Papers

In this paper, a predictive constraint-aware control scheme is formulated within the Command Governor (CG) framework for localized hydration management of a proton exchange membrane (PEM) fuel cell system. First, a comprehensive nonlinear dynamic model of the fuel cell system is presented which includes a pseudo 2-dimensional (P2D) model of the stack, reactant supply and cooling subsystems. The model captures the couplings among the various subsystems and serves as the basis for designing output feedback controllers to track the optimal set-points of the air supply and cooling systems for power optimization. The closed-loop nonlinear model is then used to analyze the dynamic behavior of membrane hydration near the anode inlet, the driest region of the membrane in a counter-flow configuration, under various operating conditions. A reduced-order linearized model is then derived to approximate hydration behavior with sufficient fidelity for constraint enforcement. This model is used within the CG framework to adjust the air supply set-points when necessary to prevent membrane dry-out. The effectiveness of the proposed approach in maintaining local membrane hydration while closely tracking the requested net power is demonstrated through realistic drive-cycle simulations.

The accelerating digitalization and decentralization of modern power systems expose critical communication infrastructures to escalating cyber risks, particularly under emerging quantum computing threats. This paper presents an integrated techno-economic framework to evaluate the feasibility of Quantum Key Distribution (QKD) for secure power-system communications. A stochastic system model is developed to jointly capture time-varying key demand, QKD supply under optical-loss constraints, station-side buffering, and post-quantum cryptography (PQC) fallback mechanisms. Analytical conditions are derived for service-level assurance, including buffer stability, outage probability, and availability bounds. Building on this, two quantitative metrics, including the Levelized Cost of Security (LCoSec) and Cost of Incremental Security (CIS), are formulated to unify capital, operational, and risk-related expenditures within a discounted net-present-value framework. Using IEEE 118-bus, 123-node, and 39-bus test systems, we conduct discrete-event simulations comparing PQC-only, QKD-only, and Hybrid architectures across multiple topologies and service profiles. Results show that Hybrid architectures dominated by QKD significantly reduce key-outage probability and SLA shortfalls, achieving near-unit availability for real-time and confidentiality-critical services. Economic analyses reveal clear breakeven zones where QKD-enhanced deployments become cost-effective, primarily in metropolitan and distribution-level networks under moderate optical loss and buffer sizing. The proposed framework provides a reproducible, risk-aware decision tool for guiding large-scale, economically justified QKD adoption in future resilient power-system infrastructures.

The proliferation of distributed energy resources (DERs) and demand-side flexibility has made virtual power plants (VPPs) central to modern grid operation. Yet their end-to-end business pipeline, covering bidding, dispatch, metering, settlement, and archival, forms a tightly coupled cyber-physical-economic system where secure and timely communication is critical. Under the combined stress of sophisticated cyberattacks and extreme weather shocks, conventional cryptography offers limited long-term protection. Quantum key distribution (QKD), with information-theoretic guarantees, is viewed as a gold standard for securing critical infrastructures. However, limited key generation rates, routing capacity, and system overhead render key allocation a pressing challenge: scarce quantum keys must be scheduled across heterogeneous processes to minimize residual risk while maintaining latency guarantees. This paper introduces a quantum-authenticated aggregation and settlement framework for VPPs. We first develop a system-threat model that connects QKD key generation and routing with business-layer security strategies, authentication strength, refresh frequency, and delay constraints. Building on this, we formulate a key-budgeted risk minimization problem that jointly accounts for economic risk, service-level violations, and key-budget feasibility, and reveal a threshold property linking marginal security value to shadow prices. Case studies on a representative VPP system demonstrate that the proposed approach significantly reduces residual risk and SLA violations, enhances key efficiency and robustness, and aligns observed dynamics with the theoretical shadow price mechanism.

