arXiv subjects

Amirreza Rouhi

Publications and source records attributed to Amirreza Rouhi.

16 recordsLinked to original sources

arXiv2026

Leveraging unstructured grids for direct numerical simulations of wall turbulence

Towards computational cost saving for direct numerical simulations (DNSs) of wall turbulence, we formulate an unstructured grid-generation framework, termed $η$-grid, where the wall-normal ($y$) and spanwise ($z$) grid sizes are proportional to the local Kolmogorov scale $η$. The framework consists of an inner layer, with a thickness $\sim 50$ viscous units, with viscous-scaled grid sizes similar to a conventional DNS grid: $0.3 \lesssim Δy^+ \lesssim 4, Δz^+ \simeq 5$ over a smooth wall, and $\ell^+/30 \lesssim Δy^+, Δz^+ \lesssim 4$ over uneven surfaces, where $\ell^+$ is the smallest surface wavelength. Above the inner layer, $Δy^+ \simeq Δz^+ \simeq 2η^+$. We test $η$-grid with finite volume and spectral element solvers, and conduct DNSs of turbulent channel flows and boundary layers over smooth wall and various streamwise-aligned riblets, up to friction Reynolds number $δ^+_0 = 1000$. We assess the accuracy of $η$-grid against the conventional Cartesian grids, through comparison with the reference DNS and experimental data. Results from $η$-grid and the Cartesian grids differ by less than $1\%$, in terms of turbulence statistics up to second-order, and the energy spectra. For turbulent channel flows with $10^3 \lesssim δ^+_0 \lesssim 10^4$, the number of grid points with $η$-grid ($N_η$) scales $\propto {δ^+_0}^{2.47}$ over a smooth wall, and $\propto {δ^+_0}^{2.0-2.47}$ over riblets, whereas the number of grid points with a Cartesian grid and hyperbolic-tangent $y$-grid ($N_\mathrm{Tanh}$) scales $\propto {δ^+_0}^{3.0}$. By $δ^+_0 = 6000$, $N_η/N_\mathrm{Tanh} \simeq 0.1$ over a smooth wall, and $N_η/N_\mathrm{Tanh} \simeq 0.04$ over typical drag-reducing riblets, with viscous-scaled spacing $15$.

physics.flu-dyn↗

arXiv2026

Direct numerical simulations of turbulent drag reduction via piezoelectric actuation

We have conducted Direct Numerical Simulations of turbulent half-channel flow over realistic surface deformations at friction Reynolds number $Re_τ=200$. We generated the surface deformations using piezoelectric actuators. We simulated the piezoelectric actuation over the practical actuation frequency range $(119Hz\le f_\mathrm{act}\le543Hz)$ and voltage range $(250V\le Q \le500V)$ beneath an Aluminum sheet using Finite Element Analysis. The sheet deformation amplitude and actuation frequency in viscous units vary within the range $2 \le η^+_\mathrm{max} \le 34$, and $-0.58 \le ω^+ \le 0.70$. The vertical surface deformations from our actuation setup generate three types of waves: travelling, hybrid, and standing waves. Surface deformations are applied as bottom-wall boundary conditions of the turbulent channel flow to generate waves in the upstream, downstream, and spanwise directions. We achieved maximum drag reductions of 1.6\%, 5.4\%, and 27.6\% for upstream, downstream, and spanwise waves, respectively. The streamwise waves generate alternating adverse and favorable pressure gradients, which locally increase and decrease drag, leading to a marginal net change in drag. In contrast, spanwise waves introduce transverse shear, accompanied by high- and low-streamwise-momentum zones that respectively attenuate and energize the near-wall turbulence. Such disruption of the near-wall turbulence-regeneration cycle produces up to $27\%$ drag reduction for the realistic spanwise hybrid wave; such an outcome demonstrates the efficacy of unconventional realistic surface deformations in achieving significant drag reduction.

physics.flu-dyn↗

arXiv2026

RetailSMV: Exocentric vs. Egocentric Adaptation of Foundation Video World Models in Retail

