Table 17 Deviation errors on circular and square paths in Q4.

From: Path-based evaluation of deep learning models for solving inverse kinematics in a revolute-prismatic robot

Model

Path shape

Deviation error (mm)

DFNN with K-fold CV (single o/p)

Square

0.715

DFNN with K-fold CV (dual o/p)

1.011

DFNN without CV

1.076

LSTM

1.055

GRU

0.772

DFNN with K-fold CV (single o/p)

Circle

0.662

DFNN with K-fold CV (dual o/p)

0.883

DFNN without CV

1.174

LSTM

1.103

GRU

0.689

Back to article page