Abstract With the continuous integration of functions in wearable devices, power consumption demands have increased significantly, posing serious challenges to conventional power supply methods. Wearable self-powered technologies offer an effective solution to this issue. This study focuses on the efficient utilization of solar energy from the human wrist and presents the design and implementation of a flexible photovoltaic wristband. The wristband employs a multi-directional parallel array of photovoltaic cells, integrated with an energy management module, enabling it to adapt effectively to the dynamic and non-uniform solar irradiance conditions on the wrist. Through both simulated sunlight and real outdoor environment tests, the energy harvesting and load-driving performances of the photovoltaic wristband were comprehensively evaluated. The results show that under a highest average outdoor illuminance of 37.38 × 10 3 lx (525 W·m −2 ), the wristband delivers an average output power of 15.88 mW, providing a stable 3.3 V supply to a wearable motion sensing node, thereby enabling self-powered operation. During a complete “energy accumulation-load activation” cycle, the sensing node can operate for 37.84 s to perceive and transmit data. By employing a one-dimensional convolutional neural networks (1D-CNN) algorithm, accurate recognition of four motion states is successfully achieved. This work presents a systematic study covering energy harvesting scenarios analysis, wristband design, performance evaluation, and sensing application. The proposed photovoltaic wristband demonstrates excellent cyclic energy accumulation and stable power supply capabilities, validating the feasibility and practicality of the wearable photovoltaic self-powered system and highlighting its promising potential for future wearable applications.

Decision-making and control are the core functionalities of high-level autonomous driving systems. Existing mainstream research, including modular and end-to-end paradigms, typically employ conservative strategies that compromise driving efficiency. However, driving efficiency constitutes a critical constraint on the transition of autonomous vehicles from mere operability to practical utility. Autonomous overtaking systems serve as a typical means to improve driving efficiency. Nevertheless, in stochastic and uncertain traffic scenarios, achieving safe and efficient continuous autonomous overtaking remains a significant challenge. In this context, this paper proposes a decision-making and control framework based on MeUAL to achieve the optimal trade-off between overtaking risk and efficiency. First, at the decision-making layer, a safe reinforcement learning method based on Uncertainty-aware Augmented Lagrangian (UAL) is developed to provide global overtaking guidance. Subsequently, the motion planning and control layer based on Model Predictive Control (MPC) closely tracks the UAL-generated guidance, while preserving the safety and constraint guarantees inherent to traditional MPC. Finally, a Policy Switching Mechanism (PSM) triggered by the safety epistemic uncertainty threshold is designed for the MeUAL-driven autonomous overtaking system. Experimental results demonstrate that MeUAL outperforms baseline algorithms with respect to reward-cost balance, sample efficiency, and learning stability. Moreover, in various test scenarios that are distinct from the training distribution, MeUAL-PSM exhibits strong robustness and interpretable overtaking maneuvers through flexible policy switching.

How to better evaluate the capabilities of Large Language Models (LLMs) is the focal point and hot topic in current LLMs research. Previous work has noted that due to the extremely high cost of iterative updates of LLMs, they are often unable to answer the latest dynamic questions well. To promote the improvement of Chinese LLMs’ ability to answer dynamic questions, in this paper, we introduce CDQA, a Chinese Dynamic QA benchmark containing question-answer pairs related to the latest news on the Chinese Internet. We obtain high-quality data through a pipeline that combines humans and models, and carefully classify the samples according to the frequency of answer changes to facilitate a more fine-grained observation of LLMs’ capabilities. We have also evaluated and analyzed mainstream and advanced Chinese LLMs on CDQA. Extensive experiments and valuable insights suggest that our proposed CDQA is challenging and worthy of more further study. We believe that the benchmark we provide will become one of the key data resources for improving LLMs’ Chinese question-answering ability in the future.

Generating long-term trajectories of dissipative chaotic systems autoregressively is a highly challenging task. The inherent positive Lyapunov exponents amplify prediction errors over time. Many chaotic systems possess a crucial property — ergodicity on their attractors, which makes long-term prediction possible. State-of-the-art methods address ergodicity by preserving statistical properties using optimal transport techniques. However, these methods face scalability challenges due to the curse of dimensionality when matching distributions. To overcome this bottleneck, we propose a scalable transformer-based framework capable of stably generating long-term high-dimensional and high-resolution chaotic dynamics while preserving ergodicity. Our method is grounded in a physical perspective, revisiting the Von Neumann mean ergodic theorem to ensure the preservation of long-term statistics in the $\mathcal{L}^2$ space. We introduce novel modifications to the attention mechanism, making the transformer architecture well-suited for learning large-scale chaotic systems. Compared to operator-based and transformer-based methods, our model achieves better performances across five metrics, from short-term prediction accuracy to long-term statistics. In addition to our methodological contributions, we introduce new chaotic system benchmarks: a machine learning dataset of 140$k$ snapshots of turbulent channel flow and a processed high-dimensional Kolmogorov Flow dataset, along with various evaluation metrics for both short- and long-term performances. Both are well-suited for machine learning research on chaotic systems.

