Hybrid action space, which combines discrete choices and continuous parameters, is prevalent in domains such as robot control and game AI. However, efficiently modeling and optimizing hybrid discrete-continuous action space remains a fundamental challenge, mainly due to limited policy expressiveness and poor scalability in high-dimensional settings. To address this challenge, we view the hybrid action space problem as a fully cooperative game and propose a Cooperative Hybrid Diffusion Policies (CHDP) framework to solve it. CHDP employs two cooperative agents that leverage a discrete and a continuous diffusion policy, respectively. The continuous policy is conditioned on the discrete action's representation, explicitly modeling the dependency between them. This cooperative design allows the diffusion policies to leverage their expressiveness to capture complex distributions in their respective action spaces. To mitigate the update conflicts arising from simultaneous policy updates in this cooperative setting, we employ a sequential update scheme that fosters co-adaptation. Moreover, to improve scalability when learning in high-dimensional discrete action space, we construct a codebook that embeds the action space into a low-dimensional latent space. This mapping enables the discrete policy to learn in a compact, structured space. Finally, we design a Q-function-based guidance mechanism to align the codebook's embeddings with the discrete policy's representation during training. On challenging hybrid action benchmarks, CHDP outperforms state-of-the-art method by up to 19.3% in success rate.

Precise environmental perception is critical for the reliability of autonomous driving systems. While collaborative perception mitigates the limitations of single-agent perception through information sharing, it encounters a fundamental communication-performance trade-off. Existing communication-efficient approaches typically assume MB-level data transmission per collaboration, which may fail due to practical network constraints. To address these issues, we propose InfoCom, an information-aware framework establishing the pioneering theoretical foundation for communication-efficient collaborative perception via extended Information Bottleneck principles. Departing from mainstream feature manipulation, InfoCom introduces a novel information purification paradigm that theoretically optimizes the extraction of minimal sufficient task-critical information under Information Bottleneck constraints. Its core innovations include: i) An Information-Aware Encoding condensing features into minimal messages while preserving perception-relevant information; ii) A Sparse Mask Generation identifying spatial cues with negligible communication cost; and iii) A Multi-Scale Decoding that progressively recovers perceptual information through mask-guided mechanisms rather than simple feature reconstruction. Comprehensive experiments across multiple datasets demonstrate that InfoCom achieves near-lossless perception while reducing communication overhead from megabyte to kilobyte-scale, representing 440-fold and 90-fold reductions per agent compared to Where2comm and ERMVP, respectively.

Autistic people often experience heightened emotional and sensory demands that affect wellbeing. Creative drawing supports expression and emotion regulation, yet an open-ended process can introduce uncertainty and overstimulation, particularly for autistic individuals who prefer structure. We present the Structured Creativity Toolkit (SCT), a set of tangible artifacts including custom laser-cut stencils, composition templates, and guided color palettes, designed to scaffold creative expression and self-regulation. We first conducted a qualitative study to identify design opportunities and user needs, then evaluated SCT through a mixed-methods approach combining a controlled study with post-study interviews to examine engagement, user experience, and self-reported wellbeing. Our findings indicate that SCT increased drawing engagement and perceived drawing quality while supporting self-reported psychological and emotional benefits. This paper contributes: (1) a creativity-support toolkit for structured drawing, and (2) evidence that order-affirming scaffolds can translate preferences for structure into design resources that support autistic individuals.

Deep Reinforcement Learning (DRL) methods have shown remarkable success in many applications, yet their high energy consumption limits their practicability. Recent studies incorporated energy-efficient Spiking Neural Networks (SNNs) to build Spiking DRL methods and lower energy consumption by setting a shorter simulation duration for SNNs to compute fewer gradients. However, these existing Spiking DRL methods fail to sample sufficient high-quality samples within a fixed-size replay buffer and perform poorly when the simulation duration is small, introducing the challenging tradeoff between energy consumption and model performance. Motivated by such observations, we develop a generic resilient experience replay method that can be seamlessly integrated into existing spiking DRL methods to effectively address the above tradeoff. Specifically, we allow the replay buffer to dynamically expand as the number of training samples increases, thereby accommodating more potentially valuable candidate samples for policy training. Meanwhile, we introduce an adaptive approach to manage the buffer size by determining when to shrink the replay buffer and removing redundant samples automatically. This strategy prevents the buffer from expanding unnecessarily, thereby mitigating the potential negative impact on model performance. Extensive experimental results demonstrate that our approach significantly enhances the performance of five state-of-the-art (SOTA) spiking DRL methods across various simulation durations in sixteen tasks, in terms of return, without compromising their energy efficiency.

