes, here we have robot in it’s “0” standing position:
Followed by it in it’s Flying Mode:
Now I come onto the animated video for the walking-flying mode transformation - I’m sorry about this but apparently SketchUp doesn’t do ‘proper’ animations so I had to do some hiding layer thing to get it to move so it’s a bit basic. 
The leg animation came out quite nicely because it was just orbiting around the model.
And just throwing this out there - here’s the servo bracket I’ll be using:
Hm … yes, I think that’s about it …
Just a few things …
The main body is going to be a square of side length ~30cm - as far as height goes, it won’t be more than 6 or 7 cm. ^___^
Also, would you guys like videos of the robot in it’s standing and flying mode being orbited?
AKdaBAOUS
P.S. Any questions?
P.P.S. I think it actually looks quite cool now. :mrgreen:
Loving this akdabaous. Most exciting project on here since Zenta’s Morphex.
Great design ideas.
And yes… more videos. 
I think you have found yourself. Just need to build it. No pressure.
Well, I certainly am glad you finally got that terrible Roomba with desk fan idea out of your head!
I have to agree with Jonny about the excitement value of your projects. They’ve all been good, but this is the best yet!
Thank you for these renders. They have really adjusted my view of what I thought you were doing. (I thought the long parts of the bracket supporting the leg/rotors would be parallel to the ground in the raised position.) Your idea of having those parts at about 45 deg to the ground and the motor mount at another 45 deg to bring the rotor shafts vertical is a much better concept. It provides much more clearance for everything which means you aren’t wasting space shrouding things inside each other. You can keep parts shorter and lighter than what I envisioned and with much less chance of them accidentally binding up like an old typewriter. (You can Google ‘typewriters’; Wikipedia probably has them under Bronze Age Technology.) You also accidentally enlightened me about your docking idea.
Definitely more videos. Even the quirky Sketch-Up worked well in your hands.
You seem to be headed for a practical payload design. I can see it as a Sky Crane or a GoPRO video platform.
Suggest you enlist DIalFONzo to keep you honest on thrust, and lift, and multicopter control factors. I think this is way beyond a Lynxmotion PS2 controller.
Ted
Thanks guys! :mrgreen:
Hopefully it’ll be even more exciting once the production is started. :mrgreen:
Of course - easy peasy lemon squeezy.
I think the world wasn’t ready for that much awesomeness yet 
… some day I hope. :mrgreen:
Yep, DiaLFonZo’s expertise will be required if the robot is to get off the ground. ^___^
Here are some videos of the robot in Standing and Flying Mode - Enjoy! ^___^
I have something to ask …
If I’m right in thinking, since this is a quadcopter, the weight of the robot will be shared equally by the four motors?
So estimating the mass of the robot (in a worse case scenario) to be 3kg, each motor would require a thrust of at least 0.75kg? Although I’m guessing it’ll be better to have something with about 1.5-2 times as much so the motors won’t have to be maxed out to be able to lift it?
So something with, say … 1.1-1.3kg of thrust would be good? Or am I totally off here?
Thanks in advance! ^___^
AKdaBAOUS
P.S. Just a question about implementing sensors …
I’ve got an idea which involves adding in some kind of distance sensor to the base of the robot which will help with landing the quadcopter in autonomous mode - it’ll check the distance between the robot and the ground and as the robot comes closer, it’ll reduce the power to the motors so the landing can be nice and soft rather than it plummeting out of the sky - which sensor would you guys recommend I use - IR or Ultrasonic?
This is just an idea - I’ll design the robot to incorporate everything but I’ll leave it out until I’ve gotten the main things sorted. ^___^
P.P.S. Some of the modifications I have planned:
A few thoughts:
multi-rotor copters have a tendency to vibrate if there is even the slightest screw loose. Given that the rotors are being mounted at the end of a leg rather than directly to the body may mean there is a LOT of things which can be even slightly misaligned and cause vibration - which will loosen screws, causing even more vibration etc. You might want to start with something simpler (described at the bottom).
Hopefully some others can weigh in, but I seem to recall that 3 bladed rotors (better suited for horizontal flight) are less efficient (although more powerful) than 2 bladed in a vertical lift situation?
In order to keep each rotor horizontal in flight, the servos will be powered and working hard - the prop wants to pivot back over the leg, and the servo has to counteract this. Assuming the distance between the “knee” and the motor is 5cm, and the thrust produced by the motor is 1.5 Kg, that servo needs to provide (a minimum of) 7.5 kg-cm . The servo at the shoulder needs to provide torque in flight as well.
Correct. Each motor + rotor combo needs to lift 1/4 of the total weight of the robot (and more in order to accelerate upwards)
4.4 Kg of thrust for a 3Kg robot should be ok in low winds
Don’t know how much noise you’ll get from the blades and if they might interfere with the ultrasonic sensor, and how much effect ambient light will have on the IR sensor (outdoors). There are many copters which use ultrasonic / sonar, so it might be the way to go.
Just an idea: Have the rotors connected to the shoulder (one degree of freedom only - cw/ccw) This means there is almost no stress on the servo in flight. Add one servo to the arm itself, separate from the connector between the motor and the body - this would add a second degree of freedom, but not add further complexity to the arm. This would create an 8DoF quad + quadcopter which lasts longer in the air or on the ground.
Thanks for all the info! ^___^
Now to design some more. :mrgreen:
AKdaBAOUS
