Hewwo. I am going to build a four legged walking robot and need some insight. I want to make the legged statically stable and resemble animal legs (loosly resemble I know). I also need to know if there is someone out there who knows about Dynamic Balance Control. Thank you.

 

Loads of info at the link below:

http://www.oricomtech.com/misc/robolnk8.htm

__________________
Two roads diverged into a woods...i took the one less travelled by. R. Frost.

 

Currently I am working on a four legged robot too. There are 2 types of walking gait implementable on this kind of robot, static or dynamic stability. Which one are you asking?

I am working on static stability where the robot will have at least 3 legs on ground while the last leg is lifted and step forward. PM me if you want to discuss more about it.

Dynamically stabilised robot use less than 3 legs. At present moment of research, scientists have developed dynamic walking gait for 2 legs, either by trot or hop. I am not working on this as it involve high complexity and effort.

 

When you say static stable, you mean when standing still right? Not when walking? For static stable when walking, I think you have to be able to shift the COG, even when you only plan to always have 3 feet on the ground at all times, since it's probably a rectangular arrangement. You also need ankles (maybe feet) and knees so that it can move the lifted foot forward without ever tipping forward.

 

Yes dknguyen. A quadruped which employ static stability has the ability to remain statically stable at all time or every interval of the walking/crawling motion. This means it will use 3 legs to form a tripod while the COG moved but will always fall within the vicinity of the tripod, and the fourth leg is lifted. The challenge in walking gait development is that the tripod will alternate between the 4 legs. One way of doing this is to form tripod, pause, shift the leg, shift the COG, form tripod again and again. Another way is to make it move along without tipping off the tripod stance. A quadruped walking gait is different from hexapod. As a hexapod has 6 legs, it can form 2 pairs of alternating tripod.

THis involve advance dynamic analysis just to formulate the relationship between the foot and the actuator response. At this moment I am trying to acquire/design a circuit which is capable of controlling 12 motors at once, and i found myself a place (this forum!) where i can learn and ask for help.

 

I've avoided even considering quadropeds because I hate having to deal with things like ankle joints, or knee joints with multiple degrees of freedom.