arXiv subjects

Hailong Huang

Publications and source records attributed to Hailong Huang.

At least 19 recordsLinked to original sources

arXiv2026

Variance-Guided Spatial Attention Fusion for Robust End-to-End Driving under Asymmetric Sensor Degradation

End-to-end multimodal driving has progressed rapidly by fusing camera and LiDAR streams. Existing pipelines remain fragile under asymmetric sensor degradation, where either an entire modality or only a localized region is corrupted while other regions remain useful. The key difficulty is not simply to add an uncertainty head, but to obtain dense reliability supervision, calibrate this reliability against physical fault severity, and use it before unreliable features bias the planner. We propose Variance-Guided Spatial Attention Fusion (VG-SAF), in which dense heteroscedastic reliability estimates act as interpretable spatial gates. The framework couples three components. First, a physically grounded augmentor simulates representative camera and LiDAR failures and emits a continuous spatial mask, providing dense supervision without additional annotation. Second, modality-specific experts predict per-pixel reliability scales through cross-branch dense distillation in log space, enforcing a monotone severity-to-scale response. Third, calibrated reliability maps drive a hybrid attention mechanism that suppresses unreliable cells with a local spatial gate and arbitrates between modalities through a cross-modal trust softmax. A Laplace uncertainty head emits a systemic waypoint uncertainty scale that signals severe or combined sensor degradation, including severities outside the training ranges. On the CARLA Longest6 benchmark, VG-SAF consistently improves closed-loop robustness over the baselines across camera-only, LiDAR-only, and joint degradation regimes, as measured by driving score, route completion, and infraction score.

cs.CV↗

arXiv2026

Online Monitoring and Risk Assessment of Non-Cooperative UAVs via STL-Aware Adaptive Fusion Kalman Filtering

This paper considers the problem of online state estimation and predictive risk assessment for non-cooperative unmanned aerial vehicles (UAVs) in the presence of asynchronous heterogeneous sensing and uncertain motion modes. To address this problem, a unified estimation and safety-assessment framework is developed by integrating an interacting multiple-model multi-rate Kalman filter with signal temporal logic (STL). The proposed framework enables simultaneous low-level state tracking and high-level safety reasoning within a common recursive architecture. Its main contribution is an STL-aware time-varying mode transition mechanism that updates model probabilities online using robustness measures induced by formal safety specifications. By embedding safety semantics directly into the mode inference and estimation process, the method improves responsiveness to maneuver variations, sensing asynchrony, and evolving threat patterns. Based on the estimated state distributions, the framework further generates multi-step state predictions and probabilistic reachable sets, which are used for finite-horizon safety evaluation and risk-triggered warning generation. Consequently, the proposed method provides not only estimates of the current target state, but also early indication of unsafe behaviors before they become fully observable. Finally, experimental results obtained from a real-time UAV monitoring platform show that the proposed approach improves estimation accuracy and produces earlier and more informative safety warnings, demonstrating its effectiveness for real-time UAV surveillance and safety monitoring applications.

eess.SY↗

arXiv2026

Do We Really Need Immediate Resets? Rethinking Collision Handling for Efficient Robot Navigation

Should a single collision necessarily terminate an entire navigation episode? In most deep reinforcement learning (DRL) frameworks for robot navigation, this remains the standard practice: every collision immediately triggers a global environment reset and is penalized as a complete task failure. While a collision during deployment naturally indicates task failure, applying the same treatment during training prevents the agent from exploring challenging obstacle configurations, which slows learning progress in the early training phase. In this work, we challenge this convention and propose a Multi-Collision reset Budget (MCB) framework that decouples local collision termination from global environment resets, allowing the agent to retry difficult configurations within the same episode. Simulation experiments show that MCB improves early-stage learning efficiency by reaching target success-rate levels with fewer interactions, with small collision budgets producing the most consistent gains. Real-world experiments on heterogeneous robot platforms further validate the deployability of the learned policies in cluttered environments.

