MOSAIC-SV: Real-Time Adaptive Identification of Vessel Dynamics for the Control and Deployment of Aquatic Robots

Wensen Liu; Jerry Peng; Shravani Vedagiri; Aaron M. Johnson

cs.ROOct 2, 2026

MOSAIC-SV: Real-Time Adaptive Identification of Vessel Dynamics for the Control and Deployment of Aquatic Robots

Authors: Wensen Liu, Jerry Peng, Shravani Vedagiri, Aaron M. Johnson

Organizations: Mechanical Engineering Department, Carnegie Mellon University, Pittsburgh, PA 15213, USA

Abstract

Model-based control of an aquatic robotic platform depends on a hydrodynamic model that is costly to identify and specific to the hull, payload, and conditions it was measured in. Here, we present MOSAIC-SV, a deployable real-time adaptive dynamics identification and control system that identifies a control-sufficient dynamics model from a spec-sheet engineering prior, without dedicated identification trials, and re-estimates it at every control step of a closed-loop mission while the controller plans on it. A physically admissible unscented Kalman filter re-estimates hydrodynamic, disturbance, and actuator parameters at every control step of the closed-loop mission; while a command-dependent consider projection withholds corrections the current command cannot attribute between actuator effectiveness and external force; and a model predictive path integral controller plans on the current estimate. In simulation on a CyberShip II plant, MOSAIC-SV recovers the transit performance of the calibrated model under static mismatch and transient changes, and stays within 20% of its own transit time at the unscaled prior when its inertia or damping prior is wrong by an order of magnitude. In on-water field trials on the Blue Robotics BlueBoat, a twin-thruster catamaran, MOSAIC-SV transits at least 25% faster and predicts its own motion with at least 56% less error than its frozen engineering prior, including under an unmodeled payload. The same MOSAIC-SV system concept was also feasibly deployed on a 6.3-tonne dual outboard monohull.

Figures & tables

Fig. 1: A Blue Robotics BlueBoat, the twin-thruster catamaran evaluated in the on-water field trials with an unmodeled extra mass (a rock).

Panel of Figure 2 — Fig. 2: (a) The general NED and FRD vessel frames. Heading $\psi$ is defined as clockwise from North. (b) The general form of the twin-hulled catamaran vessel class evaluated in this work, with location of forces and corresponding lever arms.

Fig. 3: Simulation course in the world frame. The triangle marks the start, open circles waypoints passed under way, and the square the stop waypoint.

Family	Axis	Perturbation	Levels
Static	mass	Inertia $\times$ level	$2^{-3},\ldots,2^{4}$
Static	cross	Inertia $\times$ , damping $\div$	$2^{-3},\ldots,2^{4}$
Static	damping	Damping $\times$ level	$2^{-3},\ldots,2^{4}$
Static	combined	Inertia/damping $\times$ ; 45 ∘	$2^{-3},\ldots,2^{4}$ ; force 0–3.2 N ‡
Static	current	Force at 45/0/90 ∘	0–3.2 N, step 0.4 N
Transient	mass	Inertia $\times$ level	$1/4$ , $1/2$ , $1.5$ , $2$ , $3$ , $4$ , $6$ , $8$

TABLE I: Perturbation families

Fig. 4: Static mismatch. Top: median transit time per plant level.

n

gives the number of seeds that completed the run when there were incomplete runs. Bottom: median self-prediction error. Levels are the scale applied to the unperturbed plant.

		vs. calibrated			vs. frozen
Condition	$n$	$\Delta_{\mathrm{tr}}$ %	$\Delta c$ pp	$\Delta e$ %	$\Delta_{\mathrm{tr}}$ %	$\Delta c$ pp	$\Delta e$ %
mass	96	-25.2	-11.5	+17.3	-27.5	-12.5	+3.1
cross	96	-0.6	+4.2	-29.0	-2.7	+0.0	-8.9
damping	96	+2.1	+0.0	-41.3	-26.2	+0.0	-8.5
combined	96	-5.9	+0.0	+12.8	-26.0	+0.0	+11.7
hull, all	384	-1.5	-1.8	-7.1	-26.0	-3.1	-5.8

TABLE II: Static simulation comparison

		vs. calibrated			vs. frozen
Condition	$n$	$\Delta_{\mathrm{tr}}$ %	$\Delta c$ pp	$\Delta e$ %	$\Delta_{\mathrm{tr}}$ %	$\Delta c$ pp	$\Delta e$ %
Step mass	96	-11.8	+25.0	-24.2	+5.0	+12.5	-26.7
Step damping	72	+9.9	+0.0	-38.9	-31.1	+0.0	-18.2
Step current	60	+8.6	+0.0	-31.7	-30.1	-20.0	-16.1
Step port	60	+3.4	+0.0	-23.5	-29.4	+0.0	-37.8
Step stbd	60	+1.5	+0.0	-4.7	-35.5	+0.0	-22.9

TABLE III: Transient simulation comparison

Fig. 5: MOSAIC-SV sensitivity to the engineering prior. Evaluated using an unperturbed plant and scaling the prior inertia, damping, or both. The band marks

\pm 20\%

of the transit time at the unscaled prior.