Large Audio-Language Models (LALMs) are often constrained by short audio context windows, even when their text backbones support long contexts, limiting long-form audio understanding. Prior work has introduced context-extension methods (e.g. YaRN) on unimodal LLMs, yet their application to LALMs remains unexplored. First, building on RoPE-based context extension, we introduce Partial YaRN, a training-free, modality-decoupled extension method that modifies only audio token positions, leaving text positions intact to preserve the base LLM's text capabilities. Second, we propose Virtual Longform Audio Training (VLAT), a training strategy that extends Partial YaRN into a training-time positional augmentation. VLAT simulates diverse audio lengths during training, enabling generalization to inputs far longer than those seen in training. Our experiments on SALMONN and Qwen2-Audio confirm that Partial YaRN outperforms the original models across wide range of settings, and VLAT provides substantial performance improvement on long audio of unseen lengths.

This paper presents LongCat-Audio-Codec, an audio tokenizer and detokenizer solution designed for industrial grade end-to-end speech large language models. By leveraging a decoupled model architecture and a multistage training strategy, LongCat-Audio-Codec exhibits robust semantic modeling capabilities, flexible acoustic feature extraction capabilities, and low-latency streaming synthesis capabilities. It encodes speech at an ultra-low frame rate of 16.67 Hz, with a minimum bitrate of 0.43 kbps and a maximum bitrate of 0.87 kbps. Evaluation results demonstrate that LongCat-Audio-Codec achieves strong speech intelligibility and is capable of synthesizing highquality speech at low bitrate, thus effectively balancing coding efficiency and decoding quality. The inference code and model checkpoints of LongCat-Audio-Codec are available at: https://github.com/meituan-longcat/LongCat-Audio-Codec.

Accurate interpretation of 12 lead electrocardiograms (ECGs) is critical for early detection of cardiac abnormalities, yet manual reading is error prone and existing CNN based classifiers struggle to choose receptive field sizes that generalize to the long sequences typical of ECGs. Omni Scale CNN (OS CNN) addresses this by enumerating prime sized kernels inspired by Goldbach conjecture to cover every scale, but its exhaustive design explodes computational cost and blocks deeper, wider models. We present Efficient Convolutional Omni Scale Network (EcoScale-Net), a hierarchical variant that retains full receptive field coverage while eliminating redundancy. At each stage, the maximum kernel length is capped to the scale still required after down sampling, and bottleneck convolutions inserted before and after every Omni Scale block curtail channel growth and fuse multi scale features. On the large scale CODE 15% ECG dataset, EcoScaleNet reduces parameters by 90% and FLOPs by 99% compared with OS CNN, while raising macro averaged F1 score by 2.4%. These results demonstrate that EcoScaleNet delivers SOTA accuracy for long sequence ECG classification at a fraction of the computational cost, enabling real time deployment on commodity hardware. Our EcoScaleNet code is available in GitHub Link.

Artificial intelligence and machine learning are increasingly used for forecasting, optimization, and policy design in the energy sector, yet no standardized framework exists to evaluate whether these systems reason correctly. Current validation practices focus on predictive accuracy or computational efficiency, leaving the logical integrity of analytical conclusions untested. This study introduces the Analytical Reliability Benchmark (ARB), a reproducible framework that quantifies reasoning reliability in large language models applied to energy system analysis. The benchmark integrates five submetrics: accuracy, reasoning reliability, uncertainty discipline, policy consistency, and transparency, and evaluates model performance across deterministic, probabilistic, and epistemic scenarios using open technoeconomic datasets (NREL ATB 2024, DOE H2A/H2New, IEA WEO 2024). Four frontier models (GPT-4/5, Claude 4.5 Sonnet, Gemini 2.5 Pro, Llama 3 70B) were tested under identical factual and regulatory conditions. Results show that reasoning reliability can be objectively measured. GPT-4/5 and Claude 4.5 Sonnet achieved consistent and policy-compliant reasoning (Analytical Reliability Index greater than 90), Gemini 2.5 Pro demonstrated moderate stability, and Llama 3 70B remained below professional thresholds. Statistical validation confirmed that these differences are significant and reproducible. The ARB establishes the first quantitative method in the energy literature for verifying causal, probabilistic, and policy-driven reasoning in artificial intelligence systems, providing a reference framework for trustworthy and transparent analytical applications in the global energy transition.

