Machine Learning for Robot Precision: Predicting End-of-Operation Errors Using Regression Techniques

View Full Paper

Ask AI

Bookmark

Discussion

This study compares four regression models for estimating end-effector error in robotic systems, indicating key performance metrics.

Linear regression yielded the highest accuracy, achieving an R² of 0.9838 in training, and 0.9594 in testing.
The multilayer perceptron model captured nonlinear patterns, showing training and testing R² values of 0.98075 and 0.94832, respectively.
Decision tree regression demonstrated overfitting, with a training R² of 1.0000 and a test R² of 0.9240.
Support vector regression showed consistent results, achieving R² values of 0.9606 during training and 0.9501 during testing.

A 2025 study studied this question.

View Full Paper

Ask AI

Bookmark