very smooth and Natural gait.
looks almost like a croc walk!
youtube.com/watch?v=XnmO0E2JGJ4
Yeah, very original leg design. Impressive movements too.
Great find!
Hi KÃ¥re,
Interesting gait!
I think I like your leg layout better, 'tho. Anyone else building one? I’m supposed to be designing/building a Micromouse 'bot, but I can’t get away from my legged projects!
Now I have some cash, but hardly any time to build anything.
Best regards,
Alan KM6VV
Very interesting video. The wheels on it kinda bug me though. I’m not even sure if they’re hooked up to anything. Sometimes it’s clear that they’re not being used, but other times…
i know what you mean about the wheels! they are actually used to help stablise the robot when turning at high speed to stop it from tipping.
there are motors attached to them and can allow it to convert to wheeled motion:
youtube.com/watch?v=EmRlHRKC2Gg
here is a closer look at the robot but unfortunatly there isnt much about the wheels. nice carbon fibre plates and servo brackets can be seen.
This hybrid design is so cool!
I have this sort of thing in mind for a hexapod/octapod robot. How does this do on carpet? My only problem with it is it has no ground clearance due to the way the legs are attached to the body. Maybe using a longer extension for the foot would be an alternative.
Are you going to share your design, especially for the legs? I bet something very similar could be designed using the SES. 
I’m already getting some ideas.
8-Dale
i am unfortunate to say this is not my design.!!
i agree.
after seeing this quad, i did put the leg design together using the SES to see how it would work. i can recontruct it again for you sometime if you wish?
it might be best just to look and study the legs as i did to replicate them.
this is basically a 2-wheel chassis with quad legs attached, right? it also looked like it had force sensors on the foot pads which is pretty cool.
This thing really walks great! Very fluent! I also watched one of his other 3D vids where he shows the triangle of the 3 feet being on the ground and the COG. This clearly visualize how the gait works! Good found!
Xan
I’m interested in how you built the leg from the SES. I’m thinking about a design of my own and think I have it figured out.
8-Dale
I’ve been experimenting with a design for the quadrapod leg, but I am coming out with a 4DOF or 5DOF design using SES parts. I suspect I just don’t have some of the brackets I would need to replicate your design using the SES. I keep running into places (on more than one of my designs for various things) where I need an “L” bracket that has the long side just a bit longer than the current ASB-06. I need something that will clear the end of an ASB-04 when the long end of the ASB-06 is attached to a servo mounted in it.
8-Dale
Here is an update on my version of the cool quadrapod leg:
I have a 3DOF design I believe will work well! It uses 3 ASB-04 multi-purpose brackets, two ASB-18 connector brackets, and two ASB-06 “L” brackets. I will CAD it up as soon as I can get back into Windows (it won’t switch down to the lower resolution of my 15" monitor).
Mounting the servos into two of the ASB-04’s is difficult because the ASB-06’s make a very tight clearance between the side and the servo. I need to get a proper vice for holding stuff as I am putting everything together.
8-Dale
kinda what i had…
Im having probs this end too. tried to open up my cad software and it crashed, so now i am putting it together in spare SES parts but im missing my main bit so i have also faild that too. i have taken two photos that need putting together…
the images join as presented. having the asb04 from the first image and the c bracket of the second joined together.
also missing is the 1/4 tube!
sorry for 
explaination!!!
good luck.
Pardon me, what does it mean by reverse swing gait? I try google up the term but I can’t find anything useful.
Sorry for wasting your time and miss leading you on the term i used. By reverse i mean the servos and not the gait its self. Most of the ses 3dof leg designs follow the same construction rule, having one vertical plain and two horizontal plains, where as in the video we can see these axis are reversed, having two V and one H! I hope this clears this up for you. Kind of like installing the norm ses configuration on a 90 degree angle tilt.
Iv logged in using my phone so sorry for miss spelt words. (thats dedication for ya)!
I see. That make a lot of sense now. 
This unique assembly actually has mechanical advantage over the conventional assembly. Theoretically, a low torque servo would be sufficient to support the weight! 
This is true.
i would expect that the servo arrangement would be
2x hs475’s and 1x hs645 per leg,
Rather than
1x hs475 and 2x hs645 per leg,
Big advantages…
Price,
Payload,
Stability,
Easier gait transistion,
Leg reach,
etc.
I do have plans to replicate the whole thing using SES some time soon!
Here are the servo spec for this robot:
S9602 x2
SPECIFICATIONS:
Class: Micro
Digital: No
Programmable: No
Voltage: 6.0v
General Specs:
- Bearings: Dual BB
- Gears: Metal
- 60° speed: 0.09 sec @ 4.8v
- 60° speed: 0.09 sec @ 6.0v
- Torque: 37.50 oz/in @ 4.8v
- Torque: 37.50 oz/in @ 6.0v
- 1.41 x 0.59 x 1.21 inches
- 1.09 ounces
S3102 x1
SPECIFICATIONS:
Class: Micro
Digital: No
Programmable: No
Voltage: 4.8v or 6.0v
General Specs:
- Bearings: Nylon
- Gears: Metal
- 60° speed: .25 sec @ 4.8v
- 60° speed: .20 sec @ 6.0v
- Torque: 51.40 oz/in @ 4.8v
- Torque: 64.20 oz/in @ 6.0v
– 1.10 x.51x1.17 inches - .74 ounces
In fact, 3 x HS475 would be enough for this leg assembly. The last length is only meant for Z-axis motion (aka the vertical component force). Thus the design could be optimize to maximize the servo torque.
In conventional design, the required torque is higher due to horizontal component force. The further the leg stretch, the more profound the torque required to bear the weight.
Another advantage of this unique leg, is that the IK equation would be much simpler and easier to compute. This theory is still unproven though, but theoretically, this is because of the Z coordinate is exclusively independent from the X-Y coordinate, if you get what I mean.
i know exactly what you mean.
This was the first thing i noticed! but like you say the theory is still unproven.
once i finish my current project i will open up this theory and test it.
i really want to start the build now, and as i have tons of SES parts and 20x HS475’s im very tempted! 