Massive number of simultaneous Internet of Things (IoT) uplinks strain gateways with interference and energy limits, yet devices often lack neighbors' Channel State Information (CSI) and cannot sustain centralized Mobile Edge Computing (MEC) or heavy Machine Learning (ML) coordination. Classical Bayesian solvers help with uncertainty but become intractable as users and strategies grow, making lightweight, distributed control essential. In this paper, we introduce the first-ever, novel epistemic Bayesian game for uplink power control under incomplete CSI that operates while suppressing interference among multiple uplink channels from distributed IoT devices firing at the same time. Nodes run inter-/intra-epistemic belief updates over opponents' strategies, replacing exhaustive expected-utility tables with conditional belief hierarchies. Using an exponential-Gamma SINR model and higher-order utility moments (variance, skewness, kurtosis), the scheme remains computationally lean with a single-round upper bound of $O\!\left(N^{2} S^{2N}\right)$. Precise power control and stronger coverage amid realistic interference: with channel magnitude equal to $1$ and a signal-to-interference-plus-noise ratio (SINR) threshold of $-18$ dB, coverage reaches approximately $60\%$ at approximately $55\%$ of the maximum transmit power; mid-rate devices with a threshold of $-27$ dB achieve full coverage with less than $0.1\%$ of the maximum transmit power.Under $80\%$ interference, a fourth-moment policy cuts average power from approximately $52\%$ to approximately $20\%$ of the maximum transmit power with comparable outage, outperforming expectation-only baselines. These results highlight a principled, computationally lean path to optimal power allocation and higher network coverage under real-world uncertainty within dense, distributed IoT networks.

The next-generation radio-interferometric (RI) telescopes require imaging algorithms capable of forming high-resolution high-dynamic-range images from large data volumes spanning wide frequency bands. Recently, AIRI, a plug-and-play (PnP) approach taking the forward-backward algorithmic structure (FB), has demonstrated state-of-the-art performance in monochromatic RI imaging by alternating a data-fidelity step with a regularization step via learned denoisers. In this work, we introduce HyperAIRI, its hyperspectral extension, underpinned by learned hyperspectral denoisers enforcing a power-law spectral model. For each spectral channel, the HyperAIRI denoiser takes as input its current image estimate, alongside estimates of its two immediate neighboring channels and the spectral index map, and provides as output its associated denoised image. To ensure convergence of HyperAIRI, the denoisers are trained with a Jacobian regularization enforcing non-expansiveness. To accommodate varying dynamic ranges, we assemble a shelf of pre-trained denoisers, each tailored to a specific dynamic range. At each HyperAIRI iteration, the spectral channels of the target image cube are updated in parallel using dynamic-range-matched denoisers from the pre-trained shelf. The denoisers are also endowed with a spatial image faceting functionality, enabling scalability to varied image sizes. Additionally, we formally introduce Hyper-uSARA, a variant of the optimization-based algorithm HyperSARA, promoting joint sparsity across spectral channels via the $\ell_{2,1}$-norm, also adopting FB. We evaluate HyperAIRI's performance on simulated and real observations. We showcase its superior performance compared to its optimization-based counterpart Hyper-uSARA, CLEAN's hyperspectral variant in WSClean, and the monochromatic imaging algorithms AIRI and uSARA.

Zak-OTFS provides a framework for integrated sensing & communication (ISAC) in high delay and Doppler spread environments. Pulse shaping filter design enables joint optimization of sensing and communication performance. For sensing, a localized pulse shaping filter enables input-output (I/O) relation estimates close to the physical scattering channel. For communication, orthogonality of the pulse shape on the information lattice prevents inter-symbol interference, and no time and bandwidth expansion enables full spectral efficiency. A filter simultaneously meeting all three objectives is ideal for ISAC. Existing filter designs achieve two, but not all three objectives. In this work, we design pulse shaping filters meeting all three objectives via the Isotropic Orthogonal Transform Algorithm. The proposed filters have improved spectral efficiency, data detection and sensing performance over existing filter choices.

