Actually, you would not need to add 2DOF to the end of the foot for the ankle to gain 2DOF of movement. I’d add one full DOF for forward tilt of the ankle and use a ball bearing hub for the side to side tilt letting it adjust to the terrain on its own. This would give a 4DOF leg with 5DOF of movement.
I’ve already designed a very similar thing when I was trying to figure out how to put a wheel at the end of a leg and have it work properly. That would be a failure as it would be too clunky. However, I think a very similar scheme would work for a 2DOF foot added to a leg.
I don’t have much time to type right now but I see what you’re saying. You have a lot great ideas.
As for the servos fighting each other, the thought actually crossed my mind the other day. When a Hex is crouching down, if the feet are directly under the knee and the tibia is vertical it will have a narrow width between the feet. When it stands up to a mid height, the fibula would level out and knee joint would move outward from the chassis. If the tibia was to stay vertical during this, the feet would have to slide outward to stay under the knee. If the tibia rotated to keep the feet where they are, they would be on an outward angle. So the default “foot position” would have to be under the knee with the tibia and fibula vertical and horizontal.
I’m assuming there must be some compensating movement in the Powerpod program to keep the feet in a relatively similar position under the knee while crouching and raising the body.
Vertical movement is not only the issue area. Assuming the hexapod walks by three raised legs swinging foward, then being lowered to lift the hexapod weight, then being rotated to the rear to move the hexapod foward, then binding/slippage might occurr. During this movement the lateral distance between the legs has a significant change. There are probably several environments where this issue could be demonstrated with a hexapod, such as walking on a smooth sand surface or walking on that rubber type of mat used under carpets to keep them from slipping. The same type of issue happens with tracked gizmos when they encounter rotational movement such as turning. If they try to turn on surfaces that don’t allow slippage, they then to bog down. A lot of neat drawings have been posted, but there maybe issues when they are actually implemented.
I was under the impression that the code the powerpod produced almost entirely eliminated that particular problem. I haven’t taken the time to look at any code but some of the discussions involved Inverse kinematics which lead me to believe this.
Ahh, I can see where this could be a potential issue now, depending on the surface the bot is walking on.
I think I have a potential method to combat this too, which involves using a ball bearing hub where the last leg segment attaches to the end of the leg. This would allow that segment to rotate freely as the bot turns, without changing the bot’s actual position on the surface. I am going to have to see if this works when I start building legs for Walk 'N Roll.
Ah right you are! This is why Laurents program has been called a “programming work of art”! It uses IK to calculate the leg position and trajectory in real time to make sure the legs work together to walk as smoothly as possible. The only real friction is where the tip of the foot touches the ground, and is rotational. You guys are worrying about nothing.
LOL that’s hilarious. It is almost too big to turn around in the hall.
The chassis is 27 1/2" long and 8 inches wide.
I’m painting the Hinge Mounts for the servos right now then I’ll put it together, mount the electronics, route the wiring and try her out.
I can make up duplicate mounts, brackets and complete kits if anyone is interested. I need some time to get things together before I can make a bunch more though. I’ll post some details once I’m ready to make more.
I’m not sure on payload yet. The chassis and arms weight about 14 pounds the way it sits. I’ll be cutting lightening holes in the frame and brackets to shed some ounces as well.
If the track system is too heavy for it, I’m hoping it can carry my 209 Scout around. Then it could crouch down, let him off and the scout could walk away. LOL If only I could program for that…
Sure will, I’m just working on the wiring right now. I have to plug in a Y adapter, solder a longer lead and plug to the SSC-32, solder 2 longer power wires to the Power Distribution Board, run those wires to the battery packs and braid all the extension wires…18 times…“sigh”
Ok the wiring took longer than I thought. And I’m not even done yet. Here is the mess of wires so far. They are long enough to get to the electronics. Now I have to organize them, make my Power Board, plug everything in, and setup the batteries and power wiring.
How much payload will it take? Good metal working.
