State estimation for autonomous surface vehicles based on Echo state networks

Zhouhua Peng, Jun Wang, Dan Wang

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

Abstract

This paper investigates the state estimation for autonomous surface vehicles in the presence of unknown dynamics and unmeasured states. The unknown dynamics comes from parametric model uncertainty, unmodelled hydrodynamics, and external disturbances caused by wind, waves and ocean currents. A nonlinear adaptive observer is proposed based on echo state networks, which are used to approximate the unknown dynamics using input-output data. By using the proposed observer, the unmeasured states and unknown dynamics can be simultaneously estimated in real time. The stability of the observer is analyzed via Lyapunov analysis. The proposed observer can be used in various motion control scenario, such as target tracking, trajectory tracking, path following, formation control, and even sideslip angle identification, not only for fully-actuated marine vehicles but also for under-actuated marine vehicles.

Original languageEnglish
Title of host publicationAdvances in Neural Networks - ISNN 2017 - 14th International Symposium, ISNN 2017, Proceedings
EditorsAndrew Leung, Fengyu Cong, Qinglai Wei
PublisherSpringer Verlag
Pages127-134
Number of pages8
ISBN (Print)9783319590714
DOIs
Publication statusPublished - 2017
Externally publishedYes
Event14th International Symposium on Neural Networks, ISNN 2017 - Sapporo, Hakodate, and Muroran, Hokkaido, Japan
Duration: 21 Jun 201726 Jun 2017

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10261 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference14th International Symposium on Neural Networks, ISNN 2017
Country/TerritoryJapan
CitySapporo, Hakodate, and Muroran, Hokkaido
Period21/06/1726/06/17

Keywords

  • Echo state network
  • Fully-actuated marine vehicles
  • State estimation
  • Under-actuated marine vehicles
  • Unknown dynamics

Fingerprint

Dive into the research topics of 'State estimation for autonomous surface vehicles based on Echo state networks'. Together they form a unique fingerprint.

Cite this