arXiv subjects

Junming Zhao

Publications and source records attributed to Junming Zhao.

At least 19 recordsLinked to original sources

arXiv2026

Distributed Circuit Model for Predicting the Quality Factor of Magnetic Polariton Resonance

Existing RLC circuit model leads to inaccurate predictions of the quality factor (Q-factor) of magnetic polariton (MP) resonances under imperfect absorption conditions due to the omission of radiation loss. Moreover, the lumped-parameter nature of RLC models also limits their applicability for predicting higher-order MP modes. In this letter, we propose a distributed circuit model (DCM) for predicting the Q-factor of MP resonances, which overcomes these limitations. By introducing a radiation resistance to characterize radiation loss and establishing a mapping between distributed and lumped parameters, we derive a unified analytical expression for the Q-factor of arbitrary-order MPs. Validation via rigorous coupled-wave analysis (RCWA) demonstrates that this model can accurately predict the Q-factors of MPs of all orders in various metal-insulator-metal (MIM) structures. This work provides a simple yet effective tool for designing metamaterial emitters/absorbers and advances the understanding of MP loss mechanisms.

physics.app-ph↗

arXiv2026

Predicting Multi-Order Magnetic Polariton Resonances for Radiative Properties Tailoring by Distributed Circuit Model

Surface plasmon polaritons (SPPs) and magnetic polaritons (MPs) are fundamental resonance modes that are widely used to tailor the thermal radiation properties of micro/nanostructured metamaterials. Lumped circuit models (LCMs) are usually constructed empirically to describe the MP resonance conditions, and different LCMs have to be constructed for different orders of MPs, but these are difficult to be built for high-order MP modes due to the complex electromagnetic field distribution. This work proposes a new type of circuit model, distributed circuit model (DCM), to describe and predict multi-order MP resonances inside the structure based on the minimum total impedance condition. This allows both fundamental and high-order MP resonances to be predicted with a unified circuit, significantly simplifying the analysis of high-order MPs. More importantly, the DCM shares a similar and clear physical picture as the LCM for describing MPs. The MP resonance conditions for four typical structures are derived. Theoretical predictions based on DCMs are compared with and validated by rigorous numerical simulations. This study deepens the understanding and facilitates the design of MP-based thermal radiation metamaterials.

physics.optics↗

arXiv2026

OpenHLM: An Empirical Recipe for Whole-Body Humanoid Loco-Manipulation

Whole-body humanoid loco-manipulation requires coordinating the robot's entire kinematic chain. However, most existing systems typically decouple the upper and lower bodies into separate controllers, limiting such coordination and yielding behaviors similar to those of a wheeled dual-arm platform. In this paper, we ask what it takes to build a whole-body native vision-language-action (VLA) model that maps language and pixels directly to all of the humanoid's degrees of freedom. We conduct a systematic empirical study organized as a roadmap of one-variable-at-a-time experiments across three phases: whole-body teleoperation, VLA model design, and heterogeneous co-training. Our study yields several intriguing findings: a joint-based whole-body teleoperation interface outperforms alternatives that only partially expose the humanoid's degrees of freedom; a VLA pretrained on static and wheeled dual-arm platforms transfers surprisingly well to a humanoid's full action space; and co-training with HuMI, the humanoid analog of UMI, extends the policy to new objects and instructions without additional whole-body teleoperation on those targets. Following this roadmap yields OpenHLM, an open-source recipe for whole-body humanoid loco-manipulation. In a challenging long-horizon task that spans a wide vertical range of the humanoid, OpenHLM outperforms two state-of-the-art humanoid VLA baselines (GR00T N1.6 and $\Psi_0$) using less than half the total demonstration time. Our code, training data, and model checkpoints are available at [https://openhlm-project.github.io/].