cs.RO↗

arXiv2026

SEDualVLN: A Spatially-Enhanced Dual-System for Vision-Language Navigation

Vision-Language Navigation (VLN) approaches have currently followed two primary paradigms: the end-to-end Vision-Language Model (VLM) policy fine-tuned on navigation trajectories to directly predict actions, and the zero-shot modular pipeline integrating pre-trained Multimodal Large Language Model (MLLM) for training-free generalization to unseen environments. However, end-to-end methods struggle with long-horizon navigation and lack dynamic reasoning, whereas zero-shot methods are constrained by limited spatial grounding for reliable planning and also require substantial reasoning time. To bridge this gap, we introduce SEDualVLN, a spatially-enhanced dual-system VLN framework. System 1 is a VLM model enhanced with both global and local spatial awareness, used for action generation. System 2 integrates a general MLLM with a mapping module, wherein the MLLM plans waypoints by leveraging top-down views of the real-time 3D map alongside streams of rendered path images. Both systems leverage different forms of spatial enhancement to cultivate the agent's sense of direction in VLN tasks. Ultimately, they cooperate to complete the navigation task through a fast-slow coordinated approach. SEDualVLN achieves state-of-the-art performance on VLN-CE benchmarks, and further ablation studies demonstrate the effectiveness of each system and module.

cs.RO↗

arXiv2026

LookWise: Knowing When and Where to Look for Fine-Grained Visual Reasoning in Multimodal Large Language Models

Multimodal Large Language Models (MLLMs) are shifting towards "Thinking with Images" by actively exploring image details. While effective, large-scale training is computationally expensive, which has spurred growing interest in lightweight, training-free solutions. However, existing training-free methods suffer from two flaws: perceptual redundancy from indiscriminate cropping, which increases computational cost and introduces noise; and a drift between semantic intent and spatial attention, which prevents accurate localization of user-focused regions. To address these challenges, we propose LookWise, a framework for adaptive visual reasoning. LookWise follows a two-stage pipeline: a confidence-based module decides when to look more carefully, and a semantic-guided localization module determines where to look. This design enables MLLMs to adaptively acquire fine-grained visual evidence without additional training. Experiments on fine-grained and high-resolution visual reasoning benchmarks show that LookWise consistently improves accuracy over strong baselines while achieving an approximately $4.0\times$ inference speedup over the search-based method ZoomEye, demonstrating robust cross-model generalization.

cs.CV↗

arXiv2026

A Fast Semidefinite Convex Relaxation for Optimal Control Problems With Spatio-Temporal Constraints

Solving optimal control problems (OCPs) of autonomous agents operating under spatial and temporal constraints fast and accurately is essential in applications ranging from eco-driving of autonomous vehicles to quadrotor navigation. However, the nonlinear programs approximating the OCPs are inherently nonconvex due to the coupling between the dynamics and the event timing, and therefore, they are challenging to solve. Most approaches address this challenge by predefining waypoint times or just using nonconvex trajectory optimization, which simplifies the problem but often yields suboptimal solutions. To significantly improve the numerical properties, we propose a formulation with a time-scaling direct multiple shooting scheme that partitions the prediction horizon into segments aligned with characteristic time constraints. Moreover, we develop a fast semidefinite-programming-based convex relaxation that exploits the sparsity pattern of the lifted formulation. Comprehensive simulation studies demonstrate the solution optimality and computational efficiency. Furthermore, real-world experiments on a quadrotor waypoint flight task with constrained open time windows validate the practical applicability of the approach in complex environments.

cs.RO↗

arXiv2025

Rich State Observations Empower Reinforcement Learning to Surpass PID: A Drone Ball Balancing Study

This paper addresses a drone ball-balancing task, in which a drone stabilizes a ball atop a movable beam through cable-based interaction. We propose a hierarchical control framework that decouples high-level balancing policy from low-level drone control, and train a reinforcement learning (RL) policy to handle the high-level decision-making. Simulation results show that the RL policy achieves superior performance compared to carefully tuned PID controllers within the same hierarchical structure. Through systematic comparative analysis, we demonstrate that RL's advantage stems not from improved parameter tuning or inherent nonlinear mapping capabilities, but from its ability to effectively utilize richer state observations. These findings underscore the critical role of comprehensive state representation in learning-based systems and suggest that enhanced sensing could be instrumental in improving controller performance.

cs.RO↗

arXiv2025

Locally Differentially Private Multi-Sensor Fusion Estimation With System Intrinsic Randomness