Variational data assimilation estimates the dynamical system states by minimizing a cost function that fits the numerical models with the observational data. Although four-dimensional variational assimilation (4D-Var) is widely used, it faces high computational costs in complex nonlinear systems and depends on imperfect state-observation mappings. Deep learning (DL) offers more expressive approximators, while integrating DL models into 4D-Var is challenging due to their nonlinearities and lack of theoretical guarantees in assimilation results. In this paper, we propose \textit{Tensor-Var}, a novel framework that integrates kernel conditional mean embedding (CME) with 4D-Var to linearize nonlinear dynamics, achieving convex optimization in a learned feature space. Moreover, our method provides a new perspective for solving 4D-Var in a linear way, offering theoretical guarantees of consistent assimilation results between the original and feature spaces. To handle large-scale problems, we propose a method to learn deep features (DFs) using neural networks within the Tensor-Var framework. Experiments on chaotic systems and global weather prediction with real-time observations show that Tensor-Var outperforms conventional and DL hybrid 4D-Var baselines in accuracy while achieving a 10- to 20-fold speed improvement.

Global climate change has emerged as a pressing global challenge, underscoring the imperative for governments and urban traffic management authorities to prioritize carbon emission reduction in the transportation sector. In mixed traffic environments, where internal combustion engine vehicles and electric vehicles coexist, the disparity in carbon emissions between these vehicle types poses a significant challenge for the formulation of effective transportation coordination policies. Consensus-based speed advisory systems (CSAS) have been extensively employed to enhance fleet energy efficiency and mitigate emissions. This paper develops a novel vehicle speed advisory framework for mixed traffic environments, termed the MP-CSAS, where M stands for Mixed Traffic and P for Privacy-preserving, which leverages blockchain technology, privacy-preserving mechanisms, and secure car-following strategies. By incorporating a coordination factor, the framework enables policymakers to dynamically optimize carbon emission reduction strategies while safeguarding vehicle user data privacy and ensuring operational safety. Simulation results demonstrate that the MP-CSAS framework effectively minimizes fleet carbon emissions while preserving data confidentiality and ensuring system security. This study contributes a forward-looking decision-making paradigm for road infrastructure providers and policymakers, equipping them with scientifically grounded and adaptive strategies to achieve sustainable and safe transportation objectives.

Accurate vehicle trajectory prediction process depends on seamless data sharing within the Internet of Vehicles. However, such interconnected data exchange introduces significant security risks. Specifically, network attacks can compromise data integrity, thereby degrading prediction accuracy. Concurrently, the need to protect sensitive vehicle data, such as driving trajectories and user account information, results in data silos that hinder the free flow of information essential for effective prediction. Existing studies have largely addressed either privacy preservation or attack mitigation in isolation, lacking a unified solution that simultaneously tackles both challenges. To address this gap, we propose Fed-SecTP, an integrated dual-module secure federated learning framework. The first module employs a Temporal Convolutional Network (TCN) with multi-head attention to detect and filter network attacks in real-time. The second module combines TCN with a Bidirectional Long Short-Term Memory (Bi-LSTM) network for trajectory prediction and leverages FedProx for federated learning, thereby enabling privacy-preserving model training without sharing raw data. Experimental results demonstrate that Fed-SecTP achieves high prediction accuracy and robustness even when up to 50% of the data is compromised by attacks, while ensuring secure data processing. This framework offers a reliable and comprehensive solution for autonomous vehicle trajectory prediction.