Foundation video diffusion models are increasingly viewed as world simulators for embodied agents, yet their pretraining on internet-scale generic video leaves them poorly aligned with real-world deployment domains. We study parameter-efficient adaptation of a pretrained foundation video world model to retail scenes: when synchronized egocentric and exocentric video of the same activity are available, which viewpoint of training data produces the strongest adapted model? We introduce RetailSMV (Retail Synchronized Multi-View), a corpus of 32,105 captioned retail clips from five supermarkets with synchronized ego/exo capture from the store-staff perspective (stocking, arranging, weighing, managing supply carts, scanning at checkout), rather than the customer-centric framing of prior retail video corpora, and train three matched Low-Rank Adaptation (LoRA) configurations of Cosmos3-Nano (egocentric-only, exocentric-only, combined) under identical hyperparameters. On a 200-clip held-out test set evaluated with seven complementary metrics under a strict paired statistical protocol, exocentric-only adaptation matches or exceeds combined adaptation on six of seven point estimates and is significantly better on LPIPS, PSNR, and DreamSim, despite training on only 15,985 exocentric clips (versus 32,105 for combined). A symmetric paired comparison further shows that adding exocentric data to egocentric-only training helps while adding egocentric data to exocentric-only training hurts. The absolute adaptation gap is largest at the shortest rollout time, identifying the near-horizon prediction window as the regime in which adaptation is most beneficial.

cs.CV↗

arXiv2026

SABER: A Scalable Action-Based Embodied Dataset for Real-World VLA Adaptation

Robotic deployment in real-world environments depends on rich, domain-specific action data as much as on strong model architecture. General-purpose robot foundation models show modest performance in complex unseen tasks such as manipulation in a retail domain when applied out of the box. The root cause is a data gap: retail environments are structurally absent from general robot pretraining distributions, and the path to filling that gap through teleoperation is prohibitively expensive, logistically constrained, and difficult to scale. We introduce SABER, a high-fidelity retail robotics action dataset built from over 100 hours of natural in-store capture across multiple real grocery environments. Egocentric footage from head-mounted cameras records fine-grained hand activity at the point of interaction, while exocentric 360-degree scene footage from DreamVu's ALIA camera simultaneously observes all actors and activities across the entire space. This combination yields a uniquely complete picture of human retail behavior: dexterous hand activity, whole-body motion, and scene dynamics, all captured without staging, scripting, or teleoperation overhead. The SABER corpus contains 44.8K training samples across three action representation streams: 25K latent action sequences via LAPA-style encoding, 18.6K dexterous hand-pose trajectories retargeted to robot joint space, and 1.2K whole-body synchronized motion sequences retargeted to a humanoid embodiment. When applied to GR00T N1.6 via a shared-backbone multi-task post-training recipe, SABER yields a mean success rate of 29.3% across ten retail manipulation tasks -- more than 2.19x over fine-tuning baselines (13.4%). SABER demonstrates that the path to capable retail robots runs through better data, which can be collected today, at scale, without a robot in the loop. The dataset and code are available at https://dreamvu.ai/saber

cs.RO↗

arXiv2026

Non-equilibrium effects in turbulent boundary layers over riblets: DNS of step changes in surface texture