cs.RO↗

arXiv2026

Superballistic transport of thermal photons in confined many-body systems

Ballistic transport, realized when the system size is smaller than the mean free path of energy carriers, is traditionally regarded as the ultimate limit for energy transfer. Here, we predict a superballistic radiative heat transport regime that surpasses this limit in dilute chains of plasmonic nanoparticles confined within cavities. This anomalous regime exhibits superlinear scaling of the effective thermal conductivity (k ~L^1.5) and originates from the amplification of long-range interactions mediated by cavity-guided modes. Our results establish a framework for ultrafast photonic heat transport and open pathways for thermal management, information processing and energy transfer in quantum and nanoscale systems.

physics.optics↗

arXiv2026

Humanoid Manipulation Interface: Humanoid Whole-Body Manipulation from Robot-Free Demonstrations

Current approaches for humanoid whole-body manipulation, primarily relying on teleoperation or visual sim-to-real reinforcement learning, are hindered by hardware logistics and complex reward engineering. Consequently, demonstrated autonomous skills remain limited and are typically restricted to controlled environments. In this paper, we present the Humanoid Manipulation Interface (HuMI), a portable and efficient framework for learning diverse whole-body manipulation tasks across various environments. HuMI enables robot-free data collection by capturing rich whole-body motion using portable hardware. This data drives a hierarchical learning pipeline that translates human motions into dexterous and feasible humanoid skills. Extensive experiments across five whole-body tasks--including kneeling, squatting, tossing, walking, and bimanual manipulation--demonstrate that HuMI achieves a 3x increase in data collection efficiency compared to teleoperation and attains a 70% success rate in unseen environments.

cs.RO↗

arXiv2025

EasyInsert: A Data-Efficient and Generalizable Insertion Policy

Robotic insertion is a highly challenging task that requires exceptional precision in cluttered environments. Existing methods often have poor generalization capabilities. They typically function in restricted and structured environments, and frequently fail when the plug and socket are far apart, when the scene is densely cluttered, or when handling novel objects. They also rely on strong assumptions such as access to CAD models or a digital twin in simulation. To address these limitations, we propose EasyInsert. Inspired by human intuition, it formulates insertion as a delta-pose regression problem, which unlocks an efficient, highly scalable data collection pipeline with minimal human labor to train an end-to-end visual policy. During execution, the visual policy predicts the relative pose between plug and socket to drive a multi-phase, coarse-to-fine insertion process. EasyInsert demonstrates strong zero-shot generalization capability for unseen objects in cluttered environments, robustly handling cases with significant initial pose deviations. In real-world experiments, by leveraging just 1 hour of human teleoperation data to bootstrap a large-scale automated data collection process, EasyInsert achieves an over 90% success rate in zero-shot insertion for 13 out of 15 unseen novel objects, including challenging objects like Type-C cables, HDMI cables, and Ethernet cables. Furthermore, requiring only a single manual reset, EasyInsert allows for fast adaptation to novel test objects through automated data collection and fine-tuning, achieving an over 90% success rate across all 15 objects.

cs.RO↗

arXiv2025

OneTwoVLA: A Unified Vision-Language-Action Model with Adaptive Reasoning

General-purpose robots capable of performing diverse tasks require synergistic reasoning and acting capabilities. However, recent dual-system approaches, which separate high-level reasoning from low-level acting, often suffer from challenges such as limited mutual understanding of capabilities between systems and latency issues. This paper introduces OneTwoVLA, a single unified vision-language-action model that can perform both acting (System One) and reasoning (System Two). Crucially, OneTwoVLA adaptively switches between two modes: explicitly reasoning at critical moments during task execution, and generating actions based on the most recent reasoning at other times. To further unlock OneTwoVLA's reasoning and generalization capabilities, we design a scalable pipeline for synthesizing embodied reasoning-centric vision-language data, used for co-training with robot data. We validate OneTwoVLA's effectiveness through extensive experiments, highlighting its superior performance across four key capabilities: long-horizon task planning, error detection and recovery, natural human-robot interaction, and generalizable visual grounding, enabling the model to perform long-horizon, highly dexterous manipulation tasks such as making hotpot or mixing cocktails.

