An ensemble of online estimation methods for one degree-of-freedom models of unmanned surface vehicles: applied theory and preliminary field results with eight vehicles

TL;DR

This paper evaluates three online system identification methods for unmanned surface vehicles, demonstrating their performance through experiments with eight USVs and highlighting the advantages of ensemble approaches.

Contribution

It introduces an ensemble of online estimation methods, including a stable RNN, AID, and RLS, with novel stability considerations and equilibrium point analysis for USV modeling.

Findings

AID achieved the lowest mean absolute error online

Ensemble methods outperformed individual models offline

Stable RNN with contraction stability constraints was implemented

Abstract

In this paper we report an experimental evaluation of three popular methods for online system identification of unmanned surface vehicles (USVs) which were implemented as an ensemble: certifiably stable shallow recurrent neural network (RNN), adaptive identification (AID), and recursive least squares (RLS). The algorithms were deployed on eight USVs for a total of 30 hours of online estimation. During online training the loss function for the RNN was augmented to include a cost for violating a sufficient condition for the RNN to be stable in the sense of contraction stability. Additionally we described an efficient method to calculate the equilibrium points of the RNN and classify the associated stability properties about these points. We found the AID method had lowest mean absolute error in the online prediction setting, but a weighted ensemble had lower error in offline processing.