We computationally study the response of zero-pressure-gradient (ZPG) turbulent boundary layers (TBLs) to streamwise step changes from a smooth wall to riblets (SM_RI), and vice versa (RI_SM). To quantify the departure from equilibrium due to the step changes, we conduct reference calculations of ZPG TBLs over an entirely smooth wall, and an entirely riblet-covered surface. To save the computational cost, we generate an optimal grid for an unstructured spectral-element code, consistent with the size of turbulent scales across the TBL. By the step change, the momentum thickness Reynolds number reaches $Re_{θ_0} \simeq 680$ (friction Reynolds number $Re_{τ_0} \simeq 283$), and by the domain outlet downstream of the step change, $Re_θ \simeq 1000$ ($Re_τ \simeq 400$). The TBL departure from equilibrium due to the step change, and its subsequent relaxation, recall previous studies on step changes in surface roughness. Downstream of the step change, growth of the internal equilibrium layer thickness $δ_\text{IEL}$, hence recovery to equilibrium, follows two stages. Stage I corresponds to the recovery up to the buffer region ($y^+ \simeq 10$), which is slower during the RI_SM step change than the SM_RI counterpart. For the RI_SM cases during Stage I, $δ_\text{IEL} \propto (x/k)^{0.6}$, and this stage is completed by $x \simeq 100k \simeq (5δ_0 - 20δ_0)$ downstream of the step change, where $k$ is the riblet height. Stage II recovery i.e.\ recovery of the outer region, is quite slow. Therefore, for drag-increasing riblets with $k^+ \ge 25$, $δ_\text{IEL}$ does not reach the boundary layer thickness, even up to $50δ_0$ downstream of the step change, owing to the advected frozen wake from upstream. As a result, skin-friction coefficient reaches to more than $90\%$ of its equilibrium counterpart, but does not reach its $100\%$.

physics.app-ph↗

arXiv2026

PRISM: A Multi-View Multi-Capability Retail Video Dataset for Embodied Vision-Language Models

A critical gap exists between the general-purpose visual understanding of state-of-the-art physical AI models and the specialized perceptual demands of structured real-world deployment environments. We present PRISM, a 270K-sample multi-view video supervised fine-tuning (SFT) corpus for embodied vision-language-models (VLMs) in real-world retail environments. PRISM is motivated by a simple observation - physical AI systems fail not because of poor visual recognition, but because they do not understand space, physical dynamics and embodied action well enough to operate reliably in the world. To this end, PRISM is grounded in a novel three-dimensional knowledge ontology that spans spatial knowledge, temporal and physical knowledge, and embodied action knowledge. It covers 20+ capability probes across four evaluation dimensions - Embodied Reasoning (ER), Common Sense (CS), Spatial Perception (SP), and Intuitive Physics (IP), and to our knowledge, PRISM is the first dataset to instantiate all three knowledge dimensions within a single real-world deployment domain. The corpus captures data from egocentric, exocentric and 360° viewpoints across five supermarket locations and includes open-ended, chain-of-thought, and multiple-choice supervision. At 4 fps, PRISM spans approximately 11.8M video frames and approximately 730M tokens, placing it among the largest domain-specific video SFT corpora. Fine-tuning on PRISM reduces the error rate across all 20+ probes by 66.6% over the pre-trained baseline, with significant gains in embodied action understanding where the accuracy improves by 36.4%. Our results suggest that ontology-structured, domain specific SFT can meaningfully strengthen embodied VLMs for real-world settings. The PRISM dataset and more details are available at https://dreamvu.ai/prism

cs.CV↗

arXiv2025

Application of riblets to separating turbulent boundary layers

We conduct direct numerical simulations of separating turbulent boundary layers (TBLs) over triangular riblets with tip angles $90^o$ (T9) and $60^o$ (T6). Our setup follows the separating TBL study of Wu et al.\ ({\it J. Fluid Mech.}, vol.\ 883, 2020, p.\ A45). An equilibrium zero pressure-gradient (ZPG) TBL is generated at a reference location, followed by imposition of a Gaussian suction profile to create a separation bubble. The ZPG TBLs over the riblets and the benchmark smooth case have matched momentum thickness Reynolds number $Re_{θ_0} = 583$ (friction Reynolds number 224). We employ a well-validated spectral-element solver, and leverage its unstructured-grid nature to generate an optimal grid, based on the size of turbulent scales across the TBL. At the reference location, the T9 and T6 riblets respectively increase and reduce drag, with viscous-scaled spacings $52$ and $13$. We discover that for both riblet cases, the mean separation location occurs at a distance of $140θ_0$ downstream of the reference location, $18\%$ shorter than the mean separation distance for the smooth case ($170θ_0$). This outcome is related to the progressive enhancement of the Kelvin-Helmholtz (KH) rollers over the riblets, owing to the continuous rise in the adverse pressure-gradient. The KH rollers penetrate into the turbulent separation bubble, with significantly larger size and coherence compared to their counterparts upstream of the mean separation location.

