Deep Reinforcement Learning for Robotic Manipulation of Deformable Objects

EasyChair Preprint no. 13191

Abstract

Robotic manipulation of deformable objects presents a challenging problem due to their complex and unpredictable nature. Traditional control methods often struggle to handle the inherent uncertainties and non-linearities associated with deformable objects. In recent years, Deep Reinforcement Learning (DRL) has emerged as a promising approach to address these challenges by enabling robots to learn manipulation tasks through trial and error. This paper provides a comprehensive review of recent advancements in applying DRL techniques to tackle the problem of robotic manipulation of deformable objects. We discuss the key challenges, existing methodologies, and future directions in this exciting research domain.

Keyphrases: Artificial Intelligence, control, Deep Reinforcement Learning, deformable objects, Robotic Manipulation

@Booklet{EasyChair:13191,
  author = {Aradhana Das and Wahaj Ahmed},
  title = {Deep Reinforcement Learning for Robotic Manipulation of Deformable Objects},
  howpublished = {EasyChair Preprint no. 13191},

  year = {EasyChair, 2024}}