I’ve still not decided to just settle with 5 DOF or not. There are some stuff that bugs me a bit:
The gear backlash on the 5645’s is a real pain… Do some of you know if the titanium or steal gear’s are better. Or a better question, what standard sized servos has no or very little gear backlash (with enough torque of course)?
The hub used in the LPA’s are still to loose and wobble a bit. I did thighten them really hard and used locktite. Going for only 5 DOF would solve that problem though.
I made a main body/head with cockpit for Archer. I’m pretty satisfied with the result. But its a bit heavy (195 gram), making one of plastic (or vacuum forming) would work I think. With the increased weight the gear backlash gets even worse and gets more unstable. Again, going for 5 DOF would make the result better. The problem with only 5 DOF is mainly turning, but I’ve seen many bipeds doing that fine though. Anyone with some tip about the 5 DOF turning while walking part?
We can never be sure a product will work in every configuration of course. On this part it might work better with a 3mm screw in the center. They have finer threads and this should result in more holding power. Adding a washer may help as well.
The chassis looks cool. my opinion is that its too big and as you said a lighter one would be better. i think it looks cool with the lights moving at the front. would look great with Jame’s gun turrets on top, like he did for the brat.
Always great to see your work. the craftsmanship is always impressive. Ever thought about designing and selling your own line of parts?
Thanks for your comment Jonny. Yeah, I do agree that the body is a bit too big. That struck my mind after assembling it. I do have a completely different idea for a body too, I might try that next time.
This is about the question on servo backlash. In general metal gears have more backlash than plastic gears. Over time metal gears will get more backlash as they wear while the plastic doesn’t wear so much. I guess the only middle ground would be the karbonite gears. They should wear pretty well and have decent backlash(everything has some) but still come in useable torque ranges.
I just got home today from a vacation (and leaving again in two days). So I had to try out making the 6. DOF and the hip rotation (yaw) to work. I got very inspired watching storm2k5’s work on the IK for a scout during my vacation so I’ve spent some time tinkering and writing code. So it was a bit exciting testing out the new code when I got home. I’m using a global rotation matrix for the body at first and then do a local rotation for each leg. It seem to work ok. By doing the global body rotation at first I have one common center-point for rotations, and can easily change the coordinates for the rotation point. But I’ve realized that the feet math isn’t perfect, especially when it comes to the roll ankle joint. It get visible when doing different combination of rotations. I think I’ll have to do some thinking around the feet (end-effector) math. I do wan’t the kinematics to work perfect, so at the moment I’ve to avoid some combinations of rotation.
I also have a new idea for a simple head design. But I won’t be able to work on that until September.
Sounds like you are having too much fun! Will be interesting to see what you come up with.
On a side note, I know I have mentioned webbotlib and the like on the Axon2. The same guy that develops Webotlib is also doing some gait design software. Just today someone posted a video using it showing it do dynamic adjustments to gait. Just in case you wish to take a look: societyofrobots.com/robotfor … 2.msg88807
The kinematics are improved but still not perfect but much better than it was since my last post. The body rotation work fine. Lately I’ve also made some improvements to side-walking and “walk and turn on place”. Altogether I’m pretty satisfied with the result. I just had it walking around in our living room for about 10 minutes!
I hope to post a video before I leave for my little vacation to the Netherlands…
Thanks, I did have a very nice vacation.
There was several improvements to be done when it comes to side-walking and turning, mainly avoiding the feet to crash into each-other. Ex: When starting to side-walk to the left the left leg are the first leg to be lifted. I’ve spent some time making a gait method. Its based upon Xan’s Gait sub with a lot of changes.
I mentioned before that I’ve started a blog with great help from Dale at Basic Micro. The blog are still under construction and its not updated to my current work. My main goal with the blog is to unify all robotic projects I’ve been working on, so thats its easier for everyone to follow my work instead of scrolling through endless amount of forum posts. A link to the blog: robot-kits.org/
Hopefully someone find it interesting. I do hope to get some more time for blogging. My challenges (and I guess for many others too) is getting free time for everything. If I get some free hours I prefer to spend them on making robots (or writing code) instead of typing information on a blog. So I guess it will take some time before all projects on the blog are updated though.