physics.app-ph↗

arXiv2025

LRDDv2: Enhanced Long-Range Drone Detection Dataset with Range Information and Comprehensive Real-World Challenges

The exponential growth in Unmanned Aerial Vehicles (UAVs) usage underscores the critical need of detecting them at extended distances to ensure safe operations, especially in densely populated areas. Despite the tremendous advances made in computer vision through deep learning, the detection of these small airborne objects remains a formidable challenge. While several datasets have been developed specifically for drone detection, the need for a more extensive and diverse collection of drone image data persists, particularly for long-range detection under varying environmental conditions. We introduce here the Long Range Drone Detection (LRDD) Version 2 dataset, comprising 39,516 meticulously annotated images, as a second release of the LRDD dataset released previously. The LRDDv2 dataset enhances the LRDDv1 by incorporating a greater variety of images, providing a more diverse and comprehensive resource for drone detection research. What sets LRDDv2 apart is its inclusion of target range information for over 8,000 images, making it possible to develop algorithms for drone range estimation. Tailored for long-range aerial object detection, the majority of LRDDv2's dataset consists of images capturing drones with 50 or fewer pixels in 1080p resolution. For access to the complete Long-Range Drone Detection Dataset (LRDD)v2, please visit https://research.coe.drexel.edu/ece/imaple/lrddv2/ .

cs.CV↗

arXiv2025

ADAM: Autonomous Discovery and Annotation Model using LLMs for Context-Aware Annotations

Object detection models typically rely on predefined categories, limiting their ability to identify novel objects in open-world scenarios. To overcome this constraint, we introduce ADAM: Autonomous Discovery and Annotation Model, a training-free, self-refining framework for open-world object labeling. ADAM leverages large language models (LLMs) to generate candidate labels for unknown objects based on contextual information from known entities within a scene. These labels are paired with visual embeddings from CLIP to construct an Embedding-Label Repository (ELR) that enables inference without category supervision. For a newly encountered unknown object, ADAM retrieves visually similar instances from the ELR and applies frequency-based voting and cross-modal re-ranking to assign a robust label. To further enhance consistency, we introduce a self-refinement loop that re-evaluates repository labels using visual cohesion analysis and k-nearest-neighbor-based majority re-labeling. Experimental results on the COCO and PASCAL datasets demonstrate that ADAM effectively annotates novel categories using only visual and contextual signals, without requiring any fine-tuning or retraining.

cs.CV↗

arXiv2025

Attached Decelerating Turbulent Boundary Layers over Riblets

Turbulent boundary layers over riblets subjected to adverse pressure gradients (APGs) are investigated by direct numerical simulation. Multiple APG strengths and riblet sizes are examined, permitting evaluation of drag modification by riblets, and associated physical mechanisms, in various regimes established for zero-pressure-gradient (ZPG) riblet flows. The APG strengths are selected such that the flow remains attached. It is found that during APGs, riblets reduce drag beyond what has been achieved in ZPG flows. In extreme cases, an upstream force (i.e., negative drag) is attained. The significant drag reduction is found to be a product of Kelvin-Helmholtz roller vortices forming near the riblet crest, which are augmented in size, strength, and frequency during the APG. The preliminary results reported here indicate the need to modify existing metrics to predict drag reduction and the onset of KH rollers by riblets when the pressure gradient is non-negligible. Further analysis will be documented in the final paper.

