Modeling of a Ground Vehicle Operating in Shallow Water

Project begins 2022.

Unmanned Ground Vehicle (UGV) missions necessitate navigation in highly complex environments, including standing shallow water, such as in marshy, coastal, or riverine environments. In these challenging missions, considering the changing dynamics of the vehicle with varying water depth/flow and terrain conditions is critically important to develop robust and effective vehicle navigation algorithms. In a fuel-limited operation scenario in shallow water, for example, accurate models are needed to predict fuel consumption, and in turn can be employed to compute energy-optimal or fuel-constrained control actions in these environments. However, little work has been conducted to date on the autonomous UGV navigation in shallow water due to the difficulties in predicting the coupled vehicle-water dynamics and high computational costs for tuning a controller with physics-based models. There is also little or no experimental data available describing the forces that act on ground vehicles moving in water, such as traction changes due to buoyancy, affecting the vehicle’s effective weight and tire slippage, as well as resistance of the vehicle’s body. Nonetheless, there is evidence that UGV operation in shallow water is significant and that a foundational understanding of ground vehicle performance metrics is needed. With this backdrop of ideas, this research proposal puts forward a synergistic approach that combines expertise in computational and experimental fluid dynamics (CFD and EFD), computational multibody dynamics (MBD) and control theory towards the development of mathematical models for autonomous UGV navigation in shallow water.

The research that we aim to undertake unfolds in three main steps. First, the high-fidelity simulation will be used to develop reduced-order models (ROM) for the ground vehicle operating in unstructured environment and shallow water. Second, experiments with a scaled robotic vehicle in shallow water will be conducted and the ROM will be validated against the test data. Third, a controller will be developed with the ROM and integrated into the experimental platform to demonstrate the use of the proposed ROM for ground vehicle navigation in shallow water.

Publications:

• Yamashita, H., Martin, J. E., Sugiyama, H., Tison, N., Grunin, A., & Jayakumar, P. (2023, August). Predicting Vehicle Motion in Shallow Water with Data-Driven Hydrodynamics Model. In International Design Engineering Technical Conferences and Computers and Information in Engineering Conference (Vol. 87387, p. V010T10A017). American Society of Mechanical Engineers.
• Yamashita, H., Martin, J. E., Tison, N., Grunin, A., Jayakumar, P., & Sugiyama, H. (2024). Modeling of Vehicle Mobility in Shallow Water With Data-Driven Hydrodynamics Model. Journal of Computational and Nonlinear Dynamics, 19(7).

#3.20