Sharding protocols in cryptocurrencies are dedicated to improving system throughput and reducing cross-shard transaction latency. However, they face the challenge of state explosion due to their stateful design. An intuitive solution is state sharding. Existing state sharding protocols aim to reduce the incidence of cross-shard transactions and balance transaction load. Nevertheless, it inevitably introduces the complex and sequential processing of cross-shard transactions and neglects the security of state data. To address these challenges, we propose SecChain, a secure stateless sharded blockchain via a novel state commitment in this paper. SecChain fully exploits the historical characteristics of transactions and designs the parallel cross-shard transaction execution based on asynchronous prepaid accounts. We then introduce a novel state commitment scheme to ensure the security of state data and improve the efficiency of transaction validation. In addition, we optimize the on-chain and off-chain storage by integrating an incremental partial state trie on-chain and a full trie with proofs off-chain. We perform a comprehensive evaluation using real-world workloads and results demonstrate that SecChain's throughput is 1.75× that of BrokerChain, and its latency is reduced by 7×. Furthermore, SecChain significantly reduces both proof overhead and storage overhead.

Pseudonym certificates play a crucial role in providing authorized access in vehicular networks with fine-grained privacy demands. However, the revocation of pseudonym certificates in large-scale, resource-constrained context has posed a significant challenge. The global revocation status is susceptible to disruption by localized anomalous events. Moreover, as the scale of revocation grows, the synchronization process becomes increasingly vulnerable to attacks and incurs high overhead. To decouple the global revocation status from local revocation events, we propose a spatial-temporal pseudonym revocation and verification framework that supports batch pseudonym revocation within resilient revocation cycles. Revoked pseudonyms within the event area are locally filtered, while other pseudonyms are characterized by maintaining a proof to attest their consistent and legitimate status. To improve the efficiency of revocation status updates and mitigate centralization risks, a distributed caching and proof update strategy assisted by edge nodes is presented. Extensive simulations based on real-world large-scale datasets demonstrate that PV-STR scheme significantly reduces the vulnerability window, communication overhead, and memory consumption compared to state-of-the-art approaches.

Entanglement routing plays a vital role in supporting various applications in quantum networks. Existing works on entanglement routing either ignored the Entangled Photon Source (EPS) placement issue or simply assumed a pool of EPSes at a centralized location that can provision entanglement over arbitrary quantum links. In this paper, we propose LIGHTER and fidelity-aware LIGHTER (named F-LIGHTER) to solve the joint EPS placement and entanglement routing problem based on the assumption that EPSes are co-located with quantum nodes and each EPS can send one entangled photon at a time to one of its adjacent nodes only. The salient features of LIGHTER and F-LIGHTER include (i) LIGHTER and F-LIGHTER use a demand-agnostic EPS placement scheme to maximize network throughput and fairness for all feasible Entanglement Connection EC) establishment demands, and (ii) most requested ECs can be established over Entanglement Paths (EPs) determined offline, and only a small percentage of them will be established over online calculated EPs, resulting in fast and efficient entanglement routing. Extensive simulations show that compared with schemes without proper EPS placement or entanglement routing, LIGHTER can improve the network throughput by up to 175.6% and 37.0%, respectively. When the fidelity is considered, the network throughput improvement achieved by F-LIGHTER will be up to 135.0% and 21.5%, respectively.

Multi-agent collaborative perception is expected to significantly improve perception performance by overcoming the limitations of single-agent perception through exchanging complementary information. However, training a robust collaborative perception model requires collecting sufficient training data that covers all possible collaboration scenarios, which is impractical due to intolerable deployment costs. Hence, the trained model is not robust against new traffic scenarios with inconsistent data distribution and fundamentally restricts its real-world applicability. Further, existing methods, such as domain adaptation, have mitigated this issue by exposing the deployment data during the training stage but incur a high training cost, which is infeasible for resource-constrained agents. In this paper, we propose a Parameter-Efficient Fine-Tuning-based lightweight framework, CoPEFT, for fast adapting a trained collaborative perception model to new deployment environments under low-cost conditions. CoPEFT develops a Collaboration Adapter and Agent Prompt to perform macro-level and micro-level adaptations separately. Specifically, the Collaboration Adapter utilizes the inherent knowledge from training data and limited deployment data to adapt the feature map to new data distribution. The Agent Prompt further enhances the Collaboration Adapter by inserting fine-grained contextual information about the environment. Extensive experiments demonstrate that our CoPEFT surpasses existing methods with less than 1\% trainable parameters, proving the effectiveness and efficiency of our proposed method.