physics.flu-dyn↗

arXiv2024

Spanwise wall forcing can reduce turbulent heat transfer more than drag

Direct numerical simulations are performed of turbulent forced convection in a half channel flow with wall oscillating either as a spanwise plane oscillation or to generate a streamwise travelling wave. The friction Reynolds number is fixed at $Re_{τ_0} = 590$, but the Prandtl number $Pr$ is varied from 0.71 to 20. For $Pr>1$, the heat transfer is reduced by more than the drag, 40\% compared to 30\% at $Pr=7.5$. This outcome is related to the different responses of the velocity and thermal fields to the Stokes layer. It is shown that the Stokes layer near the wall attenuates the large-scale energy of the turbulent heat-flux and the turbulent shear-stress, but amplifies their small-scale energy. At higher Prandtl numbers, the thinning of the conductive sublayer means that the energetic scales of the turbulent heat-flux move closer to the wall, where they are exposed to a stronger Stokes layer production, increasing the contribution of the small-scale energy amplification. A predictive model is derived for the Reynolds and Prandtl number dependence of the heat-transfer reduction based on the scaling of the thermal statistics. The model agrees well with the computations for Prandtl numbers up to 20.

physics.flu-dyn↗

arXiv2023

Learning Scene Context Without Images

Teaching machines of scene contextual knowledge would enable them to interact more effectively with the environment and to anticipate or predict objects that may not be immediately apparent in their perceptual field. In this paper, we introduce a novel transformer-based approach called $LMOD$ ( Label-based Missing Object Detection) to teach scene contextual knowledge to machines using an attention mechanism. A distinctive aspect of the proposed approach is its reliance solely on labels from image datasets to teach scene context, entirely eliminating the need for the actual image itself. We show how scene-wide relationships among different objects can be learned using a self-attention mechanism. We further show that the contextual knowledge gained from label based learning can enhance performance of other visual based object detection algorithm.

cs.CV↗

arXiv2023

Recovery of the wall-shear stress to equilibrium flow conditions after a rough-to-smooth step-change in turbulent boundary layers

This paper examines recovery of the wall-shear stress of a turbulent boundary layer that has undergone a sudden transition from a rough to a smooth surface. Early works of Antonia and Luxton questioned the reliability of standard smooth-wall methods to measure wall-shear stress in such conditions, and subsequent studies show significant disagreement depending on the approach used to determine the wall-shear stress downstream. Here we address this by utilising a collection of experimental databases at Re_τ\approx 4100 that have access to both `direct' and `indirect' measures of the wall-shear stress to understand the recovery to equilibrium conditions to the new surface. Our results reveal that the viscous region (z^+\lesssim 4) recovers almost immediately to an equilibrium state with the new wall conditions, however, the buffer region and beyond takes several boundary layer thicknesses before recovering to equilibrium conditions, which is longer than previously thought. A unique direct numerical simulation database of a wall-bounded flow with a rough-to-smooth wall transition is employed to confirm these findings. In doing so, we present evidence that any estimate of the wall-shear stress from the mean velocity profile in the buffer region or further away from the wall tends to underestimate its magnitude in the near vicinity of the rough-to-smooth transition, and this is likely to be partly responsible for the large scatter of recovery lengths to equilibrium conditions reported in the literature. Our results also reveal that the smaller energetic scales in the near-wall region recover to an equilibrium state associated with the new wall conditions within one boundary layer thickness downstream of the transition, while the larger energetic scales exhibit an over-energised state for several boundary layer thicknesses downstream of the transition.

physics.flu-dyn↗

arXiv2023

Turbulent drag reduction by spanwise wall forcing. Part 2: High-Reynolds-number experiments