Fig. 6: Field test run trajectories. Crosses mark waypoints with their 2 m arrival radius.

	vs. frozen
Cell	$n$	$\Delta_{\mathrm{tr}}$ %	$\Delta e$ %
Single nom.	3	-40.3	-59.9
O/B nom.	3	-40.0	-69.7
Single +9.1 kg	3	-42.9	-56.9
O/B +9.1 kg	3	-25.3	-70.3

TABLE IV: Field trial comparison

Panel of Figure 11 — Fig. 7: On-water runs of the 6.3-tonne dual outboard monohull. (a) Three-waypoint run. (b) Single-obstacle run. Crosses mark waypoints with their arrival radius; the circle marks the obstacle with its avoidance clearance.

Robot Dynamics Identification

Robot dynamics identification estimates the masses, inertias, friction, actuator behavior, and other forces that govern robot motion.

See all 57 papers

Uncertainty-Aware Adaptive Dynamics For Underwater Vehicle-Manipulator Robots
Mar 6, 2026Edward Morgan, Nenyi K Dadson, Corina BarbalataRobot Dynamics Identification Underwater Robotics
The authors are with the Department of Mechanical & Industrial Engineering, at Louisiana State University, Baton Rouge, LA 70803, USA
Optimal Excitation Trajectories for System Identification of Underwater Vehicles
Sep 15, 2026Fotis Panetsos, Kostas J. KyriakopoulosRobot Dynamics Identification Trajectory Optimization
Center for AI & Robotics (CAIR) and the Electrical Eng. Program, Engineering Division, New York University Abu Dhabi, Abu Dhabi, United Arab Emirates.
Adaptive Prescribed Performance Trajectory Tracking for Uncertain Autonomous Underwater Vehicles With Input Saturation
Oct 14, 2026Athanasios K. Gkesoulis, Christos Vlachos, George K. Fourlas +2Robot Trajectory Tracking Adaptive Control
The authors are with the Robotics Institute, Athena Research Center, Marousi, Greece · The authors are with the Division of Systems and Automatic Control, Department of Electrical and Computer Engineering, University of Patras, Patras, Greece · The authors are with the Department of Informatics and Telecommunications, School of Sciences, University of Thessaly, Lamia, Greece

Mar 6, 2026J/K move · Enter open · S save

Uncertainty-Aware Adaptive Dynamics For Underwater Vehicle-Manipulator Robots

Edward Morgan, Nenyi K Dadson, Corina Barbalata

The authors are with the Department of Mechanical & Industrial Engineering, at Louisiana State University, Baton Rouge, LA 70803, USA

Robot Dynamics Identification Underwater Robotics Robot State Estimation

Open paper AskPDF arXiv

Accurate and adaptive dynamic models are critical for underwater vehicle-manipulator systems where hydrodynamic effects induce time-varying parameters. This paper introduces a novel uncertainty-aware adaptive dynamics model framework that remains linear in lumped vehicle and manipulator parameters, and embeds convex physical consistency constraints during online estimation. Moving horizon estimation is used to stack horizon regressors, enforce realizable inertia, damping, friction, and hydrostatics, and quantify uncertainty from parameter evolution. Experiments on a BlueROV2 Heavy with a 4-DOF manipulator demonstrate rapid convergence and calibrated predictions. Manipulator fits achieve R2 = 0.88 to 0.98 with slopes near unity, while vehicle surge, heave, and roll are reproduced with good fidelity under stronger coupling and noise. Median solver time is approximately 0.023 s per update, confirming online feasibility. A comparison against a fixed parameter model shows consistent reductions in MAE and RMSE across degrees of freedom. Results indicate physically plausible parameters and confidence intervals with near 100% coverage, enabling reliable feedforward control and simulation in underwater environments.

Abstract

Robot Dynamics Identification

Explore similar work

Uncertainty-Aware Adaptive Dynamics For Underwater Vehicle-Manipulator Robots

Optimal Excitation Trajectories for System Identification of Underwater Vehicles

Adaptive Prescribed Performance Trajectory Tracking for Uncertain Autonomous Underwater Vehicles With Input Saturation