This paper addresses the longitudinal control problem of an on-ramp vehicle merging into a platoon on the main road. To tackle this challenge, a finite-time model predictive control (MPC) algorithm with a specialized feedback control law is proposed. A constraint set of the state error is designed and based on this, a decision-making scheme is established to allow the on-ramp vehicle to assess the feasibility of the merging operation at the beginning under the designed MPC strategy. If the merging is feasible, the proposed MPC strategy will be applied to drive the on-ramp vehicle towards a small neighborhood around the desired state on the basis of platoon’s velocity and position within a finite time step before joining the platoon. Furthermore, asymptotic convergence towards the desired state is achieved by a co-designed feedback control law. Otherwise, the MPC strategy will not be triggered, instead an alternative method such as slowing down the on-ramp vehicle to create space and allow the vehicles on the main road to proceed ahead. Under the proposed method, the recursive feasibility of the MPC optimization problem is achieved at all time steps and the finite time convergence to the small neighborhood of the desired state can be proved under the MPC algorithm. An upper bound on the convergence time step is also derived, which is used to prove the effectiveness of the decision-making mechanism. In addition, the closed-loop constraints satisfaction and asymptotic stability of the on-ramp vehicle are also guaranteed. The effectiveness of the proposed MPC method is demonstrated through simulation examples.

Automotive energy management systems (EMSs) are evolving to achieve intelligence across their entire lifecycle, from initial product development to real-world customer usage. This paper proposes a bi-level transfer approach with model-agnostic meta-learning (MAML) to realize cross-platform transferable and online-adaptive EMS. In the development phase, MAML calibrates the heuristic control maps of an instantaneous optimization-based EMS, with a per-unit state-action space design ensuring seamless knowledge transfer between vehicle platforms. During the usage phase, real driving data are adopted to refine onboard control parameters. The framework is validated through real vehicle experiments. Firstly, the entire MAML-assisted V-cycle development is implemented to validate the optimality and knowledge transfer of the EMS, resulting in zero-shot transfer for EMS calibration on new vehicle products. Additionally, real vehicle experimental tests show that a correction of 8.0%~9.5% fuel economy is improved against the convention reinforcement learning-based EMS during usage via online-adaptation, effectively bridging the gap between pre-trained policies and real-world optimal energy management.

With the rapid growth of geographic applications, automatable and intelligent models are essential to be designed to handle the large volume of information. However, few researchers focus on geographic natural language processing, and there has never been a benchmark to build a unified standard. In this work, we propose a GeoGraphic Language Understanding Evaluation benchmark, named GeoGLUE. We collect data from open-released geographic resources and introduce six natural language understanding tasks, including geographic textual similarity on recall, geographic textual similarity on rerank, geographic elements tagging, geographic composition analysis, geographic where what cut, and geographic entity alignment. We also provide evaluation experiments and analysis of general baselines, indicating the effectiveness and significance of the GeoGLUE benchmark ( https://modelscope.cn/datasets/iic/GeoGLUE/summary ).

Yong Cao, Ruixue Ding, Boli Chen, Xianzhi Li, Min Chen, Daniel Hershcovich, Pengjun Xie, Fei Huang. Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.

Implanted devices are increasingly used in chronic disease monitoring, but face challenges in energy autonomy. This article presents a novel wireless power transfer (WPT) method for self-sustained medical implants using Halbach array-based magnetic plucking and piezoelectric transduction. The wearable-implantable coupled system consists of a piezoelectric receiver within the implant to receive power and a near-field magnetic power transmitter as a wearable device. To deliver power over greater distances through the human body, the transmitter features a rotating magnetic Halbach array powered by a miniature motor, or by human motion, to generate an alternating magnetic field. The use of low-frequency rotating magnetic fields periodically excites a cantilevered piezoelectric beam with a tip magnet to realize WPT. A theoretical model that includes magnetic coupling, piezoelectric transduction and receiver beam dynamics has been established to study the electro-magneto-mechanical dynamics of this WPT system. The effectiveness of the Halbach array for extended power transfer is examined through theoretical modeling and numerical simulation, showing a 37.2% enhancement of the magnetic forces. A prototype was also fabricated and tested to examine the WPT performance. The established wireless power link can provide sufficient power ($\sim 32~\mu $W) over a large transmission distance (22 mm), providing a potential battery-free solution for the self-sustained Internet of Implanted Things (IoIT) for personalized healthcare.

With the increasing penetration of inverter-based renewable energy resources, deep reinforcement learning (DRL) has been proposed as one of the most promising solutions to realize real-time and autonomous control for future carbon-neutral power systems. In particular, DRL-based frequency control approaches have been extensively investigated to overcome the limitations of model-based approaches, such as the computational cost and scalability for large-scale systems. Nevertheless, the real-world implementation of DRLbased frequency control methods is facing the following fundamental challenges: 1) safety guarantee during the learning and decision-making processes; 2) adaptability against the dynamic system operating conditions. To this end, this is the first work that proposes an Adaptive and Safe-Certified DRL (AdapSafe) algorithm for frequency control to simultaneously address the aforementioned challenges. In particular, a novel self-tuning control barrier function is designed to actively compensate the unsafe frequency control strategies under variational safety constraints and thus achieve guaranteed safety. Furthermore, the concept of meta-reinforcement learning is integrated to significantly enhance its adaptiveness in non-stationary power system environments without sacrificing the safety cost. Experiments are conducted based on GB 2030 power system, and the results demonstrate that the proposed AdapSafe exhibits superior performance in terms of its guaranteed safety in both training and test phases, as well as its considerable adaptability against the dynamics changes of system parameters.

