SearcharxivSearch

arXiv subjects

Linhao Jin

Publications and source records attributed to Linhao Jin.

2 recordsLinked to original sources

A reconfigurable multi-axis cyber-physical framework for multi-regime fluid--structure interaction experiments

Changing the mechanical impedance and constraints of a fluid--structure interaction (FSI) experiment often requires modifying or rebuilding the physical apparatus. Here we present a reconfigurable cyber-physical water-tunnel system in which these properties can instead be defined and reassigned in software. Three translational degrees of freedom and one rotational degree of freedom can each independently prescribe motion, respond to measured fluid loads through user-defined virtual dynamics, or remain fixed, allowing active, passive, and constrained motions to be combined within the same experiment. A six-axis force/torque transducer supplies hydrodynamic loads to the real-time controller, while a common control architecture coordinates motion, virtual dynamics, mode switching, data acquisition, and flow-field measurements. We validate prescribed-motion operation using pitching-foil measurements that reproduce established thrust and power scaling trends, and force-responsive operation using an active-heave/passive-pitch oscillator that recovers the resonance trend of a published benchmark. We then reconfigure the same apparatus for intra-cycle active--passive pitching, coordinated vertical-axis turbine kinematics, active-heave/passive-surge locomotion, and body-fitted multilayer stereoscopic particle image velocimetry. These experiments span different kinematic configurations, mechanical impedances, constraints, and measurement sequences without changes to the core motion and sensing hardware. The system therefore provides an experimental architecture for studying multiple regimes of unsteady FSI by making mechanical impedance and constraints software-reconfigurable.

physics.flu-dyn

Write-Safe Flow Field Mapping under Ambiguous Onboard Sensing and Localization Drift

Mobile robots can infer local flow structure from onboard sensing, but a locally plausible estimate is not always safe to write into a global map. Similar flow structures may produce ambiguous observations, while localization drift causes predicted patches to be written at incorrect locations. Repeated misregistered updates then accumulate into persistent ghost structures. We address this failure mode with a map-reference-aware conservative fusion framework. The model predicts a local velocity patch and a learned write-safety score that continuously attenuates uncertain map updates while permitting initialization when no reliable map reference is available. Across synthetic jet and crossflow environments, the proposed method reduces average ghost contamination by 42% relative to ungated fusion. A zero-shot hardware replay using real pressure and optical-flow measurements from a thruster wake further reduces ghost contamination by 39% while retaining 81% map coverage. These results show that safe map writing is critical for flow mapping under ambiguous sensing and localization drift.

cs.RO