Sliding Mode and Neural Networks Integration

EasyChair Preprint 13238

Abstract

This study delves into the exploration of adaptive control methodologies customized for manipulator visual servoing, a crucial element in robotic systems where precise manipulation of the manipulator's motion is essential, particularly in tasks reliant on visual feedback. Visual servoing poses distinct challenges due to uncertainties stemming from environmental factors, camera calibration intricacies, and fluctuating lighting conditions. Conventional control approaches often encounter difficulties in sustaining performance amid such uncertainties, prompting the necessity for the innovation of adaptive control techniques. This paper introduces original adaptive control methodologies that empower manipulators to dynamically adjust their control parameters based on real-time visual feedback, thereby enhancing resilience and performance in dynamic settings. The efficacy of these proposed techniques is substantiated through simulations and experimental findings, underscoring their potential to augment the effectiveness and dependability of manipulator visual servoing systems.

Keyphrases: Rigid-Deformable Objects, adaptive control, manipulation, manipulator, visual servoing

@booklet{EasyChair:13238,
  author    = {Wahaj Ahmed and Rahul Agarwal},
  title     = {Sliding Mode and Neural Networks Integration},
  howpublished = {EasyChair Preprint 13238},
  year      = {EasyChair, 2024}}