Hi Jeroen, Could you describe what you're trying to model in more detail? In particular, where are the encoders, motors, springs, gears, and joints in relation to each other? -Stu On Tue, Dec 14, 2010 at 4:37 AM, Jeroen Willems wrote: > Okay I'm looking into the flexibility between the actuator and the joint by > adding a custom transmission but I'm having some difficulties. > > I've looked into the existing transmissions and those are pretty > straightforward (except for the pr2_gripper). > I want to add the (linear, rotational) flexibility, I will neglect the > damping for simplicity. > I'm used to use the equations of motion: > Load:  Jl theta_L_ddot + k ( theta_L - theta_M ) = 0 > Motor: Jm theta_M_ddot - k ( theta_L - theta_M ) = u > > I don't get it how to implement this in the transmission. I was thinking of > a little bit of rewriting as in the attachment but it did not influence > anything at all. > Would be great if someone has an idea, solution or suggestion on how > implementing this flexibility. > > Thanks in advance > > Jeroen > > On Sat, Nov 20, 2010 at 21:07, Stuart Glaser > wrote: >> >> Hi Joeren, >> >> For both of these, the solution is to write a transmission.  Both >> issues require a specific translation between actuators and joints, >> and the best way to encode that translation is in a transmission. >> >> -Stu >> >> On Fri, Nov 19, 2010 at 1:36 AM, Jeroen Willems >> wrote: >> > Hi All, >> > >> > I'm currently modeling a hexapod in Gazebo but I'm currently having some >> > difficulties. >> > >> > I have basically 2 questions, I hope that this is not a problem for one >> > post >> > but both have to do with the transmission: >> > >> > 1) Between the actuators and the joints are torsion springs. I haven't >> > found >> > something that incorporates flexibility between actuators and joints. >> > Is there some solution or workaround to obtain this or should I try an >> > alternative solution like writing a custom transmission? >> > >> > 2) As you can imagine I have 6 legs consisting of three links, basically >> > an >> > elbow manipulator. >> > The last two links can be regarded as in a 2D-plane. The third link is >> > remotely driven. >> > This means that when rotating the second link the third link does not >> > change >> > the rotation with respect to the body. >> > So in a 'standard' situation the third link rotates with the second link >> > so >> > the angle between those two does not change. >> > I hope this is a bit clear. >> > So again is there some solution or workaround? >> > >> > Thanks! >> > >> > Jeroen >> > >> > _______________________________________________ >> > ros-users mailing list >> > ros-users@code.ros.org >> > https://code.ros.org/mailman/listinfo/ros-users >> > >> > >> >> >> >> -- >> Stuart Glaser >> sglaser -at- willowgarage -dot- com >> www.willowgarage.com >> _______________________________________________ >> ros-users mailing list >> ros-users@code.ros.org >> https://code.ros.org/mailman/listinfo/ros-users > > > _______________________________________________ > ros-users mailing list > ros-users@code.ros.org > https://code.ros.org/mailman/listinfo/ros-users > > -- Stuart Glaser sglaser -at- willowgarage -dot- com www.willowgarage.com