cs.RO↗

arXiv2025

Verifying Device Drivers with Pancake

Device driver bugs are the leading cause of OS compromises, and their formal verification is therefore highly desirable. To the best of our knowledge, no realistic and performant driver has been verified for a non-trivial device. We propose Pancake, an imperative language for systems programming that features a well-defined and verification-friendly semantics. Leveraging the verified compiler backend of the CakeML functional language, we develop a compiler for Pancake that guarantees that the binary retains the semantics of the source code. Usng automatic translation of Pancake to the Viper SMT front-end, we verify a performant driver for an Ethernet NIC.

cs.PL↗

arXiv2023

Effect of graphene grating coating on near-field radiative heat transfer

In this work we analyze the near-field radiative heat transfer (NFRHT) between finite-thickness planar fused silica slabs coated with graphene gratings. We go beyond the effective medium approximation by using an exact Fourier Modal Method (FMM) equipped with specific Local Basis Functions (LBF), and this is needed for realistic experimental analysis. In general, coating a substrate with a full graphene sheet has been shown to decrease the NFRHT at short separations (typically for d<100 nm) compared to the bare substrates, where the effective medium approximation consistently overestimates the radiative heat flux, with relative errors exceeding 50%. We show that, by patterning the graphene sheet into a grating, the topology of the plasmonic graphene mode changes from circular to hyperbolic, allowing to open more channels for the energy transfer between the substrates. We show that, at short separations, the NFRHT between slabs coated with graphene gratings is higher than that between full-graphene-sheet coated slabs and also than that between uncoated ones. We show a significant dependence of the radiative heat transfer on the chemical potential, which can be applied to modulate in situ the scattering properties of the graphene grating without any geometric alterations. We also compare the exact calculation with an approximate additive one and show that this approximation performs quite well for low chemical potentials. This work has the potential to unveil new avenues for harnessing non-additive heat transfer effects in graphene-based nanodevices.

physics.app-ph↗

arXiv2023

Favorable and unfavorable many-body interactions for near-field radiative heat transfer in nanoparticle networks

Near-field radiative heat transfer (NFRHT) in nanoparticle networks is complicated due to the multiple scattering of thermally excited electromagnetic wave (namely, many-body interaction, MBI). The MBI regime is analyzed using the many-body radiative heat transfer theory at the particle scale for networks of a few nanoparticles. Effect of MBI on radiative heat diffusion in networks of a large number of nanoparticles is analyzed using the normal-diffusion radiative heat transfer theory at the continuum scale. An influencing factor $\psi$ is defined to numerically figure out the border of the different many-body interaction regimes. The whole space near the two nanoparticles can be divided into four zones, non-MBI zone, enhancement zone, inhibition zone and forbidden zone, respectively. Enhancement zone is relatively smaller than the inhibition zone, so many particles can lie in the inhibiting zone that the inhibition effect of many-body interaction on NFRHT in nanoparticle networks is common in literature. Analysis on the radiative thermal energy confirms that multiple scattering caused by the inserted scatter accounts for the enhancement and inhibition of NFRHT. By arranging the nanoparticle network in aspect of structures and optical properties, the MBI can be used to modulate radiative heat diffusion characterized by the radiative effective thermal conductivity ($k_{\rm eff}$) over a wide range, from inhibition (over 55% reduction) to amplification (30 times of magnitude). To achieve a notable MBI, it is necessary to introduce particles that have resonances well-matched with those of the particles of interest, irrespective of their match with the Planckian window. This work may help for the understanding of the thermal radiation in nanoparticle networks.

physics.app-ph↗

arXiv2023

Many-body interaction on near-field radiative heat transfer between two nanoparticles caused by proximate particle ensembles

Near-field radiative heat transfer (NFRHT) has received growing attention because of its high intensity far beyond the Planck's black-body limit. Insertion of a third object in proximity of the two articles can significantly influence and manipulate its NFRHT. However, for the system composed of many particles, the effect of many-body interaction (MBI) on NFRHT between arbitrary two particles is still not well understood. In this work, the MBI is studied for two particles with three typical proximate ensembles: particle chain, plane and grating. With the increasing of proximate particle size, the MBI on NFRHT will experience a radical change from inhibition to enhancement. The polarizability of the proximate particle increases with particle radius, which enhances the interaction between the proximate particles and the main particle, and then results in enhancement of NFRHT between the main particles. When twisting the proximate particle ensemble, the proximate MBI accounts for a smooth and non-oscillated twisting angle dependence of NFRHT, different from the oscillation phenomenon of NFRHT for particle gratings. This work deepens the understanding of NFRHT in dense particulate systems.