Here, we present measurements of turbulent drag reduction in boundary layers at high friction Reynolds numbers in the range of $4500 \le Re_τ\le 15000$. The efficacy of the approach, using streamwise travelling waves of spanwise wall oscillations, is studied for two actuation regimes: (i) inner-scaled actuation (ISA), as investigated in Part 1 of this study, which targets the relatively high-frequency structures of the near-wall cycle, and (ii) outer-scaled actuation (OSA), which was recently presented by Marusic et al. (Nat. Commun., vol. 12, 2021) for high-$Re_τ$ flows, targeting the lower-frequency, outer-scale motions. Multiple experimental techniques were used, including a floating-element balance to directly measure the skin-friction drag force, hot-wire anemometry to acquire long-time fluctuating velocity and wall-shear stress, and stereoscopic-PIV (particle image velocimetry) to measure the turbulence statistics of all three velocity components across the boundary layer. Under the ISA pathway, drag reduction of up to 25% was achieved, but mostly with net power saving losses due to the high-input power cost associated with the high-frequency actuation. The low-frequency OSA pathway, however, with its lower input power requirements, was found to consistently result in positive net power savings of 5 - 10%, for moderate drag reductions of 5 - 15%. The results suggest that OSA is an attractive pathway for energy-efficient drag reduction in high Reynolds number applications. Both ISA and OSA strategies are found to produce complex inter-scale interactions, leading to attenuation of the turbulent fluctuations across the boundary layer for a broad range of length and time scales.

physics.flu-dyn↗

arXiv2023

Turbulent drag reduction by spanwise wall forcing. Part 1: Large-eddy simulations

Turbulent drag reduction through streamwise travelling waves of spanwise wall oscillation is investigated over a wide range of Reynolds numbers. Here, in Part 1, wall-resolved large-eddy simulations in a channel flow are conducted to examine how the frequency and wavenumber of the travelling wave influence the drag reduction at friction Reynolds numbers $Re_τ= 951$ and $4000$. The actuation parameter space is restricted to the inner-scaled actuation (ISA) pathway, where drag reduction is achieved through direct attenuation of the near-wall scales. The level of turbulence attenuation, hence drag reduction, is found to change with the near-wall Stokes layer protrusion height $\ell_{0.01}$. A range of frequencies is identified where the Stokes layer attenuates turbulence, lifting up the cycle of turbulence generation and thickening the viscous sublayer; in this range, the drag reduction increases as $\ell_{0.01}$ increases up to $30$ viscous units. Outside this range, the strong Stokes shear strain enhances near-wall turbulence generation leading to a drop in drag reduction with increasing $\ell_{0.01}$. We further find that, within our parameter and Reynolds number space, the ISA pathway has a power cost that always exceeds any drag reduction savings. This motivates the study of the outer-scaled actuation (OSA) pathway in Part 2, where drag reduction is achieved through actuating the outer-scaled motions.

physics.flu-dyn↗

arXiv2020

Near wall coherence in wall-bounded flows and implications for flow control

Opposition-control of the energetic cycle of near wall streaks in wall-bounded turbulence, using numerical approaches, has shown promise for drag reduction. For practical implementation, opposition control is only realizable if there is a degree of coherence between the sensor--actuator pairs of the control system (these sensors and actuators should typically be wall-based to avoid parasitic drag). As such, we here inspect the feasibility of real-time control of the near-wall cycle, by considering the coherence between a measurable wall-quantity, being the wall-shear stress fluctuations, and the streamwise and wall-normal velocity fluctuations in a turbulent boundary layer. Synchronized spatial and temporal velocity data from numerical simulations at $Re_τ\approx 590$ and $ 2000$ are employed. It is shown that the spectral energy of the streamwise velocity fluctuations that is stochastically incoherent with wall signals is independent of Reynolds number in the near wall region. Consequently, the streamwise energy-fraction that is stochastically wall-coherent grows with Reynolds number due to the increasing range of energetic large scales. This thus implies that a wall-based control system has the ability to manipulate a larger portion of the total turbulence energy at off-wall locations, at higher Reynolds numbers. Coherence values of 0.55 and 0.4, which are considerably lower than the maximum possible coherence 1, were found between the streamwise and wall-normal velocity fluctuations at the near wall peak in the energy spectrogram, respectively, and the streamwise fluctuating friction velocity. This suggests that a closed-loop drag reduction scheme targeting near wall cycle of streaks alone will be of limited success in practice as the Reynolds number grows.

physics.flu-dyn↗