As connected and autonomous vehicles become more widespread, platooning has emerged as a key strategy to improve road capacity, reduce fuel consumption, and enhance traffic flow. However, the benefits of platoons strongly depend on their ability to maintain stability. Instability can lead to unsafe spacing and increased energy usage. In this work, we study platoon instability and analyze the root cause of its occurrence, as well as its impacts on the following vehicle. To achieve this, we propose a comparative study between different car-following models such as the Intelligent Driver Model (IDM), the Optimal Velocity Model (OVM), the General Motors Model (GMM), and the Cooperative Adaptive Cruise Control (CACC). In our approach, we introduce a disruption in the model by varying the velocity of the leading vehicle to visualize the behavior of the following vehicles. To evaluate the dynamic response of each model, we introduce controlled perturbations in the velocity of the leading vehicle, specifically, sinusoidal oscillations and discrete velocity changes. The resulting vehicle trajectories and variations in inter-vehicle spacing are analyzed to assess the robustness of each model to disturbance propagation. The findings offer insight into model sensitivity, stability characteristics, and implications for designing resilient platooning control strategies.

The Koopman operator, originally defined for dynamical systems without input, has inspired many applications in control. Yet, the theoretical foundations underpinning this progress in control remain underdeveloped. This paper investigates the theoretical structure and connections between two extensions of Koopman theory to control: (i) Koopman operator via infinite input sequences and (ii) the Koopman control family. Although these frameworks encode system information in fundamentally different ways, we show that under certain conditions on the function spaces they operate on, they are equivalent. The equivalence is both in terms of the actions of the Koopman-based formulations in each framework as well as the function values on the system trajectories. Our analysis provides constructive tools to translate between the frameworks, offering a unified perspective for Koopman methods in control.

Prompt caching is critical for reducing latency and cost in LLM inference: OpenAI and Anthropic report up to 50-90% cost savings through prompt reuse. Despite its widespread success, little is known about what constitutes an optimal prompt caching policy, particularly when optimizing tail latency, a metric of central importance to practitioners. The widely used Least Recently Used (LRU) policy can perform arbitrarily poor on this metric, as it is oblivious to the heterogeneity of conversation lengths. To address this gap, we propose Tail-Optimized LRU, a simple two-line modification that reallocates KV cache capacity to prioritize high-latency conversations by evicting cache entries that are unlikely to affect future turns. Though the implementation is simple, we prove its optimality under a natural stochastic model of conversation dynamics, providing the first theoretical justification for LRU in this setting, a result that may be of independent interest to the caching community. Experimentally, on real conversation data WildChat, Tail-Optimized LRU achieves up to 27.5% reduction in P90 tail Time to First Token latency and 23.9% in P95 tail latency compared to LRU, along with up to 38.9% decrease in SLO violations of 200ms. We believe this provides a practical and theoretically grounded option for practitioners seeking to optimize tail latency in real-world LLM deployments.

Advances in leveraging Gaussian processes (GP) have enabled learning and inferring dynamic grid behavior from scarce PMU measurements. However, real measurements can be corrupted by various random and targeted threats, leading to inaccurate and meaningless results. This paper develops robust transient learning to overcome this challenge by exploiting the sparse corruption patterns in the data flow. Specifically, we integrate sparse optimization with method of moments (MoM) to make learning robust to a sparse distribution of data corruptions; then, we optimize sparse weights to identify corrupted meter locations. To improve inference speed on large-scale systems, we further adopt K-medoid clustering of locations to develop dimension reduction (DR) and aggregate representation (AR) heuristics. Experimental results demonstrate robustness against random large errors, targeted false data injections, and local PMU clock drifts. On a 1354-bus system, inference turns out to be 18x faster using DR and 400x faster when further combined with AR heuristics.

