Filtering Method for Location Estimation of an Underwater Robot

Nak Yong Ko, Tae Gyun Kim


This paper describes an application of extended Kalman filter(EKF) for localization of an underwater robot. For the application, linearized model of robot motion and sensor measurement are derived. Like usual EKF, the method is recursion of two main steps: the time update(or prediction) and measurement update. The measurement update uses exteroceptive sensors such as four acoustic beacons and a pressure sensor. The four beacons provide four range data from these beacons to the robot and pressure sensor does the depth data of the robot. One of the major contributions of the paper is suggestion of two measurement update approaches. The first approach corrects the predicted states using the measurement data individually. The second one corrects the predicted state using the measurement data collectively. The simulation analysis shows that EKF outperforms least squares or odometry based dead-reckoning in the precision and robustness of the estimation. Also, EKF with collective measurement update brings out better accuracy than the EKF with individual measurement update.

Full Text:




  • There are currently no refbacks.

Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

IAES International Journal of Robotics and Automation (IJRA)
ISSN 2089-4856, e-ISSN 2722-2586
This journal is published by the Institute of Advanced Engineering and Science (IAES) in collaboration with Intelektual Pustaka Media Utama (IPMU).

Web Analytics Made Easy - Statcounter IJRA Visitor Statistics