This paper focuses on the privacy-preserving multi-sensor fusion estimation (MSFE) problem with differential privacy considerations. Most existing research efforts are directed towards the exploration of traditional differential privacy, also referred to as centralized differential privacy (CDP). It is important to note that CDP is tailored to protect the privacy of statistical data at fusion center such as averages and sums rather than individual data at sensors, which renders it inappropriate for MSFE. Additionally, the definitions and assumptions of CDP are primarily applicable for large-scale systems that require statistical results mentioned above. Therefore, to address these limitations, this paper introduces a more recent advancement known as \emph{local differential privacy (LDP)} to enhance the privacy of MSFE. We provide some rigorous definitions about LDP based on the intrinsic properties of MSFE rather than directly presenting the assumptions under CDP. Subsequently, the LDP is proved to be realized with system intrinsic randomness, which is useful and has never been considered before. Furthermore, the Gaussian mechanism is designed when the intrinsic randomness is insufficient. The lower bound of the covariance for extra injected Gaussian noises is determined by integrating system information with privacy budgets. Moreover, the optimal fusion estimators under intrinsic and extra disturbances are respectively designed in the linear minimum variance sense. Finally, the effectiveness of the proposed methods is verified through numerical simulations, encompassing both one-dimensional and high-dimensional scenarios.

eess.SY↗

arXiv2025

MAER-Nav: Bidirectional Motion Learning Through Mirror-Augmented Experience Replay for Robot Navigation

Deep Reinforcement Learning (DRL) based navigation methods have demonstrated promising results for mobile robots, but suffer from limited action flexibility in confined spaces. Conventional DRL approaches predominantly learn forward-motion policies, causing robots to become trapped in complex environments where backward maneuvers are necessary for recovery. This paper presents MAER-Nav (Mirror-Augmented Experience Replay for Robot Navigation), a novel framework that enables bidirectional motion learning without requiring explicit failure-driven hindsight experience replay or reward function modifications. Our approach integrates a mirror-augmented experience replay mechanism with curriculum learning to generate synthetic backward navigation experiences from successful trajectories. Experimental results in both simulation and real-world environments demonstrate that MAER-Nav significantly outperforms state-of-the-art methods while maintaining strong forward navigation capabilities. The framework effectively bridges the gap between the comprehensive action space utilization of traditional planning methods and the environmental adaptability of learning-based approaches, enabling robust navigation in scenarios where conventional DRL methods consistently fail.

cs.RO↗

arXiv2025

Risk-Aware Reinforcement Learning for Autonomous Driving: Improving Safety When Driving through Intersection

Applying reinforcement learning to autonomous driving has garnered widespread attention. However, classical reinforcement learning methods optimize policies by maximizing expected rewards but lack sufficient safety considerations, often putting agents in hazardous situations. This paper proposes a risk-aware reinforcement learning approach for autonomous driving to improve the safety performance when crossing the intersection. Safe critics are constructed to evaluate driving risk and work in conjunction with the reward critic to update the actor. Based on this, a Lagrangian relaxation method and cyclic gradient iteration are combined to project actions into a feasible safe region. Furthermore, a Multi-hop and Multi-layer perception (MLP) mixed Attention Mechanism (MMAM) is incorporated into the actor-critic network, enabling the policy to adapt to dynamic traffic and overcome permutation sensitivity challenges. This allows the policy to focus more effectively on surrounding potential risks while enhancing the identification of passing opportunities. Simulation tests are conducted on different tasks at unsignalized intersections. The results show that the proposed approach effectively reduces collision rates and improves crossing efficiency in comparison to baseline algorithms. Additionally, our ablation experiments demonstrate the benefits of incorporating risk-awareness and MMAM into RL.

cs.RO↗

arXiv2025

Enhancing Deep Reinforcement Learning-based Robot Navigation Generalization through Scenario Augmentation

This work focuses on enhancing the generalization performance of deep reinforcement learning-based robot navigation in unseen environments. We present a novel data augmentation approach called scenario augmentation, which enables robots to navigate effectively across diverse settings without altering the training scenario. The method operates by mapping the robot's observation into an imagined space, generating an imagined action based on this transformed observation, and then remapping this action back to the real action executed in simulation. Through scenario augmentation, we conduct extensive comparative experiments to investigate the underlying causes of suboptimal navigation behaviors in unseen environments. Our analysis indicates that limited training scenarios represent the primary factor behind these undesired behaviors. Experimental results confirm that scenario augmentation substantially enhances the generalization capabilities of deep reinforcement learning-based navigation systems. The improved navigation framework demonstrates exceptional performance by producing near-optimal trajectories with significantly reduced navigation time in real-world applications.