Trajectory prediction plays a crucial role in autonomous driving systems, and exploring its vulnerability has garnered widespread attention. However, existing trajectory prediction attack methods often rely on single-point attacks to make efficient perturbations. This limits their applications in real-world scenarios due to the transient nature of single-point attacks, their susceptibility to filtration, and the uncertainty regarding the deployment environment. To address these challenges, this paper proposes a novel LiDAR-induced attack framework to impose multi-frame attacks by optimization-driven adversarial location search, achieving endurance, efficiency, and robustness. This framework strategically places objects near the adversarial vehicle to implement an attack and introduces three key innovations. First, successive state perturbations are generated using a multi-frame single-point attack strategy, effectively misleading trajectory predictions over extended time horizons. Second, we efficiently optimize adversarial objects’ locations through three specialized loss functions to achieve desired perturbations. Lastly, we improve robustness by treating the adversarial object as a point without size constraints during the location search phase and reduce dependence on both the specific attack point and the adversarial object’s properties. Extensive experiments confirm the superior performance and robustness of our framework.

Federated Unlearning (FU) enables clients to selectively remove the influence of specific data from a trained federated learning model, addressing privacy concerns and regulatory requirements. However, existing FU methods often struggle to balance effective erasure with model utility preservation, especially for class-level unlearning in non-IID settings. We propose Federated Unlearning via Class-aware Representation Transformation (FUCRT), a novel method that achieves unlearning through class-aware representation transformation. FUCRT employs two key components: (1) a transformation class selection strategy to identify optimal forgetting directions, and (2) a transformation alignment technique using dual class-aware contrastive learning to ensure consistent transformations across clients. Extensive experiments on four datasets demonstrate FUCRT's superior performance in terms of erasure guarantee, model utility preservation, and efficiency. FUCRT achieves complete (100\%) erasure of unlearning classes while maintaining or improving performance on remaining classes, outperforming state-of-the-art baselines across both IID and Non-IID settings. Analysis of the representation space reveals FUCRT's ability to effectively merge unlearning class representations with the transformation class from remaining classes, closely mimicking the model retrained from scratch.

Collaborative perception significantly enhances individual vehicle perception performance through the exchange of sensory information among agents. However, real-world deployment faces challenges due to bandwidth constraints and inevitable calibration errors during information exchange. To address these issues, we propose mmCooper, a novel multi-agent, multi-stage, communication-efficient, and collaboration-robust cooperative perception framework. Our framework leverages a multi-stage collaboration strategy that dynamically and adaptively balances intermediate- and late-stage information to share among agents, enhancing perceptual performance while maintaining communication efficiency. To support robust collaboration despite potential misalignments and calibration errors, our framework prevents misleading low-confidence sensing information from transmission and refines the received detection results from collaborators to improve accuracy. The extensive evaluation results on both real-world and simulated datasets demonstrate the effectiveness of the mmCooper framework and its components.

Vertical Federated Learning (VFL) has emerged as a popular machine learning paradigm, enabling model training between the data and the task parties with different features about the same user set while preserving data privacy. In a production environment, VFL usually involves one task party and one data party. Fair and economically efficient feature trading is crucial to the commercialization of VFL, where the task party is considered the data consumer who buys the data party's features. However, current VFL feature trading practices often price the data party's data as a whole and assume transactions occur before performing VFL. Neglecting the performance gains resulting from traded features may lead to underpayment and overpayment issues. In this study, we propose a bargaining-based feature trading approach in VFL to facilitate economically efficient transactions. Our model incorporates performance gain-based pricing, taking into account the revenue-based optimization objectives of both parties. We analyze the proposed bargaining model under perfect and imperfect performance information settings, proving the existence of an equilibrium that optimizes the parties' objectives. Moreover, we develop performance gain estimation-based bargaining strategies for imperfect performance information scenarios and discuss potential security concerns and solutions. Experiments on three real-world datasets demonstrate the effectiveness of the proposed bargaining model.

