I have two sets of forces and moments, measured by two 6 DOF force/torque sensors at position A and B on a robot arm (B at end effector). My goal is to use this measurements to relate the forces/moments at position A to those at position B, i.e. find a transformation of $F_A, M_A$ to $F_B, M_B$. The transformation is static and no other external forces apply.
The related formulas for that are: $$ F_B \approx RF_A \\ M_B \approx RM_A+t\times F_B $$ Is that correct (it's addaped from this)?
My goal is now to find an optimal (in a least squares way) matrix R and vector t. How would I do this? Is the Kabsch algorithm suitable for this?

Below a small sketch of the setup. It is some sort of handle on the robot arm. A force is applied (and measured) at Sensor A, the goal is to use the measured force at sensor B to find the transformation. Sketch of the setup

  • $\begingroup$ I updated the question, hope that helps. Unfortunately I don't have a photo or real technical drawing at the moment due to the current crisis. $\endgroup$ – Sebastian E Apr 15 '20 at 9:26
  • $\begingroup$ If I understand well your problem, the sensor B is at a "free edge" of the robot's arm. If it is the case, no forces or moments will ever be registered by your sensor, since by definition a free edge is stress free. $\endgroup$ – ClariB Apr 15 '20 at 10:15
  • $\begingroup$ It is at the end effector and measures all forces applied to the robot. Maybe my drawing is not perfect :D $\endgroup$ – Sebastian E Apr 15 '20 at 10:39
  • $\begingroup$ Maybe that is not s clear, but of cause this is just a detail. At the right would be a joint and other links of the robot. $\endgroup$ – Sebastian E Apr 15 '20 at 12:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.