Skip to main navigation Skip to search Skip to main content

Discounted Sampling Policy Gradient for Robot Multi-objective Visual Control

Research output: Chapters, Conference Papers, Creative and Literary WorksRGC 32 - Refereed conference paper (with host publication)peer-review

Abstract

Robot visual control often involves multiple objectives such as achieving high efficiency, maintaining stability, and avoiding failure. This paper proposes a novel Vision-Based Control method (VBC) with the Discounted Sampling Policy Gradient (DSPG) and Cosine Annealing (CA) to achieve excellent multi-objective control performance. In our proposed visual control framework, a DSPG learning agent is employed to learn a policy estimating continuous kinematics for VBC. The deep policy maps the visual observation to a specific action in an end-to-end manner. The DSPG agent finally can update the policy to obtain the optimal or near-optimal solution using shaped rewards from the environment. The proposed VBC-DSPG model is optimized using a heuristic method. Experimental results demonstrate that the proposed method performs very well compared with some classical competitors in the multi-objective visual control scenario.
Original languageEnglish
Title of host publicationEvolutionary Multi-Criterion Optimization
Subtitle of host publication11th International Conference, EMO 2021, Shenzhen, China, March 28–31, 2021, Proceedings
EditorsHisao Ishibuchi, Qingfu Zhang, Ran Cheng, Ke Li, Hui Li, Handing Wang, Aimin Zhou
Place of PublicationCham
PublisherSpringer 
Pages441-452
Number of pages12
ISBN (Electronic)9783030720629
ISBN (Print)9783030720612
DOIs
Publication statusPublished - 28 Mar 2021
Event11th International Conference on Evolutionary Multi-Criterion Optimization (EMO 2021) - Hampton by Hilton Hotel (on-site & on-line), Shenzhen, China
Duration: 28 Mar 202131 Mar 2021

Publication series

NameLecture Notes in Computer Science
Volume12654
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference11th International Conference on Evolutionary Multi-Criterion Optimization (EMO 2021)
PlaceChina
CityShenzhen
Period28/03/2131/03/21

Research Keywords

  • Multi-objective visual control
  • Kinematics
  • Discounted sampling policy gradient
  • Cosine annealing

Fingerprint

Dive into the research topics of 'Discounted Sampling Policy Gradient for Robot Multi-objective Visual Control'. Together they form a unique fingerprint.

Cite this