The Coupling Effect: Experimental Validation of the Fusion of Fossen and Featherstone to Simulate UVMS Dynamics in Julia

TL;DR

This work addresses the challenge of accurately modeling dynamic coupling in UVMSs, especially for lightweight vehicles where manipulator motion can induce vehicle attitude changes. It introduces a Julia-based simulator that couples Featherstone's open-chain rigid-body dynamics with Fossen's underwater hydrodynamics, solving the system as an open chain with a floating base and validating against pool tests with a ten-DOF UVMS. The study demonstrates that the model reproduces the qualitative behavior and steady-state orientations, achieving RMSE on the order of a couple degrees, though transients are faster in the real system and exhibit a consistent temporal offset; analysis points to added-mass parameter choices, unmodeled actuator dynamics, and environmental stochasticity as core sources of discrepancy. The results highlight the value of an open, transparent, and high-fidelity tool for UVMS design and control development, while outlining concrete avenues—such as incorporating actuator dynamics and refining hydrodynamic parameters—for improving sim-to-real accuracy. By aligning hydrodynamics with rigid-body chain dynamics and validating against hardware data, the work lays groundwork for broader application to UVMS analysis and to back-end solvers like Stonefish.

Abstract

As Underwater Vehicle Manipulator Systems (UVMSs) have gotten smaller and lighter over the past years, it is becoming increasingly important to consider the coupling forces between the manipulator and the vehicle when planning and controlling the system. A number of different models have been proposed, each using different rigid body dynamics or hydrodynamics algorithms, or purporting to consider different dynamic effects on the system, but most go without experimental validation of the full model, and in particular, of the coupling effect between the two systems. In this work, we return to a model combining Featherstone's rigid body dynamics algorithms with Fossen's equations for underwater dynamics by using the Julia package RigidBodyDynamics.jl. We compare the simulation's output with experimental results from pool trials with a ten degree of freedom UVMS that integrates a Reach Alpha manipulator with a BlueROV2. We validate the model's usefulness and identify its strengths and weaknesses in studying the dynamic coupling effect.

Figures (6)

• Figure 1: A lightweight UVMS. (Left) The equilibrium position if ballast is added to offset the manipulator weight. If no ballast is added, the vehicle pitches significantly at the home position (center) and even more in the extended position (right).
• Figure 2: Images of the BlueROV2/Alpha UVMS, as visualized in MeshCat from the Julia simulation. (Left) A right-handed Cartesian coordinate frame is assigned at the vehicle's center of mass, using forward-left-up notation. (Right) The Alpha's four joints: a base joint parallel to yaw, two joints to extend the arm, and a rotational joint at the wrist.
• Figure 3: (Top) Snapshots of the MeshCat visualization of the UVMS performing an example trajectory. (Bottom left) Manipulator joint angles over time. (Bottom right) Roll, pitch, and yaw of the vehicle over time.
• Figure 4: Left A diagram of the underwater Motion Capture System cameras as placed in the pool. Center An image of the BlueROV2 in the pool during experimental trials. Right The body frame of the BlueROV2, using a forward - left - up right handed frame. The MoCap markers used to localize the rigid body are in an "L" shape on the top of the vehicle.
• Figure 5: Vehicle orientation obtained from the Motion Capture system and the onboard IMU for one trial in which the manipulator followed a trajectory. The MoCap orientation switches frequently between pose estimations based on which of the markers are in view.