In emerging networked systems, mobile edge devices such as ground vehicles and unmanned aerial system (UAS) swarms collectively aggregate vast amounts of data to make machine learning decisions such as threat detection in remote, dynamic, and infrastructure-constrained environments where power and bandwidth are scarce. Federated learning (FL) addresses these constraints and privacy concerns by enabling nodes to share local model weights for deep neural networks instead of raw data, facilitating more reliable decision-making than individual learning. However, conventional FL relies on a central server to coordinate model updates in each learning round, which imposes significant computational burdens on the central node and may not be feasible due to the connectivity constraints. By eliminating dependence on a central server, distributed federated learning (DFL) offers scalability, resilience to node failures, learning robustness, and more effective defense strategies. Despite these advantages, DFL remains vulnerable to increasingly advanced and stealthy cyberattacks. In this paper, we design sophisticated targeted training data poisoning and backdoor (Trojan) attacks, and characterize the emerging vulnerabilities in a vehicular network. We analyze how DFL provides resilience against such attacks compared to individual learning and present effective defense mechanisms to further strengthen DFL against the emerging cyber threats.

We introduce the PenduMAV, an exactly actuated (6-input) omnidirectional multirotor that structurally eliminates internal forces at equilibria. The vehicle features one actively-tilting propeller and three propellers mounted on passive pendulum links via universal joints. This architecture achieves full 6D wrench generation while avoiding the structural and energetic costs of input redundancy and internal forces. After deriving the full multibody dynamics, we demonstrate that a forced equilibrium exists for every main platform pose. To asymptotically stabilize the closed-loop system, we design a coordinate-invariant nonlinear controller based on dynamic feedback linearization and backstepping, utilizing the left-trivialized error on SE(3). System stability is formally guaranteed through Lyapunov analysis of the zero dynamics. Finally, Gazebo simulations validate the approach, showcasing fully decoupled attitude and translational tracking under parametric uncertainty and actuator noise.

This work presents a novel structured family of Grassmannian constellations for multiple-input multiple-output (MIMO) noncoherent communications over Rayleigh block-fading channels, where neither the transmitter nor the receiver has channel state information (CSI). The proposed constellation design is built upon the geodesic curves of the Grassmann manifold, thereby exploiting its underlying geometric structure. The resulting solution is limited in spectral efficiency (with a maximum constellation size of $4M^2$ points, where $M$ is the number of transmit antennas), targeting a rate in the range of $0.25$-$1$ bps/Hz. However, all space-time matrices resulting from this design exhibit the remarkable property of having a single nonzero entry per row, meaning that only one transmit antenna is active per time slot. This property significantly reduces hardware complexity and implementation cost, while also lowering power consumption, as only a single radio-frequency (RF) chain and power amplifier are required for transmission. Furthermore, within the constellation size limits, the proposed design achieves error performance comparable to state-of-the-art optimization-based unstructured designs, as validated through symbol error rate (SER) numerical results. It also enables simple yet effective bit labeling, confirmed by comparisons of bit error rate (BER) and SER, and reduces the computational complexity of the maximum-likelihood (ML) detector for Grassmannian constellations by a factor of $M$.

The rapid decentralization and digitalization of local electricity markets have introduced new cyber-physical vulnerabilities, including key leakage, data tampering, and identity spoofing. Existing blockchain-based solutions provide transparency and traceability but still depend on classical cryptographic primitives that are vulnerable to quantum attacks. To address these challenges, this paper proposes Q-EnergyDEX, a zero-trust distributed energy trading framework driven by quantum key distribution and blockchain. The framework integrates physical-layer quantum randomness with market-level operations, providing an end-to-end quantum-secured infrastructure. A cloud-based Quantum Key Management Service continuously generates verifiable entropy and regulates key generation through a rate-adaptive algorithm to sustain high-quality randomness. A symmetric authentication protocol (Q-SAH) establishes secure and low-latency sessions, while the quantum-aided consensus mechanism (PoR-Lite) achieves probabilistic ledger finality within a few seconds. Furthermore, a Stackelberg-constrained bilateral auction couples market clearing with entropy availability, ensuring both economic efficiency and cryptographic security. Simulation results show that Q-EnergyDEX maintains robust key stability and near-optimal social welfare, demonstrating its feasibility for large-scale decentralized energy markets.