Query and point of interest (POI) matching is a core task in location-based services~(LBS), e.g., navigation maps. It connects users' intent with real-world geographic information. Lately, pre-trained language models (PLMs) have made notable advancements in many natural language processing (NLP) tasks. To overcome the limitation that generic PLMs lack geographic knowledge for query-POI matching, related literature attempts to employ continued pre-training based on domain-specific corpus. However, a query generally describes the geographic context (GC) about its destination and contains mentions of multiple geographic objects like nearby roads and regions of interest (ROIs). These diverse geographic objects and their correlations are pivotal to retrieving the most relevant POI. Text-based single-modal PLMs can barely make use of the important GC and are therefore limited. In this work, we propose a novel method for query-POI matching, namely Multi-modal Geographic language model (MGeo), which comprises a geographic encoder and a multi-modal interaction module. Representing GC as a new modality, MGeo is able to fully extract multi-modal correlations to perform accurate query-POI matching. Moreover, there exists no publicly available query-POI matching benchmark. Intending to facilitate further research, we build a new open-source large-scale benchmark for this topic, i.e., Geographic TExtual Similarity (GeoTES). The POIs come from an open-source geographic information system (GIS) and the queries are manually generated by annotators to prevent privacy issues. Compared with several strong baselines, the extensive experiment results and detailed ablation analyses demonstrate that our proposed multi-modal geographic pre-training method can significantly improve the query-POI matching capability of PLMs with or without users' locations. Our code and benchmark are publicly available at https://github.com/PhantomGrapes/MGeo.

Cooperative vehicle management emerges as a promising solution to improve road traffic safety and efficiency. This paper addresses the speed planning problem for connected and autonomous vehicles (CAVs) at an unsignalized intersection with consideration of turning maneuvers. The problem is approached by a hierarchical centralized coordination scheme that successively optimizes the crossing order and velocity trajectories of a group of vehicles so as to minimize their total energy consumption and travel time required to pass the intersection. For an accurate estimate of the energy consumption of each CAV, the vehicle modeling framework in this paper captures 1) friction losses that affect longitudinal vehicle dynamics, and 2) the powertrain of each CAV in line with a battery-electric architecture. It is shown that the underlying optimization problem subject to safety constraints for powertrain operation, cornering and collision avoidance, after convexification and relaxation in some aspects can be formulated as two second-order cone programs, which ensures a rapid solution search and a unique global optimum. Simulation case studies are provided showing the tightness of the convex relaxation bounds, the overall effectiveness of the proposed approach, and its advantages over a benchmark solution invoking the widely used first-in-first-out policy. The investigation of Pareto optimal solutions for the two objectives (travel time and energy consumption) highlights the importance of optimizing their trade-off, as small compromises in travel time could produce significant energy savings.

This paper proposes a distributed control scheme for a platoon of heterogeneous vehicles based on the mechanism of model predictive control (MPC). The platoon composes of a group of vehicles interacting with each other via inter-vehicular spacing constraints, to avoid collision and reduce communication latency, and aims to make multiple vehicles driving on the same lane safely with a close range and the same velocity. Each vehicle is subject to both state constraints and input constraints, communicates only with neighboring vehicles, and may not know a priori desired setpoint. We divide the computation of control inputs into several local optimization problems based on each vehicle’s local information. To compute the control input of each vehicle based on local information, a distributed computing method must be adopted and thus the coupled constraint is required to be decoupled. This is achieved by introducing the reference state trajectories from neighboring vehicles for each vehicle and by employing the interactive structure of computing local problems of vehicles with odd indices and even indices. It is shown that the feasibility of MPC optimization problems is achieved at all time steps based on tailored terminal inequality constraints, and the asymptotic stability of each vehicle to the desired trajectory is guaranteed even under a single iteration between vehicles at each time. Finally, a comparison simulation is conducted to demonstrate the effectiveness of the proposed distributed MPC method for heterogeneous vehicle control with respect to normal and extreme scenarios.

