Here is a large prototype of a robot inspired by the Theo Jansen walkers. The four legs are controlled by servo and interfaced through a Parallax server controller. Because it is only a prototype, I am using the Parallax sumo bot microcontroller. The final bot will have a styren half circular shell with close proximity object detection towards it's left and right. Perhaps I will add a swivel mounted sonar for fun, but probably not as this is a prototype for a much smaller (less then 2"x2"x2") version. The legs are made of aluminum bars and fasteners. The chassis is made of Lexan (I think, it was in the scrap pile)
I was inspired to build this robot when I first saw Jansen's sand creatures. The organic motion of those legs captivated me.
My robot differs from Jansen's designs in that each leg has it's own pivot. This was to help experement, although on the final small design, i will use only two drives for all four legs.
The legs are not synched in anyway. I wanted to be able to create a more versatile gait to the robot, so instead of driving all the legs on one shaft, they have been given 4 independent full rotation servos
As far as the design, I believe that that I had taken the leg proportions from a Phun file where I had originally saw the design. I have since then mad a lot of experiments with the ratios of the smaller struts the connectors to the drive. Also, I have played with the throw of the drive shaft.
The name is a joke of of Theo from the Cosby show. Yes I know, it is a stretch.
As far as stability, the bot is very stable. The carriage has been expanded so now the bot is even more stable. there is a slight horizontal wobble to the legs but that will be corrected with a larger washer holding in the stationary pivot. Know what I mean? (I’m not sure if there is any Jansen teminalogy to explain the different joint of the legs, but there sure should be!)
And finally, yes I have seen your bot. That is one of the reasons I signed up for this site
About terminology: I use the letters from an original Jansen drawing which also dictated dimensions. I added my own letters for all the pivots. I’m sure I published them on this site. Somewhere. Use the tags Luke!
I found the drawings you were speaking about. My interpretation of the of the leg designs was much more rudimentary. By your numbers, my B C D F G lengths are all equal. So are my J and K lengths. My I length may also be the same dimensions as the J and K, with the H section meeting somewhere mid leg on it. My H and E are also equal in length. I am not sure if this has lead to an improvement in stability because I haven’t tested both yet, but I will hypothesize that it will because the arch of my legs is much shorter then in your design. This design as well as my new mini design are both quadrupeds, so keeping the legs stable is important.
What I gain though in stability will be lost in grace as the shorter gait of the bot will make a much less elegant look movement. For my projects that is ok, as I aesthetically I want the motion to be similar in look to that of an old windup toy.
Thanks for cluing me in on your measurements. I will use your terminology moving forward. The only number that I was unsure of on your design, was that of M. That is also a number I need to find my maximum of, because on this bot i think i played it rather on the safe side and could have had a much longer throw off of my servos. Any advice?
Make M too large and the whole system will seize up. Shorter is always safer. But too short will handicap your gait. Any good crank needs at least some radius. I tested mine with a variable prototype. Just an arm with multiple holes in i. Kinda multiple choice.
Make M the final adjustment of your leg. That’s all I can say. Any thoughts about feet yet? I have not developed that far yet. But they have been on my mind from the beginning.