The multi-agent pathfinding problem seeks to generate low-cost paths for agents to reach their targets, playing a crucial role in advancing the development of smart warehousing. Traditional methods do not utilize neural networks, resulting in high computational costs and poor scalability. In contrast, learning-based approaches leverage reinforcement learning to handle large-scale scenarios more effectively. However, they often depend heavily on expert demonstrations, and their plain reward structures can lead to suboptimal path planning. To address these challenges, we propose a novel multi-agent reinforcement learning framework: Hierarchical Decision-making Pathfinding Based on Social Value Orientation (HIPS). Specifically, HIPS adopts a hierarchical decision-making structure in which the lower-level interaction policy generates actions to interact with the environment, guided by egoistic rewards to promote low-cost path planning. Moreover, the upper-level policy is environmentoriented, learning dynamic social value orientations that enable agents to plan paths while accounting for interactions with other agents, thereby achieving an adaptive balance between individual self-interest and long-term team benefits. Extensive experimental validation demonstrates that HIPS increases the number of finished targets by at least 6% compared to existing methods without relying on expert experience.

Federated Learning (FL) enables collaborative model training while preserving data privacy, but it is highly vulnerable to backdoor attacks. Most existing defense methods in FL have limited effectiveness due to their neglect of the model's over-reliance on backdoor triggers, particularly as the proportion of malicious clients increases. In this paper, we propose FedBAP, a novel defense framework for mitigating backdoor attacks in FL by reducing the model's reliance on backdoor triggers. Specifically, first, we propose a perturbed trigger generation mechanism that creates perturbation triggers precisely matching backdoor triggers in location and size, ensuring strong influence on model outputs. Second, we utilize these perturbation triggers to generate benign adversarial perturbations that disrupt the model's dependence on backdoor triggers while forcing it to learn more robust decision boundaries. Finally, we design an adaptive scaling mechanism to dynamically adjust perturbation intensity, effectively balancing defense strength and model performance. The experimental results demonstrate that FedBAP reduces the attack success rates by 0.22%-5.34%, 0.48%-6.34%, and 97.22%-97.6% under three types of backdoor attacks, respectively. In particular, FedBAP demonstrates outstanding performance against novel backdoor attacks.

Infrastructure-assisted autonomous driving has emerged as a pivotal technology to overcome the challenges posed by occlusions and limited fields of view for individual vehicles. Vehicles can fuse perception information from the infrastructure with their own in real-time, thereby enhancing their perception ability. However, our real-world experiments demonstrate that such an approach could introduce artifacts such as ghost objects, resulting in unsafe and unreliable planning outcomes. Besides, the system integration complexity and communication overhead are typically considerable, posing challenges to practical deployment. Therefore, we propose VI-Planning, an innovative infrastructure-assisted system that effectively optimizes autonomous vehicle planning in real time. The core idea of VI-Planning is to leverage the scene-level future occupancy grid maps constructed by the infrastructure as future drivable area references to directly optimize planning outcomes of autonomous vehicles. Since VI-Planning operates only at the autonomous vehicle's final output stage, without modifying the vehicle's underlying system architecture, it can be plug-and-play for most autonomous driving systems, whether they are modular or end-to-end architectures. Moreover, VI-Planning employs a novel bitwise encoding mechanism to efficiently compress these maps, enabling practical transmission. We implement VI-Planning end-to-end on a real-world testbed. The results of closed-loop and open-loop experiments indicate that VI-Planning can achieve real-time planning optimization (62.54 ms on average) and 817 × data transmission efficiency compared to the state-of-the-art baseline. A video demo of VI-Planning on our real-world testbed is available at: https://youtu.be/DXl5BhDEvFQ.

We propose VI-Planning, an innovative system that leverages the scene-level future occupancy grid maps predicted by the infrastructure as future drivable area references to directly optimize the planning trajectories of autonomous vehicles in real time. Since VI-Planning operates only at the autonomous vehicle's final output stage, without modifying the underlying system architecture, it can be plug-and-play for most autonomous driving systems. Moreover, VI-Planning employs a novel bitwise encoding mechanism to efficiently compress these maps, enabling practical transmission. The experimental results demonstrate that VI-Planning can achieve real-time planning optimization with extremely low bandwidth consumption, significantly enhancing the driving safety of autonomous systems. The source code and video demonstration of VI-Planning are available on GitHub: https://github.com/YANG-Deep/VI-Planning.

