A machine learning approach for dead-reckoning navigation at sea using a single accelerometer

Roee Diamant, Yunye Jin

Research output: Contribution to journalArticlepeer-review


Dead-reckoning (DR) navigation is used when Global Positioning System (GPS) reception is not available or its accuracy is not sufficient. At sea, DR requires the use of inertial sensors, usually a gyrocompass and an accelerometer, to estimate the orientation and distance traveled by the tracked object with respect to a reference coordinate system. In this paper, we consider the problem of DR navigation for vessels located close to or on the sea surface, where motion is caused by ocean waves. In such cases, the vessel pitch angle is fast time varying and its estimation by direct measurements of orientation is prone to drifts and noises of the gyroscope. Regarding this problem, we propose a method to compensate for the vessel pitch angle using a single acceleration sensor. Using a constraint expectation-maximization (EM) algorithm, our method classifies acceleration measurements into states of similar pitch angles. Subsequently, for each class, we project acceleration measurements into the reference coordinate system along the vessel heading direction, and obtain distance estimations by integrating the projected measurements. Results in both simulated and actual sea environments demonstrate that, by using only acceleration measurements, our method achieves accurate results.

Original languageEnglish
Article number6646319
Pages (from-to)672-684
Number of pages13
JournalIEEE Journal of Oceanic Engineering
Issue number4
StatePublished - 1 Oct 2014
Externally publishedYes

Bibliographical note

Publisher Copyright:
© 2014 IEEE.


  • Dead reckoning (DR)
  • expectation-maximization (EM) classification
  • naval navigation

ASJC Scopus subject areas

  • Ocean Engineering
  • Mechanical Engineering
  • Electrical and Electronic Engineering


Dive into the research topics of 'A machine learning approach for dead-reckoning navigation at sea using a single accelerometer'. Together they form a unique fingerprint.

Cite this