Visual Servoing Gain Tuning by Sarsa: An Application with a Manipulator

Jie Liu, Yang Zhou, Jian Gao, Weisheng Yan

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Scopus citations

Abstract

This paper investigates a Sarsa-based visual servoing control gain tuning method and the application on a manipulator. For a typical visual servo controller, fixed control gains will not provide the best performance. Therefore, state action reward state action (SARSA) algorithm, one of learning-based methods from reinforcement learning (RL), is introduced to select control gains in every control step. The norm of the visual error is used to define the state space. The positive gain of the controller is discretized as the actions. A reward function is defined to evaluate the performance of every action. Both a numerical test and a robot experiment are carried out to validate the presented algorithm.

Original languageEnglish
Title of host publicationProceedings - 2023 3rd International Conference on Robotics and Control Engineering, RobCE 2023
EditorsAiguo Song, Maki Habib
PublisherAssociation for Computing Machinery
Pages103-107
Number of pages5
ISBN (Electronic)9781450398107
DOIs
StatePublished - 12 May 2023
Event3rd International Conference on Robotics and Control Engineering, RobCE 2023 - Nanjing, China
Duration: 12 May 202314 May 2023

Publication series

NameACM International Conference Proceeding Series

Conference

Conference3rd International Conference on Robotics and Control Engineering, RobCE 2023
Country/TerritoryChina
CityNanjing
Period12/05/2314/05/23

Keywords

  • manipulator
  • reinforcement learning
  • SARSA
  • visual servoing

Fingerprint

Dive into the research topics of 'Visual Servoing Gain Tuning by Sarsa: An Application with a Manipulator'. Together they form a unique fingerprint.

Cite this