To tackle long-horizon tasks, recent hierarchical vision-language-action (VLAs) frameworks employ vision-language model (VLM)-based planners to decompose complex manipulation tasks into simpler sub-tasks that low-level visuomotor policies can easily handle. Typically, the VLM planner is finetuned to learn to decompose a target task. This finetuning requires target task demonstrations segmented into sub-tasks by either human annotation or heuristic rules. However, the heuristic subtasks can deviate significantly from the training data of the visuomotor policy, which degrades task performance. To address these issues, we propose a Retrieval-based Demonstration Decomposer (RDD) that automatically decomposes demonstrations into sub-tasks by aligning the visual features of the decomposed sub-task intervals with those from the training data of the low-level visuomotor policies. Our method outperforms the state-of-the-art sub-task decomposer on both simulation and real-world tasks, demonstrating robustness across diverse settings. Code and more results are available at rdd-neurips.github.io.

Reinforcement learning (RL), while powerful and expressive, can often prioritize performance at the expense of safety. Yet safety violations can lead to catastrophic outcomes in real-world deployments. Control Barrier Functions (CBFs) offer a principled method to enforce dynamic safety – traditionally deployed online via safety filters. While the result is safe behavior, the fact that the RL policy does not have knowledge of the CBF can lead to conservative behaviors. This paper proposes CBF-RL, a framework for generating safe behaviors with RL by enforcing CBFs in training. CBF-RL has two key attributes: (1) minimally modifying a nominal RL policy to encode safety constraints via a CBF term, (2) and safety filtering of the policy rollouts in training. Theoretically, we prove that continuous-time safety filters can be deployed via closed-form expressions on discrete-time roll-outs. Practically, we demonstrate that CBF-RL internalizes the safety constraints in the learned policy – both enforcing safer actions and biasing towards safer rewards – enabling safe deployment without the need for an online safety filter. We validate our framework through ablation studies on navigation tasks and on the Unitree G1 humanoid robot, where CBF-RL enables safer exploration, faster convergence, and robust performance under uncertainty, enabling the humanoid robot to avoid obstacles and climb stairs safely in real-world settings without a runtime safety filter.

Robust humanoid locomotion in unstructured environments requires architectures that balance fast low-level stabilization with slower perceptual decision-making. We show that a simple layered control architecture (LCA), a proprioceptive stabilizer running at high rate, coupled with a compact low-rate perceptual policy, enables substantially more robust performance than monolithic end-to-end designs, even when using minimal perception encoders. Through a two-stage training curriculum (blind stabilizer pretraining followed by perceptual fine-tuning), we demonstrate that layered policies consistently outperform one-stage alternatives in both simulation and hardware. On a Unitree G1 humanoid, our approach succeeds across stair and ledge tasks where one-stage perceptual policies fail. These results highlight that architectural separation of timescales, rather than network scale or complexity, is the key enabler for robust perception-conditioned locomotion.

Deployment of efficient and accurate Deep Learning models has long been a challenge in autonomous navigation, particularly for real-time applications on resource-constrained edge devices. Edge devices are limited in computing power and memory, making model efficiency and compression essential. In this work, we propose EdgeNavMamba, a reinforcement learning-based framework for goal-directed navigation using an efficient Mamba object detection model. To train and evaluate the detector, we introduce a custom shape detection dataset collected in diverse indoor settings, reflecting visual cues common in real-world navigation. The object detector serves as a pre-processing module, extracting bounding boxes (BBOX) from visual input, which are then passed to an RL policy to control goal-oriented navigation. Experimental results show that the student model achieved a reduction of 67% in size, and up to 73% in energy per inference on edge devices of NVIDIA Jetson Orin Nano and Raspberry Pi 5, while keeping the same performance as the teacher model. EdgeNavMamba also maintains high detection accuracy in MiniWorld and IsaacLab simulators while reducing parameters by 31% compared to the baseline. In the MiniWorld simulator, the navigation policy achieves over 90% success across environments of varying complexity.

