Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Reinforcement Learning-Based Bionic Reflex Control for Anthropomorphic Robotic Grasping exploiting Domain Randomization

Hirakjyoti Basumatary, Daksh Adhar, Atharva Shrawge, Prathamesh Kanbaskar, Shyamanta M. Hazarika

TL;DR

This study introduces an innovative bionic reflex control pipeline, leveraging reinforcement learning (RL); thereby eliminating the need for human intervention during control design, and underscores the promise of RL as a potent tool for advancing bionic Reflex control within anthropomorphic robotic hands.

Abstract

Achieving human-level dexterity in robotic grasping remains a challenging endeavor. Robotic hands frequently encounter slippage and deformation during object manipulation, issues rarely encountered by humans due to their sensory receptors, experiential learning, and motor memory. The emulation of the human grasping reflex within robotic hands is referred to as the ``bionic reflex". Past endeavors in the realm of bionic reflex control predominantly relied on model-based and supervised learning approaches, necessitating human intervention during thresholding and labeling tasks. In this study, we introduce an innovative bionic reflex control pipeline, leveraging reinforcement learning (RL); thereby eliminating the need for human intervention during control design. Our proposed bionic reflex controller has been designed and tested on an anthropomorphic hand, manipulating deformable objects in the PyBullet physics simulator, incorporating domain randomization (DR) for enhanced Sim2Real transferability. Our findings underscore the promise of RL as a potent tool for advancing bionic reflex control within anthropomorphic robotic hands. We anticipate that this autonomous, RL-based bionic reflex controller will catalyze the development of dependable and highly efficient robotic and prosthetic hands, revolutionizing human-robot interaction and assistive technologies.

Reinforcement Learning-Based Bionic Reflex Control for Anthropomorphic Robotic Grasping exploiting Domain Randomization

TL;DR

This study introduces an innovative bionic reflex control pipeline, leveraging reinforcement learning (RL); thereby eliminating the need for human intervention during control design, and underscores the promise of RL as a potent tool for advancing bionic Reflex control within anthropomorphic robotic hands.

Abstract

Achieving human-level dexterity in robotic grasping remains a challenging endeavor. Robotic hands frequently encounter slippage and deformation during object manipulation, issues rarely encountered by humans due to their sensory receptors, experiential learning, and motor memory. The emulation of the human grasping reflex within robotic hands is referred to as the ``bionic reflex". Past endeavors in the realm of bionic reflex control predominantly relied on model-based and supervised learning approaches, necessitating human intervention during thresholding and labeling tasks. In this study, we introduce an innovative bionic reflex control pipeline, leveraging reinforcement learning (RL); thereby eliminating the need for human intervention during control design. Our proposed bionic reflex controller has been designed and tested on an anthropomorphic hand, manipulating deformable objects in the PyBullet physics simulator, incorporating domain randomization (DR) for enhanced Sim2Real transferability. Our findings underscore the promise of RL as a potent tool for advancing bionic reflex control within anthropomorphic robotic hands. We anticipate that this autonomous, RL-based bionic reflex controller will catalyze the development of dependable and highly efficient robotic and prosthetic hands, revolutionizing human-robot interaction and assistive technologies.
Paper Structure (17 sections, 8 equations, 8 figures, 1 algorithm)

This paper contains 17 sections, 8 equations, 8 figures, 1 algorithm.

Table of Contents

  1. INTRODUCTION
  2. Related Work
  3. Fundamentals
  4. Reinforcement Learning
  5. Actor critic model
  6. Soft Actor-Critic
  7. Discrete Wavelet Transform - Haar Wavelet
  8. Design Methodology
  9. PyBullet and OpenAI Gym
  10. Slip Detection
  11. Deformation Calculation
  12. Rewards
  13. Results and Discussion
  14. Reward Plots for Nominal and DR Agents
  15. Success Rates for Nominal and DR Agent
  16. ...and 2 more sections

Figures (8)

  • Figure 1: RL based Bionic Reflex Control Pipeline
  • Figure 2: Force sensor signal while grasping and lifting an object. Fifth level Haar decomposition of raw force sensor signal is used to detect slip. The positive gradient is a reflection of the load being applied. The opposing variation trend is a representation of slip.
  • Figure 3: Triangular mesh description of the deformable cylinder. Point O represents reference origin.
  • Figure 4: Average Reward Plots of Nominal and DR Agent
  • Figure 5: Learned Grasp simulation in PyBullet: (a) Initial Grasp Pose, (b) Grasping the object (c) Object Lift without Slippage
  • ...and 3 more figures