Named Entity Recognition (NER) from speech is among Spoken Language Understanding (SLU) tasks, aiming to extract semantic information from the speech signal. NER from speech is usually made through a two-step pipeline that consists of (1) processing the audio using an Automatic Speech Recognition (ASR) system and (2) applying an NER tagger to the ASR outputs. Recent works have shown the capability of the End-to-End (E2E) approach for NER from English and French speech, which is essentially entity-aware ASR. However, due to the many homophones and polyphones that exist in Chinese, NER from Chinese speech is effectively a more challenging task. In this paper, we introduce a new dataset AISEHLL-NER for NER from Chinese speech. Extensive experiments are conducted to explore the performance of several state-of-the-art methods. The results demonstrate that the performance could be improved by combining entity-aware ASR and pretrained NER tagger, which can be easily applied to the modern SLU pipeline. The dataset is publicly available at github.com/Alibaba-NLP/AISHELL-NER.

This paper investigates the path tracking control problem of autonomous vehicles subject to modelling uncertainties and external disturbances. The problem is approached by employing a 2-degree of freedom vehicle model, which is reformulated into a newly defined parametric form with the system uncertainties being lumped into an unknown parametric vector. On top of the parametric system representation, a novel robust adaptive learning control (RALC) approach is then developed, which estimates the system uncertainties through iterative learning while treating the external disturbances by adopting a robust term. It is shown that the proposed approach is able to improve the lateral tracking performance gradually through learning from previous control experiences, despite only partial knowledge of the vehicle dynamics being available. It is noteworthy that a novel technique targeting at the non-square input distribution matrix is employed so as to deal with the under-actuation property of the vehicle dynamics, which extends the adaptive learning control theory from square systems to non-square systems. Moreover, the convergence properties of the RALC algorithm are analysed under the framework of Lyapunov-like theory by virtue of the composite energy function and the$\lambda $-norm. The effectiveness of the proposed control scheme is verified by representative simulation examples and comparisons with existing methods.

Different from the traditional classification tasks which assume mutual exclusion of labels, hierarchical multi-label classification (HMLC) aims to assign multiple labels to every instance with the labels organized under hierarchical relations. Besides the labels, since linguistic ontologies are intrinsic hierarchies, the conceptual relations between words can also form hierarchical structures. Thus it can be a challenge to learn mappings from word hierarchies to label hierarchies. We propose to model the word and label hierarchies by embedding them jointly in the hyperbolic space. The main reason is that the tree-likeness of the hyperbolic space matches the complexity of symbolic data with hierarchical structures. A new Hyperbolic Interaction Model (HyperIM) is designed to learn the label-aware document representations and make predictions for HMLC. Extensive experiments are conducted on three benchmark datasets. The results have demonstrated that the new model can realistically capture the complex data structures and further improve the performance for HMLC comparing with the state-of-the-art methods. To facilitate future research, our code is publicly available.

Although deep neural networks are effective at extracting high-level features, classification methods usually encode an input into a vector representation via simple feature aggregation operations (e.g.pooling).Such operations limit the performance.For instance, a multi-label document may contain several concepts.In this case, one vector can not sufficiently capture its salient and discriminative content.Thus, we propose Hyperbolic Capsule Networks (HYPERCAPS) for Multi-Label Classification (MLC), which have two merits.First, hyperbolic capsules are designed to capture fine-grained document information for each label, which has the ability to characterize complicated structures among labels and documents.Second, Hyperbolic Dynamic Routing (HDR) is introduced to aggregate hyperbolic capsules in a label-aware manner, so that the label-level discriminative information can be preserved along the depth of neural networks.To efficiently handle large-scale MLC datasets, we additionally present a new routing method to adaptively adjust the capsule number during routing.Extensive experiments are conducted on four benchmark datasets.Compared with the state-of-the-art methods, HY-PERCAPS significantly improves the performance of MLC especially on tail labels.

Lin Xiao, Xin Huang, Boli Chen, Liping Jing. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). 2019.

This work introduces a new algorithm, named Global Quadrature PLL (GQPLL) for tracking a sinusoidal signal and for estimating its frequency and amplitude. The proposed technique derives from the well-known PLL architecture based on Quadrature Signal Generation, that is widely used for tracking the fundamental of single-phase electrical signals. The proposed algorithm improves the existing quadrature-PLL solutions from two different perspectives. First, the cancellation of the DC-bias is embedded by construction. Moreover, a Lyapunov-based stability analysis guarantees the global convergence of the estimates for arbitrarily large adaptation gains, enabling fast adaptation transients. Simulations show that the proposed algorithm is able to deal with sudden variations of the fundamental frequency and of the DC-bias magnitude.