Hexapod Makeover

[innerbreed]

SWWEEET! i really like this.

yeah the tibia is very eye catching but now the whole this is. it looks really good. well done.

i think the tibias are good as there not as bulky and heavy looking like others iv seen. looks very nimble.
maybe you could intergrate some stronger. how about just bigger threaded rod. you could also make then a tad shorter, this will help.

[BillB]

Looks fabulous. Great the design, and the choice of colour works very well.

The great thing about the Robotis servos is that they can be daisy chained togther, so there is not a spagetti mess techicolor wires.

[blegas78]

i think the tibias are good as there not as bulky and heavy looking like others iv seen. looks very nimble.
maybe you could intergrate some stronger. how about just bigger threaded rod. you could also make then a tad shorter, this will help.

This is one of the ideas I had, as there is a tad bit of slop between the brass and the thread rod. Perhaps a metric thread rod would fit better, so I don't have to remake all of the brass pieces.

Looks fabulous. Great the design, and the choice of colour works very well.

The great thing about the Robotis servos is that they can be daisy chained togther, so there is not a spagetti mess techicolor wires.

I like the color too, it's my favorite color that the machine offers. Unfortunately, I used up all of the material for this color, so until we decide to buy more cartridges, I will be printing the material in white, which I'm not a big fan of...

[blegas78]

I'm very curious about your plans for the head and all those crammed servos.

Time to see what this baby can do...

On the top and bottom plate towards the front, you can see two holes. This is where I plan to mount a geared bracket system:


This is more of a proof of concept than it is a final design. the two holes in the back line up with the holes on the top and bottom plate.

The square frame that holds all of the gears rotates up and down, giving the camera tilt.

The shaft on the right connects to the right gear, which rotates the second largest gear on the bottom. This gear is mounted to the inner square frame. When I rotate the right shaft, the result is a pan motion for the camera.

The left shaft connects to the left large gear, which is meshed with the lowest gear. The lowest gear is connected by a shaft to the inner small gear. There is then an identical gear meshed with this, which is connected to the shaft that goes to the front. The rotation of the left shaft results in rotation of the camera.

[wattsie]

Hi,

Love the idea, are you going to be motorising the movements by electric motors?

Having a closer look, if you rotate the right hand shaft to pan the camera, wont this also produce a rotation of the camera? as the lower bevel gear will not be rotating and this would make the 'rotation' shaft rotate through the panning action?

Also, tilting the camera will create a rotation and panning movement unless this is compensated using a negative pan and rotate to balance out the transfered rotations.

Intrigued to see this in action though, looks pretty cool!

wattsie

[tom_chang79]

sexy!

sexy +1 more!

I LOVE the custom job on it and the legs... Skinny legs on a hexapod definitely earns a gold star!!!

[blegas78]

Well, School has started, so things have been a bit slow. I have printed the gear setup, and for my first time designing gears, I am amazed at how well they fit together! They are a bit stiff, so I might have to build a little setup with a DC motor to break in the gears.

Lately I have been working on getting it walking instead of head design. I think I may use a PS3 sixaxis controller in bluetooth mode to operate the hexapod. Since I have a full computer onboard, it's easy enough to add a bluetooth dongle.

Hi,

Love the idea, are you going to be motorising the movements by electric motors?

Having a closer look, if you rotate the right hand shaft to pan the camera, wont this also produce a rotation of the camera? as the lower bevel gear will not be rotating and this would make the 'rotation' shaft rotate through the panning action?

Also, tilting the camera will create a rotation and panning movement unless this is compensated using a negative pan and rotate to balance out the transfered rotations.

Intrigued to see this in action though, looks pretty cool!

wattsie

I will be making some sort of linkage to operate the different axis. The motors are mounted in body already, as can be seen in the previous pictures.

You are right, if I move one of the axis, then that will cause others to move. This isn't a major error though, because it is easy enough to do some more kinematics calculations to make sure the other axis stay still as only one moves. I think it will be much easier than the inverse kinematics for all of the legs.

[blegas78]

Here's a quick video of the hexapod. It's a terrible dynamic walking gait right now, as you will see when the hexapod translates and almost trips on itself. I haven't been polishing off the dynamics yet since I have been focusing on getting hardware situated. My board also tends to have a problem where communication to the RX series motors cut out. This probably has to do with fact that I stuffed a wire-wrap board in the body. I am close to getting the PS3 sixaxis controller to operate the hex.

I hope you all get a good laugh when it accidentally kills itself.

[Zenta]

Great to see it moves! I see the movement need some tuning, but overall it looks awesome.
And you are correct, the last part made me laugh

-Zenta

[blegas78]

Thanks!

The right middle leg's tibia motor also had a loose signal connection, so it was little flaky, especially towards the end when it looks like it pauses. I fixed it and it now works great. Maybe I can record another video tomorrow, with a little bit better dynamics.