Dong, Z;
Ding, Y;
Liu, W;
Hu, Z;
Lu, S;
Liu, Y;
(2025)
Maneuverability parameter identification of a water-jet USV based on truncated weighted LSSVM integrated with adaptive mutation PSO algorithm.
Ocean Engineering
, 321
, Article 120474. 10.1016/j.oceaneng.2025.120474.
![[thumbnail of Maneuverability parameter identification of a water-jet USV based on truncated weighted LSSVM integrated with adaptive mutation PSO algorithm.pdf]](https://discovery.ucl.ac.uk/style/images/fileicons/text.png) |
Text
Maneuverability parameter identification of a water-jet USV based on truncated weighted LSSVM integrated with adaptive mutation PSO algorithm.pdf - Accepted Version Access restricted to UCL open access staff until 26 January 2026. Download (6MB) |
Abstract
The maneuverability parameter identification of a water-jet unmanned surface vehicle (USV) is investigated in this paper based on experimental data. A multivariable function is employed to calculate the propulsive force generated by the dual water-jet propulsion system. Subsequently, a three-degree-of-freedom (3-DOF) mathematical model is established to capture the USV's dynamic characteristics. A novel approach integrating a truncated weighted least-squares support vector machine (TWLSSVM) with an adaptive mutation particle swarm optimization (AMPSO) algorithm is proposed for accurate parameter estimation. Truncated singular value decomposition (SVD) is incorporated into the LSSVM framework to enhance numerical stability and reduce noise interference. Unlike traditional LSSVM, this approach assigns weighting factors to data points based on Lagrange multipliers, reflecting their contributions. The AMPSO algorithm, enhanced with adaptive and mutation strategies, optimizes key TWLSSVM parameters, improving global search efficiency and avoiding local optima. Experimental steering and turning test data are used for parameter identification and subsequent motion prediction. Compared to traditional LSSVM, the root mean square error (RMSE) and combined accuracy (CA) of the predicted heading angle are reduced by more than 38%, while trajectory prediction metrics decrease by over 15%, reflecting the superiority of the proposed AMPSO-LSSVM algorithm in USV's parameter identification and motion prediction.
| Type: | Article |
|---|---|
| Title: | Maneuverability parameter identification of a water-jet USV based on truncated weighted LSSVM integrated with adaptive mutation PSO algorithm |
| DOI: | 10.1016/j.oceaneng.2025.120474 |
| Publisher version: | https://doi.org/10.1016/j.oceaneng.2025.120474 |
| Language: | English |
| Additional information: | This version is the author accepted manuscript. For information on re-use, please refer to the publisher’s terms and conditions. |
| Keywords: | Water-jet unmanned surface vehicle; Parameter identification; Motion prediction; Truncated weighted least square support vector machine; Adaptive mutation particle swarm optimization; Full-scale trial data |
| UCL classification: | UCL UCL > Provost and Vice Provost Offices > UCL BEAMS UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Mechanical Engineering |
| URI: | https://discovery.ucl.ac.uk/id/eprint/10204793 |
Archive Staff Only
 |
View Item |
