would this work:
using the gait section of the code and a mix of BodyPosY +/- help to adjust the gait relitive to the tilted coxa.
let me explain:
imagine adding a code or part that would have a normal phoenix to walk along changing its body height everytime a leg lifts. when no leg is lifted the body is at normal body height, and when the flighted leg begins to lift the body starts to lower, as the leg moves to the center point of rotation the body is at its lowest point, and then as the leg begins to return back to the ground the body rises up again. got that…?.ok this is how it works for the Stalker:
as the tilted coxa on each leg moves though the gait in a downward arc it will raise the body.
can we use body axis Y as explained above to compensate for this.
;Check IF the Gait is in motion
GaitInMotion = ((ABS(TravelLengthX)>ps2Tolerance) | (ABS(TravelLengthZ)>ps2Tolerance) | (ABS(TravelRotationY)>ps2Tolerance) )
;Leg middle up position
;Gait in motion Gait NOT in motion, return to home position
IF (GaitInMotion & (NrLiftedPos=1 | NrLiftedPos=3) & GaitStep=GaitLegNr) | (GaitInMotion=NEGITIVE & GaitStep=GaitLegNr & ((ABS(GaitPosX)>2) | (ABS(GaitPosZ)>2) | (ABS(GaitRotY)>2))) THEN ;Up
GaitPosX = 0
GaitPosY = -LegLiftHeight
GaitPosZ = 0
GaitRotY = 0
BodyPosY = 15
ELSE
;Advance to the next step
IF LastLeg THEN ;The last leg in this step
GaitStep = GaitStep+1
IF GaitStep>StepsInGait THEN
GaitStep = 1
BodyPosY = 0 ;Normal hieght
ENDIF
ENDIF[/code]
these numbers are only an example as it would have to be tried and tested and i would need to find the best way to calculate the arc values!
I don’t think that will work. The computation of x,y,z is continuous, it’s more than just raising and lowering the body. You will also notice that the tip of leg will drift outward and inward.
Unless I’m missing something between Xan’s code, I still think my transformation is correct. 8) Let me find time to add it onto my bot to verify it.
Yes you are correct there. The work you produced is spot on. Just need to find the right way to implement that into the code. Would like like me to post you the current code? I need to do a bit more work on it tonight to tidy up the lose ends. Thanks for you help. J
i would like to create another init pos for the legs so i can allow the body to move into a better posistion before initializing walk to get a better COG.
Would this be the correct way to do it. i wish to use a button to select it into this mode, like this:
[code]StartPosSwitch var bit ;Start2 Positions
RFPosX var sword ;Actual Position of the Right Front Leg
RFPosY var sword
RFPosZ var sword
RRPosX var sword ;Actual Position of the Right Rear Leg
RRPosY var sword
RRPosZ var sword
LFPosX var sword ;Actual Position of the Left Front Leg
LFPosY var sword
LFPosZ var sword
LRPosX var sword ;Actual Position of the Left Rear Leg
LRPosY var sword
LRPosZ var sword
IF (DualShock(2).bit3 = 0) and LastButton(1).bit3 THEN ;R1 Button test
IF StartPosSwitch = 0 THEN
StartPosSwitch = 1
sound P9,[100\4000]
ELSE
StartPosSwitch = 0
sound P9,[250\3000]
ENDIF
ENDIF
;-----main
IF (StartPosSwitch>0) THEN
gosub StartPosMode
ENDIF
StartPosMode
RFPosX = # ;Start2 positions of the Right Front leg
RFPosY = #
RFPosZ = -#
RRPosX = # ;Start2 positions of the Right Rear leg
RRPosY = #
RRPosZ = #
LFPosX = # ;Start2 positions of the Left Front leg
LFPosY = #
LFPosZ = -#
LRPosX = # ;Start2 positions of the Left Rear leg
LRPosY = #
LRPosZ = #
return[/code]
obviously the sub would need to step into this position and not just snap into position. how would this look.
There are a few different ways to get the legs to the position like in your drawing. But I doubt that changing the init position isn’t really what you want. Looking at the picture I think you want to create a start position for walking. It’s a good thing when you start working on the gait but at the end you want to start (and change) the gait dynamic. This way you don’t need to step in position first and then start to walk. This even makes it possible to change direction while walking.
Don’t underestimate it; this was one of the hardest things to do. I suggest you use the XXGaitPos variables to get the legs in the correct position so you get the gait right. Check one of my first vids where I did some walking tests. At 0:30 You will see that I place the legs in the init position first, then lower all legs and start walking.
I also saw Zenta’s reaction in the code topic. This defiantly is a great way to change the init positions without the legs snapping in to position. But if you only gonna use it for tests to get the gait started I should lift the bot when the legs snap. And just focus on the gait for now and get a dynamic starting position in the future.
I think the latest design you have there looks more rigid. That one design with the bent 45° bracket looked very “springy”.
The evoloution of robotics 8)
To make the new design look similar to the older design, you could flip the hip servos 180° so that the servo horn is on top instead of the bottom. Or leave the servos the way they are and just flip the “C” brackets, then reattach the legs. The body would then be lower like the first design. Unless having the body lower causes problems with the IK, you could look about trying that configuration.
On the PC boards. You may be able to reduce the height of the stack by turning the SSC-32 over. Then you can stack the Bot Board on it with a 1/4" spacer. There should be ample room for the wires with the “H” design of the carrier. You may need to flip it so the DB9 is facing the away from the Sharp sensor.
This design are going to make the leg stronger, espescially when the coxa servo are far from its center position. To be honest I liked the previous design a little better, but your new design looked great too and you did a great job making Stalker look similar to what you want. I also liked the detail Jim pointed out with the brackets! A smart and easy mod.
I’m sorry for not beeing able to help you out with the IK. Good thing the IK isn’t any trouble anymore though.
very good idea i will try this to see how it looks. thanks for the idea.
i wouldnt have thought it would be a problem for the ik and if anything having low body would help with the COG.
yeah i actually went to do this but am going to order your carrier to do
this with. the one i made didnt really give enough room for the wires. so you’ve sold another one there!
i have plenty of SES. collected over the years. iv got enough to build a hexapod completly out of SES plus a few other parts but not as much as jim!
thanks, i wouldnt worry bout that, im happy with it anyway and my be sometime in the futre we can work on a robot with tilted servos for experiments or so.
thanks for your help on this anyway.
**
thanks guys.**
i have flipped the SSC b2b with the ABB reducing the stack. but i have also used the old Chassis plate again on the top.
here includes:
*PS2 receiver,
ABB 9v clip,
SSC 9v clip,
servo power switch,
head directional servo.
*