physics.optics↗

arXiv2022

Residual surface charge mediated near-field radiative energy transfer: A topological insulator analog

We study the modifications of near-field radiative energy transfer (NFRET) caused by residual surface charges, which are common in micro- and nano-systems like NEMS/MEMS. The host object with the residual surface charges and the inherent bulk state can be treated as an analog of the real three-dimensional topological insulator, which is inherent of also both surface states and bulk states and is promising to modulate NFRET. Through constructing such a topological insulator analog, we aim to modulate NFRET concerning only common trivial materials. Besides the well-known resonant modes (surface polariton and localized surface polariton) supported by the bulk state, the residual surface charges give rise to an additional temperature-dependent mode providing a new heat flux channel. For low temperatures we find a giant surface-charge-induced enhancement of the NFRET due to a good match between the surface-charge-induced resonance and the Planck window. However, for relative high temperatures where the Fr\"{o}hlich resonance dominates the heat transfer rather the surface-charge-induced resonance, the residual charges result in a weakening of the NFRET. This work paves way for understanding and modulating the near-field radiative energy transfer for micro- and nano-systems.

physics.app-ph↗

arXiv2022

Photonic p-n Junction: An Ideal Near-Field Heat Flux Modulator

Using a pair of p- and n-type semiconductors separated by a nanoscale vacuum gap, we introduce an optoelectronics element prototype, "photonic p-n junction", as an analogue of the electronic p-n junction, which is demonstrated to serve as an ideal near-field heat flux modulator. The high modulation performance relies on the switch among three fundamental photon-carrier interaction modes (i.e., surface plasmon polaritons, symmetric depleted internal plasmon polaritons and symmetric accumulated surface plasmon polaritons) caused by the changes in hole and electron densities near the surfaces under a tunable bias. This prototype offers new thoughts not only for contactless thermal management at nanoscale but also for design of optoelectronics devices processing information carried by plasmon polaritons.

physics.optics↗

arXiv2021

Photothermal behavior for two-dimensional nanoparticle ensembles: Multiple scattering and thermal accumulation effects

Light-assisted micro-nanoscale temperature control in complex nanoparticle network attracts lots of research interests. Many efforts have been put on the optical properties of the nanoparticle networks and only a few investigations on its light-induced thermal behavior was reported. We consider two-dimensional (2D) square-lattice nanoparticle ensemble made of typical metal Ag with a radius of 5 nm. The effect of complex optical coupling and thermal accumulation on the light-induced thermal behavior in plasmonic resonance frequency (around 383 nm) is analyzed by means of the Green\textquotesingle s function approach. Regime borders of both optical coupling and thermal accumulation effects on the photothermal behavior of 2D square-lattice nanoparticle ensemble are figured out clearly and quantitatively. A dimensionless parameter $\varphi$ is defined as the ratio of full temperature increase to that without considering the optical coupling or thermal accumulation to quantify the optical coupling and thermal accumulation effects on photothermal behavior. The more compact the nanoparticle ensemble is, the stronger the optical coupling on thermal behavior is. When the lattice spacing increases to tens of nanoparticle radius, the optical coupling becomes insignificant. When $\varphi \approx 1$ (lattice spacing increases to hundreds of nanoparticle radius), the thermal accumulation effects are weak and can be neglected safely. The polarization-dependent distribution of temperature increase of nanoparticles is observed only in the compact nanoparticle ensemble, while for dilute ensemble, such polarization-dependent temperature increase distribution can not be observed anymore. This work may help for the understanding of the light-induced thermal transport in the 2D particle ensemble.