Inter symbol interference (ISI), which occurs in a wide variety of channels, is a result of time dispersion. It can be mitigated by equalization, which results in noise coloring. Inspired by the development of Approximate Independence in statistical physics, for such colored noise we propose a decoder called Ordered Reliability Bits Guessing Random Additive Noise Decoding (ORBGRAND-AI) that operates without the need for turbo equalization or interleaving. By foregoing interleaving, ORBGRAND-AI can deliver the same, or lower, block error rate (BLER) for the same amount of energy per information bit in an ISI channel as a state-of-the-art soft input decoder, such as Cyclic Redundancy Check Assisted-Successive Cancellation List (CA-SCL) decoding, with an interleaver. To assess the decoding performance of ORBGRAND-AI, we consider delay tap models and their associated colored noise. In particular, we examine a two-tap dicode ISI channel as well as an ISI channel derived from data from RFView, a physics-informed modeling and simulation tool. We investigate the dicode and RFView channel under a variety of imperfect channel state information assumptions and show that a second order autoregressive model adequately represents the RFView channel effect.

In this paper, we address the stabilization problem for force-controlled nonholonomic mobile robots under safety-critical constraints. We propose a continuous, time-invariant control law based on the gamma m-quadratic programming (gamma m-QP) framework, which unifies control Lyapunov functions (CLFs) and control barrier functions (CBFs) to enforce both stability and safety in the closed-loop system. For the first time, we construct a global, time-invariant, strict Lyapunov function for the closed-loop nonholonomic mobile robot full-dynamic system with a nominal stabilization controller in polar coordinates; this strict Lyapunov function then serves as the CLF in the QP design. Next, by exploiting the inherent cascaded structure of the vehicle dynamics, we develop a CBF for the mobile robot via an integrator backstepping procedure. Our main results guarantee both asymptotic stability and safety for the closed-loop system. Both the simulation and experimental results are presented to illustrate the effectiveness and performance of our approach.

Depression is a widespread mental health disorder, yet its automatic detection remains challenging. Prior work has explored unimodal and multimodal approaches, with multimodal systems showing promise by leveraging complementary signals. However, existing studies are limited in scope, lack systematic comparisons of features, and suffer from inconsistent evaluation protocols. We address these gaps by systematically exploring feature representations and modelling strategies across EEG, together with speech and text. We evaluate handcrafted features versus pre-trained embeddings, assess the effectiveness of different neural encoders, compare unimodal, bimodal, and trimodal configurations, and analyse fusion strategies with attention to the role of EEG. Consistent subject-independent splits are applied to ensure robust, reproducible benchmarking. Our results show that (i) the combination of EEG, speech and text modalities enhances multimodal detection, (ii) pretrained embeddings outperform handcrafted features, and (iii) carefully designed trimodal models achieve state-of-the-art performance. Our work lays the groundwork for future research in multimodal depression detection.

Near-field blockage changes the beam-design objective in millimeter-wave links: maximizing the unblocked on-axis gain does not necessarily maximize blocked-link performance. This paper studies when phase-only, aperture-constrained non-diffracting (ND) beams provide a blocked-link advantage over equal-aperture, equal-power conventional reference beams. We develop a unified annular-spectrum framework that generates isotropic Bessel-like and anisotropic Mathieu-like beams under discrete phased-array constraints, and a geometry-aware analysis centered on three propagation landmarks: the peak-intensity distance, the crossover distance, and an effective post-blockage recovery distance. Their relationship yields a recovery-before-crossover condition linking blockage size, depth, cone angle, and usable ND range, and motivates a blocked-link gain ratio that maps directly onto an achievable-rate gap at every operating SNR. The analysis also explains why anisotropic Mathieu-like beams can outperform isotropic ones under direction-dependent blockage. Monte Carlo simulations verify the predicted advantage regimes, an auxiliary comparison against a near-field focusing baseline confirms that the advantage persists against an unblocked-optimal array, and sensitivity studies over cone-angle choice and partial-transmission blockers show that the opaque-screen picture is a conservative reading of the underlying physics. The results identify Bessel-like and Mathieu-like beams as practical candidates for blockage-resilient near-field communications.