cs.RO↗

arXiv2024

Linking Order to Strength in Metals

The metallurgy and materials communities have long known and exploited fundamental links between chemical and structural ordering in metallic solids and their mechanical properties. The highest reported strength achievable through the combination of multiple metals (alloying) has rapidly climbed and given rise to new classifications of materials with extraordinary properties. Metallic glasses and high-entropy alloys are two limiting examples of how tailored order can be used to manipulate mechanical behavior. Here, we show that the complex electronic-structure mechanisms governing the peak strength of alloys and pure metals can be reduced to a few physically-meaningful parameters based on their atomic arrangements and used (with no fitting parameters) to predict the maximum strength of any metallic solid, regardless of degree of structural or chemical ordering. Predictions of maximum strength based on the activation energy for a stress-driven phase transition to an amorphous state is shown to accurately describe the breakdown in Hall-Petch behavior at the smallest crystallite sizes for pure metals, intermetallic compounds, metallic glasses, and high-entropy alloys. This activation energy is also shown to be directly proportional to interstitial (electronic) charge density, which is a good predictor of ductility, stiffness (moduli), and phase stability in high-entropy alloys, and in solid metals generally. The proposed framework suggests the possibility of coupling ordering and intrinsic strength to mechanisms like dislocation nucleation, hydrogen embrittlement, and transport properties. It additionally opens the prospect for greatly accelerated structural materials design and development to address materials challenges limiting more sustainable and efficient use of energy.

cond-mat.mtrl-sci↗

arXiv2024

Privacy-Preserving State Estimation in the Presence of Eavesdroppers: A Survey

Networked systems are increasingly the target of cyberattacks that exploit vulnerabilities within digital communications, embedded hardware, and software. Arguably, the simplest class of attacks -- and often the first type before launching destructive integrity attacks -- are eavesdropping attacks, which aim to infer information by collecting system data and exploiting it for malicious purposes. A key technology of networked systems is state estimation, which leverages sensing and actuation data and first-principles models to enable trajectory planning, real-time monitoring, and control. However, state estimation can also be exploited by eavesdroppers to identify models and reconstruct states with the aim of, e.g., launching integrity (stealthy) attacks and inferring sensitive information. It is therefore crucial to protect disclosed system data to avoid an accurate state estimation by eavesdroppers. This survey presents a comprehensive review of existing literature on privacy-preserving state estimation methods, while also identifying potential limitations and research gaps. Our primary focus revolves around three types of methods: cryptography, data perturbation, and transmission scheduling, with particular emphasis on Kalman-like filters. Within these categories, we delve into the concepts of homomorphic encryption and differential privacy, which have been extensively investigated in recent years in the context of privacy-preserving state estimation. Finally, we shed light on several technical and fundamental challenges surrounding current methods and propose potential directions for future research.

cs.CR↗

arXiv2023

PAI at SemEval-2023 Task 2: A Universal System for Named Entity Recognition with External Entity Information

The MultiCoNER II task aims to detect complex, ambiguous, and fine-grained named entities in low-context situations and noisy scenarios like the presence of spelling mistakes and typos for multiple languages. The task poses significant challenges due to the scarcity of contextual information, the high granularity of the entities(up to 33 classes), and the interference of noisy data. To address these issues, our team {\bf PAI} proposes a universal Named Entity Recognition (NER) system that integrates external entity information to improve performance. Specifically, our system retrieves entities with properties from the knowledge base (i.e. Wikipedia) for a given text, then concatenates entity information with the input sentence and feeds it into Transformer-based models. Finally, our system wins 2 first places, 4 second places, and 1 third place out of 13 tracks. The code is publicly available at \url{https://github.com/diqiuzhuanzhuan/semeval-2023}.

cs.CL↗

arXiv2023

An End-to-End Vehicle Trajcetory Prediction Framework

Anticipating the motion of neighboring vehicles is crucial for autonomous driving, especially on congested highways where even slight motion variations can result in catastrophic collisions. An accurate prediction of a future trajectory does not just rely on the previous trajectory, but also, more importantly, a simulation of the complex interactions between other vehicles nearby. Most state-of-the-art networks built to tackle the problem assume readily available past trajectory points, hence lacking a full end-to-end pipeline with direct video-to-output mechanism. In this article, we thus propose a novel end-to-end architecture that takes raw video inputs and outputs future trajectory predictions. It first extracts and tracks the 3D location of the nearby vehicles via multi-head attention-based regression networks as well as non-linear optimization. This provides the past trajectory points which then feeds into the trajectory prediction algorithm consisting of an attention-based LSTM encoder-decoder architecture, which allows it to model the complicated interdependence between the vehicles and make an accurate prediction of the future trajectory points of the surrounding vehicles. The proposed model is evaluated on the large-scale BLVD dataset, and has also been implemented on CARLA. The experimental results demonstrate that our approach outperforms various state-of-the-art models.

