Analysis of Robot Skill Learning Methods and Secondary Approaches to Enhance Learning

Abstract

Robot skill learning is a long-existing problem that has gained attention due to the success of deep learning applications. When child learning is analyzed, they use a combination of methods for learning manipulation tasks. This situation inspired us to analyze different learning approaches. We present a comprehensive analysis of robot skill learning. The paper's primary focus is on exploring the effect of robot skill learning based on various learning methods, such as representation learning, predictive learning, and multi-modal learning. This paper also analyzes the advantages and disadvantages of these algorithms.