physics.optics↗

arXiv2021

Near-field radiation assisted smart skin for spacecraft thermal control

Thermal control is of critical importance for normal operation of spacecraft. Given thermal radiation is the only means of heat dissipation in space, an efficient thermal control approach for spacecraft is to coat the radiator with a tunable-emittance "skin" that can tune its heat dissipation according to various thermal conditions. The existing schemes solely relying on far-field thermal radiation, which are based on mechanical, electrochromic or thermochromic working principles, are difficult to combine the advantages of all-solid-state structure, actively and accurate tuning, and large tuning range of heat flux. In this work, we propose a near-field radiation assisted (NFRA) smart skin for thermal control which can tune the heat rejection accurately and in a large range. It contains a metal-insulator-semiconductor (MIS) structure, where the carrier distribution in the semiconductor layer can be electrically altered. In this way, the near-field heat flux, and ultimately the skin emission power expressed using effective emittance, can be controlled as a function of the applied voltage. The variation range of the effective emittance can exceed 0.7 when adjusting the applied voltage from -10 V to 100 V with our preliminary design. This work opens a new way of smart skin design for active spacecraft thermal control.

physics.app-ph↗

arXiv2020

Normal heat diffusion in many-body system via thermal photons

A normal-diffusion theory for heat transfer in many-body systems via carriers of thermal photons is developed. The thermal conductivity tensor is rigorously derived from fluctuational electrodynamics as a coefficient of diffusion term for the first time. In addition, a convection-like heat transfer behavior is revealed in systems of asymmetric distribution of particles, indicating violation of Fourier's law for such system. Considering the central role of thermal conductivity in heat transfer, this work paves a way for understanding, analysis and manipulation of heat transfer in nanoparticle system via thermal photons with many-body interactions.

physics.app-ph↗

arXiv2020

Many-body effective thermal conductivity in phase-change nanoparticle chains due to near-field radiative heat transfer

In dense systems composed of numerous nanoparticles, direct simulations of near-field radiative heat transfer (NFRHT) require considerable computational resources. NFRHT for the simple one-dimensional nanoparticle chains embedded in a non-absorbing host medium is investigated from the point of view of the continuum by means of an approach combining the many-body radiative heat transfer theory and the Fourier law. Effects of the phase change of the insulator-metal transition material (VO$_2$), the complex many-body interaction (MBI) and the host medium relative permittivity on the characteristic effective thermal conductivity (ETC) are analyzed. The ETC for VO$_2$ nanoparticle chains below the transition temperature can be as high as 50 times of that above the transition temperature due to the phase change effect. The strong coupling in the insulator-phase VO$_2$ nanoparticle chain accounts for its high ETC as compared to the low ETC for the chain at the metallic phase, where there is a mismatch between the characteristic thermal frequency and resonance frequency. The strong MBI is in favor of the ETC. For SiC nanoparticle chains, the MBI even can double the ETC as compared to those without considering the MBI effect. For the dense chains, a strong MBI enhances the ETC due to the strong inter-particles couplings. When the chains go more and more dilute, the MBI can be neglected safely due to negligible couplings. The host medium relative permittivity significantly affects the inter-particles couplings, which accounts for the permittivity-dependent ETC for the VO$_2$ nanoparticle chains.

cond-mat.mes-hall↗

arXiv2020

Near-field radiative heat transfer between twisted nanoparticle gratings

We study the near-field radiative heat transfer between two twisted finite-size polar dielectric nanoparticle gratings. Differently from previous studies of the same configuration, we do not rely on any approximated effective medium theory to describe the gratings. By the full many-body radiative heat transfer theory we are able to investigate how the size, distance and relative orientation between the gratings influence the radiative heat flux. By changing the twisting angle $θ$, we show a significant oscillation of the thermal conductance $G(θ)$, due to the size effect for gratings of both square and circular shapes. The distance- and twisting-dependent coupling between the gratings accounts for a strong and characteristic modulation of radiative thermal conductance with implications for the energy management, sensing, and NEMS/MEMS devices.

cond-mat.mes-hall↗