cs.CV↗

arXiv2022

Navigation of a UAV Equipped with a Reconfigurable Intelligent Surface for LoS Wireless Communication with a Ground Vehicle

Unmanned aerial vehicles (UAVs) have been successfully adopted to enhance the flexibility and robustness of wireless communication networks. And recently, the reconfigurable intelligent surface (RIS) technology has been paid increasing attention to improve the throughput of the fifth-generation (5G) millimeter-wave (mmWave) wireless communication. In this work, we propose an RIS-outfitted UAV (RISoUAV) to secure an uninterrupted line-of-sight (LoS) link with a ground moving target (MT). The MT can be an emergency ambulance and need a secure wireless communication link for continuous monitoring and diagnosing the health condition of a patient, which is vital for delivering critical patient care. In this light, real-time communication is required for sending various clinical multimedia data including videos, medical images, and vital signs. This significant target is achievable thanks to the 5G wireless communication assisted with RISoUAV. A two-stage optimization method is proposed to optimize the RISoUAV trajectory limited to UAV motion and LoS constraints. At the first stage, the optimal tube path of the RISoUAV is determined by taking into account the energy consumption, instant LoS link, and UAV speed/acceleration constraints. At the second stage, an accurate RISoUAV trajectory is obtained by considering the communication channel performance and passive beamforming. Simulation results show the accuracy and effectiveness of the method.

eess.SY↗

arXiv2022

Machine Learning-Aided Operations and Communications of Unmanned Aerial Vehicles: A Contemporary Survey

The ongoing amalgamation of UAV and ML techniques is creating a significant synergy and empowering UAVs with unprecedented intelligence and autonomy. This survey aims to provide a timely and comprehensive overview of ML techniques used in UAV operations and communications and identify the potential growth areas and research gaps. We emphasise the four key components of UAV operations and communications to which ML can significantly contribute, namely, perception and feature extraction, feature interpretation and regeneration, trajectory and mission planning, and aerodynamic control and operation. We classify the latest popular ML tools based on their applications to the four components and conduct gap analyses. This survey also takes a step forward by pointing out significant challenges in the upcoming realm of ML-aided automated UAV operations and communications. It is revealed that different ML techniques dominate the applications to the four key modules of UAV operations and communications. While there is an increasing trend of cross-module designs, little effort has been devoted to an end-to-end ML framework, from perception and feature extraction to aerodynamic control and operation. It is also unveiled that the reliability and trust of ML in UAV operations and applications require significant attention before full automation of UAVs and potential cooperation between UAVs and humans come to fruition.

cs.RO↗

arXiv2021

A Diversity-Enhanced and Constraints-Relaxed Augmentation for Low-Resource Classification

Data augmentation (DA) aims to generate constrained and diversified data to improve classifiers in Low-Resource Classification (LRC). Previous studies mostly use a fine-tuned Language Model (LM) to strengthen the constraints but ignore the fact that the potential of diversity could improve the effectiveness of generated data. In LRC, strong constraints but weak diversity in DA result in the poor generalization ability of classifiers. To address this dilemma, we propose a {D}iversity-{E}nhanced and {C}onstraints-\{R}elaxed {A}ugmentation (DECRA). Our DECRA has two essential components on top of a transformer-based backbone model. 1) A k-beta augmentation, an essential component of DECRA, is proposed to enhance the diversity in generating constrained data. It expands the changing scope and improves the degree of complexity of the generated data. 2) A masked language model loss, instead of fine-tuning, is used as a regularization. It relaxes constraints so that the classifier can be trained with more scattered generated data. The combination of these two components generates data that can reach or approach category boundaries and hence help the classifier generalize better. We evaluate our DECRA on three public benchmark datasets under low-resource settings. Extensive experiments demonstrate that our DECRA outperforms state-of-the-art approaches by 3.8% in the overall score.

cs.CL↗