Discrete learning control for robots: strategy, convergence and robustness (Q4036143)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 168399
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | Discrete learning control for robots: strategy, convergence and robustness |
scientific article; zbMATH DE number 168399 |
Statements
Discrete learning control for robots: strategy, convergence and robustness (English)
0 references
16 May 1993
0 references
discrete time iterative learning control
0 references
0.8490656018257141
0 references
0.8207467198371887
0 references
0.8178048133850098
0 references
0.7984034419059753
0 references
0.7886316180229187
0 references
The paper presents a discrete time iterative learning control for the accurate tracking of repetitive trajectories in rigid robots, including a convergence proof and experimental results. The basic update for the torque to be applied at the next iteration of the learning strategy involves the previously applied torque and two further terms. As the sampling time goes to zero, these two terms approximate respectively a velocity error feedback and the actual acceleration of the robot at the previous iteration. The convergence proof uses standard contraction mapping arguments, together with an appropriate functional norm definition. It should be pointed out that here, as in many other papers on the same subject, a somewhat erroneous assumption is used: the Coriolis and centrifugal terms in the robot dynamic model are assumed to be Lipschitz, though their quadratic dependence on joint velocities. A local Lipschitz property holds instead if one proves additionally that the joint velocities remain upper bounded. With simple modifications the thesis of the main (sufficient) result still holds. This is confirmed by the reported experimental results, where saturation of the actuators induces a practical velocity bound.
0 references