Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Uncertainty-Aware Adaptive Dynamics For Underwater Vehicle-Manipulator Robots

Edward Morgan, Nenyi K Dadson, Corina Barbalata

Abstract

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.

Uncertainty-Aware Adaptive Dynamics For Underwater Vehicle-Manipulator Robots

Abstract

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.
Paper Structure (18 sections, 14 equations, 10 figures, 1 table)

This paper contains 18 sections, 14 equations, 10 figures, 1 table.

Table of Contents

  1. Introduction
  2. Related Work
  3. Contribution
  4. Problem Formulation
  5. UVMS Description
  6. Vehicle Subsystem
  7. Manipulator Subsystem
  8. Problem Statement
  9. Methodology
  10. Underwater Vehicle Regressor Formulation
  11. Manipulator Regressor Formulation
  12. Complete System Regressor
  13. Parameter Estimation
  14. Uncertainty Quantification
  15. Use Cases of Estimated Parameters
  16. ...and 3 more sections

Figures (10)

  • Figure 1: Adaptive dynamics in UVMS: prior knowledge of system parameters is updated using new experimental data, resulting in an updated belief consistent with observed dynamics.
  • Figure 2: Framework overview of the Uncertainty-Aware Adaptive Dynamics scheme for UVMS, combining regressor models, horizon stacking, moving horizon estimation, and physical consistency constraints to produce adaptive parameter estimates with uncertainty for control and planning.
  • Figure 3: Identified fit for excited and recorded vehicle (a) surge force, (b) heave force and (c) roll moment DOF.
  • Figure 4: Identified fit for excited and recorded torque profile of manipulator: (a) joint 0 and (b) joint 1.
  • Figure 5: Identified fit for excited and recorded torque profile of manipulator: (a) joint 2 and (b) joint 3.
  • ...and 5 more figures