I believe I have stumbled onto a new type of biped, which I will call a bipod. It is just a pair of 3DOF SES hexapod type legs attached to a U channel body.
I noticed that this would actually be able to walk when I assembled the first 3DOF of my large hybrid biped called Mega Scout. Right now there is much more movement range to the rear than towards the front, but this could easily be corrected if I change back to long C brackets to hold the first two joints.
My bipod can be powered by a single 6.0V 1600 mAH battery pack and will be a 6 servo walker. I can even add the Mega Scout style sensor turret to the front of the body and the battery pack and electronics holders to the rear of the body without having to make the body longer.
I will post some CAD model pictures of this new design shortly.
Obviously this is still in the modelling stage, but Iām not sure how the ābipodā will walk. The CoG is very far from each foot - based on the CAD drawing even if the bipod leaned you would not be able to get the CoG above a foot to get it to balance properly. It is likely that you will need a VERY wide foot that extends underneath the body to be able to support a walking gait in this configuration. The other option would be to include the body as a āfootā to support the weight while one foot is in the air.
I,d have to agree ,if ur gonna have the leggs spaced so far a part you,ll to need a tail or an arm swinging back and forth to shift ur balance,mabe u could have it scream āALL IS LOST ALL IS LOSTā
I have a model of a version of the Bipod with the same type of foot as Mega Scout, but without the 2DOF movement. I kind of like the unique wide foot design.
I can see the Bipod walking in such a way as to lower the body all the way to the ground, pick up a leg, move it to the next step, pick up the other leg, move it to the next step, then pick the body off the grould and move it to be back over the legs. It would be a very strange sight to see it walk this way, but it would sure be interesting to implement.
Just put a foot on the bottom of the body and have both legs move at once.
For that matter why use legs? Use two arms that can be depressed far enough to lift the base and then you have a mobile two-armed robot without having to go all spendy for leg servos.
The wide feet should solve the problem nicely, just like for Mega Scout. I want to keep the legs at 4DOF so the Bipod is an eight servo walker, just a bit more than a BRAT, but less than a full scale biped like the Scout. I may remove the foot rotation and bring the Bipod down to a standard 3DOF type hexapod leg.
With the addition of the wider feet, will this Bipod will be able to walk now? I will eventually find out for sure once I have completed construction of the Bipod, but I am interested in finding out what others think also and what, if any modifications, would be suggested to enable full walking.
I can maybe see it ācrab-walkingā sideways, with both feet in a line, and shuffling along with a foot in front and one in back, but it seems to me that to try to walk front-to-back with those broad feet and the hips rotating like hexapod legs, youād need another lateral joint to rotate the ankle and counteract the rotation at the hip joint. Either that, or the bodyās facing will swing wildly from side to side with each step that it takes. Maybe a rotatry joint where the vertical tube joins either the center of the wide foot, or the ākneeā, but Iām not sure.
Itās one of those things where I think playing with it would be the best way to discover the abilities/limitations, and possibly new solutions to make it work in ways you didnāt think of before.
This is already a part of the leg. It is the fourth joint, and allows the foot to be rotated. Is this what you mean?
I tend to agree. This is a new design for a biped as far as I know, and sometimes it is really hard to visualize how it might walk. I do believe it can walk without falling, and am anxious to get it constructed so I can try it out.
I believe it will be possible to have the Bipod lean over far enough towards the grounded leg to allow it to take a step.
Itās good to at least have brackets to fiddle with when I donāt have servos to make a robot actually move. I was doing some fiddling to see just what range of motion the Bipod would have on the legs and came up with a new configuration. This new configuration of the legs makes the Bipod almost the same as a BRAT except for the lack of ankle rotation of the foot.
As much as I really want to keep the Bipod an eight servo walker, I am tempted to add that ankle movement. I will have to think on this a bit. I am not sure the Bipod could manage a reasonable walk in this configuration, which just rotates the hips and feet to reposition the legs.
Here are a couple pictures of the Bipod with the legs forward.
Just a suggestion, but you really need to closely study how other bipeds walk instead of just drawing pictures. The ankle servos usually are one of the key components for biped walking. The ankle servos generally tilt the biped side to side placing the the biped center of gravity over the standing leg so the other leg can take a step.
I do understand how bipeds walk. I just had to figure out how to get the ankle rotation in without increasing the number of DOF. I want to keep the Bipod as an eight servo walker. I believe I have solved this problem now.
Yes, they usually are, but that doesnāt necessarily mean it has to always be that way. I know and understand how the ankle servos of a biped work.
The Bipod is a project under development, and I have specific goals for it:
Keep the first three leg segments the same as a standard hexapod leg to see if this type of leg is viable in a biped.
Keep the number of DOF at 4 for each leg to keep the Bipod in between a BRAT and a Scout as far as size and mobility is concerned.
Itās just a matter of getting the motion of the legs to where the Bipod can walk reliably while staying within the goals I set for this project.
Iāve just completed a minor design change to the Bipodās legs, which I believe solves the walking problem and stays true to my goals for it. Iāll have new pictures and a description of what Iāve done up soon.
Iāve figured out how to get a standard biped stance out of this Bipod without adding any more servos. I just put the fourth servo and the feet in slightly different configurations. I had to go back to the narrow feet or it would be tripping over itself when it tries to walk in biped stance.
The crab type stance is proving to be a real challenge so far though, but I am not giving up yet.
I think I have finally solved the problem of the Bipod being able to lean over far enough in both the Hexapod Stance and regular Biped (forward walking) Stance. In Bipod v2.0, it could not lean at all when in the Hexapod Stance, but could lean properly in regular Biped (forward walking) Stance. I had to redesign the wide feet a bit so they would not trip over each other when in the Biped Stance, yet still allow the Bipod to lean over and balance when in the Hexapod Stance. I still have to make one adjustment to the legs so the ankle rotation servo will be closer to the foot, but that is an easy change. The Bipod still has only 4DOF per leg too!
I know this sounds confusing, but it should be much more clear once I post new pictures of Bipod v3.0 with the new legs and feet.
This is Bipod 3.0, with several changes. It has a new version of the wide feet, modified so the feet donāt trip over each other when in regular biped walking mode. These wide feet will also hopefully allow the Bipod to lean over far enough and stay balanced while walking in the Hexapod mode. I wonāt really know how this all works until I build it and start working with it, because this is a brand new design as far as I know.
I think this is the final version of the Bipod I am actually going to build. I canāt think of any other design changes I can make that would make it more possible for the Bipod to walk in the Hex style mode and Biped modes.
Now itās time to redesign the sensor turret for the Bipod and Mega Scout.
I think I finally have the range of motion of all the joints I have been looking for. This one should be able to walk in both Bipod (Hexapod style) and Biped (Humanoid style) stances.
It is just a pair of 3DOF SES hexapod type legs attached to a U channel body.
I know and understand how the ankle servos of a biped work.