Dear everyone:
How does the damping factor work in the DLS method? Is there any relevant document?
About the damping of the DLS method
About the damping of the DLS method
Last edited by zhangm365 on 06 Jun 2022, 08:15, edited 1 time in total.
Re: About the damping of the DLS method
What I want to know is how the damping factor affects the control performance of DLS algorithm. For example, what is the principle of adjusting this value?
Re: About the damping of the DLS method
Sorry, I didn't realize that you were talking about the Ik module...
If the Jacobian pseudo inverse is expressed as \(J^{T}\cdot (J\cdot J^{T})^{-1}\), then the damped version is: \(J^{T}\cdot (J\cdot J^{T}+I\cdot f^{2})^{-1}\), with f the damping factor
Cheers
If the Jacobian pseudo inverse is expressed as \(J^{T}\cdot (J\cdot J^{T})^{-1}\), then the damped version is: \(J^{T}\cdot (J\cdot J^{T}+I\cdot f^{2})^{-1}\), with f the damping factor
Cheers
Re: About the damping of the DLS method
Thank you.
Is it true that the greater the damping factor within a certain range, the better the control accuracy of DLS algorithm?
Is it true that the greater the damping factor within a certain range, the better the control accuracy of DLS algorithm?
Re: About the damping of the DLS method
The only thing you can say I believe is that the greater the damping, the slower but more stable the convergence
Cheers
Cheers