Magnetic induction (MI) communication (MIC) has emerged as a promising candidate for underground communication networks due to its excellent penetration capabilities. Integration with Space-Air-Ground-Underground (SAGUI) networks in next-generation mobile communication systems requires a well-defined network architecture. A recent discovery in MIC research, MI fast fading, remains in its early stages and presents unique challenges. This paper provides a comprehensive survey on through-the-earth (TTE) MIC, covering MI applications, channel modeling, point-to-point MIC design, relay techniques, network frameworks, and emerging technologies. We compare various MIC applications to highlight TTE-specific challenges and review the principles of channel modeling, addressing both MI slow fading and MI fast fading, along with its potential impact on existing MIC theories. We conduct a fine-grained decomposition of MI channel power gain into four distinct physical parameters, and propose a novel geometric model to analyze MI fast fading. We also summarize MI relay techniques, examine crosstalk effects in relay and high-density networks, and explore key research tasks within the OSI framework for a holistic MI network protocol in SAGUI. To bridge the gaps identified, we propose a MIC framework that supports TCP/IP and Linux, enabling full implementation of existing and emerging MIC solutions. This framework empowers researchers to leverage Linux resources and deep learning platforms for accelerated development of MIC in SAGUI networks. Remaining research challenges, open issues, and promising novel techniques are further identified to advance MIC research.

To address growing cybersecurity challenges in modern power dispatch systems, this paper proposes a multi-layer modeling and optimization framework for SCADA systems enhanced with quantum key distribution (QKD). While most existing applications of QKD in the power sector focus on building secure point-to-point communication tunnels, they rarely consider the system-level coupling between key dynamics and control scheduling. In contrast, our approach integrates quantum key generation, consumption, inventory prediction, and control latency into a unified model, enabling key-aware reconfiguration of SCADA control chains based on task security demands and real-time resource constraints. To resolve conflicts in key resource allocation between transmission system operators (TSOs) and distribution system operators (DSOs), we formulate a bi-level Stackelberg game and transform it into a mathematical program with complementarity constraints (MPCC). We further develop an efficient Level Decomposition-Complementarity Pruning (LD-CP) algorithm to solve the problem. To support reproducible evaluation, we build an end-to-end co-simulation platform that integrates physical-layer disruptions via OpenQKD-Sim, Q3P/IEC-104 protocol stack binding, and real-time control-chain monitoring through Grafana. Experimental results on the IEEE 39- and 118-bus systems show that our method increases task success rate by 25%, reduces peak frequency deviation by 70%, and improves key utilization to 83%. This work lays the foundation for future quantum-secure control systems in power grid operations.

Integration of distributed energy resources has created a need for autonomous, dynamic voltage regulation. Decentralized Volt-VAr Control (VVC) of grid-connected inverters presents a unique opportunity for voltage management but, if designed poorly, can lead to unstable behavior when in feedback with the grid. We model the grid-VVC closed-loop dynamics with a linearized power flow approach, leveraging historical data, which shows improvement over the commonly used LinDistFlow model. This model is used to design VVC slopes by minimizing steady-state voltage deviation from the nominal value, subject to a non-convex spectral radius stability constraint, which has not been previously implemented within this context. We compare this constraint to existing convex restrictions and demonstrate, through simulations on a realistic feeder, that using the spectral radius results in more effective voltage regulation.

Complete awareness of the wireless environment, crucial for future intelligent networks, requires sensing all transmitted signals, not just the strongest. A fundamental barrier is estimating the target signal when it is buried under strong co-channel interference from other transmitters, a failure of which renders the signal unusable. This work proposes a maximum likelihood (ML)-based cross-preamble estimation framework that exploits carrier frequency offset (CFO) constancy across beam-swept synchronization signals (SS), coherently aggregating information across multiple observations to reinforce the desired signal against overwhelming interference. Cramer-Rao lower bound (CRLB) analysis and simulation demonstrate reliable estimation even when the signal is over a thousand times weaker than the interference. A low-altitude radio-map case study further verifies the framework's practical effectiveness.