Session: 04-06-02 Underwater Vehicles and Subsea Communications II

Submission Number: 181202

Hydrodynamic Wake Tracking for Autonomous Underwater Vehicles Using Bio-Inspired Pressure Sensing in a Potential-Flow Simulation 

Aquatic animals rely heavily on their lateral line systems to interpret hydrodynamic cues, using flow disturbances such as wakes to perform complex behaviors including prey detection, schooling, and rheotaxis.  These wakes, often composed of coherent vortex structures, carry detailed information about the motion, size, and propulsion characteristics of their source.  Inspired by this biological capability, recent research has sought to develop artificial lateral line sensors capable of identifying hydrodynamic patterns for underwater navigation and target detection.  Prior studies have demonstrated that pressure sensor arrays can distinguish canonical von Kármán vortex streets from uniform flow and extract dominant wake features such as shedding frequency and convective velocity.  Building on these advances, the next step is to enable dynamic tracking, allowing a mobile platform to identify and follow a wake as it evolves downstream.  Such a capability would represent a critical shift from stationary flow characterization to real-time, motion-based sensing. However, existing systems rely on large, fixed arrays positioned directly within the flow, which limits their integration into compact autonomous underwater vehicles (AUVs) and constrains their effectiveness for tracking scenarios where the wake may approach from arbitrary angles.

The ability to passively identify and track wakes offers significant potential for underwater robotics, particularly in surveillance, swarm coordination, and bio-inspired navigation.  Unlike active sonar, which emits detectable signals, passive hydrodynamic sensing enables stealthy operation and reduces environmental impact.  By observing local pressure variations induced by nearby flow structures, an AUV could infer the presence and trajectory of a swimming body or another vehicle.  Such capabilities are particularly advantageous in acoustically cluttered or light-limited environments, where conventional sensing methods are unreliable.

This study presents a two-dimensional simulation framework for target trajectory tracking based on hydrodynamic wake cues.  The simulation uses a potential-flow formulation to generate a reverse von Kármán vortex street with the vortices represented as point vortices.   At each time step, the flow field is computed as the gradient of the velocity potential, representing an ideal, inviscid potential flow.  A virtual AUV equipped with distributed pressure sensors is placed within this simulated flow, and its guidance system uses the pressure distribution over the vehicle’s surface as feedback to infer wake direction and dynamically adjust its motion.  The simulation environment enables evaluation of wake interception performance under varying approach angles and sensor configurations, supporting the future development of robust control algorithms such as pure pursuit, line-of-sight, or lookahead-based control laws.

By combining accurate flow modeling with simulated pressure-based sensing, this work provides a realistic and flexible environment for developing and testing passive wake-tracking strategies.  This framework not only captures the essential physics of unsteady vortex dynamics but also reproduces the spatial and temporal variations in pressure that an onboard lateral line system would experience.  Preliminary results demonstrate the system’s ability to identify wake signatures and adjust its trajectory to converge toward the wake centerline, even when approaching from a variety of angles. These findings suggest that flow-based feedback can enable robust guidance behavior without requiring direct knowledge of the target’s position or motion state.

Beyond trajectory convergence, the framework supports extensive parametric studies and sensitivity analyses.  Variations in vortex strength, spacing, and background flow conditions can be introduced systematically, revealing how each influences detection range, sensor density requirements, and control response.  The resulting data will guide sensor placement and algorithm design for future hardware implementations.

This research advances the concept of bio-inspired wake tracking for autonomous underwater systems, bridging the gap between lateral line-based sensing and motion control. The resulting model functions as both a validation platform for artificial lateral line designs and a testbed for evaluating emerging control laws that exploit flow information directly.  In the future, such an approach may inform the development of small, low-power AUVs capable of tracking biological or engineered swimmers using only the hydrodynamic signatures they leave behind, offering a step toward stealthier, more adaptive, and environmentally aware underwater vehicles.

 

 

Presenting Author: Gary Glass University of Hawaii at Manoa

Presenting Author Biography: Gary Glass is a Ph.D. candidate in Ocean and Resources Engineering at the University of Hawaiʻi, studying bio-inspired underwater sensing.

Authors:

Gary Glass University of Hawaii at Manoa
Mike Krieg University of Hawaii at Manoa