Thanks for your tip Kurt! I did try the java app but somehow I didn’t get it to work …
I’m not sure if he is using inverse kinematics, to be honest I’ve not studied his work yet. But it looked very interesting though.
Very cool blog, very good idea too. people can follow this very easily. its good that a blog can give you nice structure and form when searching though it.
A-Pod looks very cool holding that Atom chip in its mandibles. nice photo.
I think its a good idea posting your blog, and im sure it wont take up your time needed for your projects. you always seem to take your time by creating professional looking designs and parts. your talent is definitely shown in your work.
We can only sit back and see what comes walking out from your workshop.
Iv also been looking at gait methods regarding my t-hex inspired quad. it’s really nice working on a project that doesn’t just involve simple gaits. ie hexapods.
I have tried to explain to some of my friends “and also to my wife” (but that conversation didn’t last long) that the more legs you have the easier it is to get walking. Thats probably why i like quadrupeds. Ok its not as much of a challenge as i biped but fun all the same. I mean some of the factors involved in bipeds also need to be applied to quads.
i wish you all the best with any of your future projects. We look forward to seeing what else you have in mind.
Ps congratulation on the pregnancy. wish you the best mate.
Thanks a lot for your comment about the blog Jonny. I did just update it with a new post, but its still not updated to my last work though. You sure did a great job on your quad code! I’m not sure if I can promise a new video before I pay Xan a personal visit though.
Working with biped is a very different task and challenge than the hexapods. I must admit that my hearth belongs to them though
Here is one picture I took yesterday of Archer doing a combination of Yaw and Pitch rotation:
looks to me like its updated. good reading. i like the pictures to. i think adding tip and idea is a good idea for other robot followers. after a last few years of knowing you, i have noticed you are a Farly good teacher so i think you should implement some of this into your blog.
i like the new feet. makes it look more mech! 8)
yes xan said you guys were meeting up next week or something. iv been testing Terrain Adaption code this evening. im getting close, i hope!
Ps glad you have added a link in your signature. was a pain minimizing windows to get to the shortcut icon for your blog on my desktop.
I’ve had some robotic time now and then lately. When it comes to Archer I’m a little stuck since I’ve still some issues with the BB hub used for the LPA-01. Beside that it perform OK and the single leg control also work fine.
After fiddling around with a biped I started missing working with hexapods simply because they are so much easier to make walk and perform so much more natural and smooth. Yesterday I did some changes to the gait code for Phoenix, beside supporting NrLiftedPos= 1,2 and 3 it also support 5 lifting states. You may ask whats the point in that? The main reason is that I wanted to create another hexapod gait method. I’m calling the new method for Triple gait, its basically a tripod gait but only one leg are lifted or lowered at the time (within the tripod). Its like combining the wave and tripod gait. It was possible to use the 3 stated (NrLiftedPos=3) method but 5 lifting states gave a better result (IMHO).
I hope to post a video of that too when shooting a new workshop video of the Archer project, I just want to replace that BB hub first (hint hint Jim… ). The new gait method uses 16 steps and even that is pretty fast thanks to all the improvements done by Xan and Kurt.
I luckily got some robotic time in my workshop today. Inspired by Jonny’s work on his latest biped I decided to replace the short and simple femur on Archer with a ASB-10 (long C) + ASB-11 (offset) brackets. I do like this combination of brackets as a femur part for a hexapod (like T-hex) and here as the femur part for a biped. By using these brackets the legs got much more freedom, the ankle and the hip joint can almost touch each-other:
So far I’m very impressed by the servos, after about 8 minutes of walking and body movement the servos doesn’t even get hot! They are not cold of course, but they seem to handle the load pretty well.
(I just answered your PM)
By doing the global body rotation at first I have one common center-point for rotations, and can easily change the coordinates for the rotation point. But I’ve realized that the feet math isn’t perfect, especially when it comes to the roll ankle joint. It get visible when doing different combination of rotations. I think I’ll have to do some thinking around the feet (end-effector) math. I do wan’t the kinematics to work perfect, so at the moment I’ve to avoid some combinations of rotation. 
…