We propose VI-Planning, an innovative system that leverages the scene-level future occupancy grid maps predicted by the infrastructure as future drivable area references to directly optimize the planning trajectories of autonomous vehicles in real time. Since VI-Planning operates only at the autonomous vehicle's final output stage, without modifying the underlying system architecture, it can be plug-and-play for most autonomous driving systems. Moreover, VI-Planning employs a novel bitwise encoding mechanism to efficiently compress these maps, enabling practical transmission. The experimental results demonstrate that VI-Planning can achieve real-time planning optimization with extremely low bandwidth consumption, significantly enhancing the driving safety of autonomous systems. The source code and video demonstration of VI-Planning are available on GitHub: https://github.com/YANG-Deep/VI-Planning.

Multi-agent collaboration enhances the perception capabilities of individual agents through information sharing. However, in real-world applications, differences in sensors and models across heterogeneous agents inevitably lead to domain gaps during collaboration. Existing approaches based on adaptation and reconstruction fail to support *pragmatic heterogeneous collaboration* due to two key limitations: (1) Intrusive retraining of the encoder or core modules disrupts the established semantic consistency among agents; and (2) accommodating new agents incurs high computational costs, limiting scalability. To address these challenges, we present a novel **Gen**erative **Comm**unication mechanism (GenComm) that facilitates seamless perception across heterogeneous multi-agent systems through feature generation, without altering the original network, and employs lightweight numerical alignment of spatial information to efficiently integrate new agents at minimal cost. Specifically, a tailored Deformable Message Extractor is designed to extract spatial message for each collaborator, which is then transmitted in place of intermediate features. The Spatial-Aware Feature Generator, utilizing a conditional diffusion model, generates features aligned with the ego agent's semantic space while preserving the spatial information of the collaborators. These generated features are further refined by a Channel Enhancer before fusion. Experiments conducted on the OPV2V-H, DAIR-V2X and V2X-Real datasets demonstrate that GenComm outperforms existing state-of-the-art methods, achieving an 81\% reduction in both computational cost and parameter count when incorporating new agents. Our code is available at https://github.com/jeffreychou777/GenComm.

Federated learning (FL), as an emerging distributed machine learning paradigm, aims to achieve collaborative training of a global model without sharing clients’ local data, thereby addressing user privacy and data silo issues. However, due to the distributed architecture of FL and the invisibility of local training, it is highly susceptible to poisoning attacks. While numerous poisoning defense methods have been proposed, they often suffer from limited defense effectiveness and overly strong security assumptions. To address this problem, we propose FedTPD, a dual gradient evaluation-based defense method against poisoning attacks in FL. Specifically, we first design an adaptive representative gradient selection mechanism, which involves adaptively clustering local gradients and selecting representative gradients from different clusters. Secondly, we propose a client-assisted model evaluation mechanism, wherein a reliable client is selected as an evaluator to assess the representative gradients from various clusters. Finally, we develop a bias-corrected model aggregation mechanism that aggregates the local gradients of each cluster based on the trust scores of their representative gradients, thereby reducing the impact of poisoning attacks on the global model. Experimental results on three typical datasets demonstrate that the proposed FedTPD can effectively resist various poisoning attacks. In particular, compared with state-of-the-art methods, the global model’s maximum and average accuracies on the CIFAR-10 and CIFAR-100 datasets are improved by 2.79%–28.15% and 0.79%–31.34%, and 3.48%–22.18% and 2.56%–22.06%, respectively.

In automated warehouses, efficient management and economic benefits hinge on the effective scheduling of automated guided vehicles (AGVs) to transport diverse packets. Emerging technologies such as artificial intelligence and automation control have greatly contributed to the development of packet transport schemes for AGVs. However, the development of the logistics industry results in a massive amount of packets with diverse deadlines, which brings new challenges for the AGV scheduling system. To address this, this paper treats each AGV as an agent and designs a novel hierarchical transformer reinforcement learning (HTRL) framework to generate efficient AGV scheduling policies. Specifically, this framework consists of one encoder and two decoders to produce the packet selection and path improvement actions. These two decoders are equipped with masked self-attention mechanisms to learn efficient packet selection and path improvement policies, facilitating AGV transport efficiency to meet the deadlines of packets. Moreover, we consider the kinetic features of AGVs and design a model predictive control (MPC)-based speed control method for AGVs to prevent frequent stop-and-wait of AGVs and enhance their transport efficiency. We build up a simulated warehouse environment containing packets with different deadlines and conduct extensive experiments. Experimental results validate that the proposed HTRL framework increases the delivered packets within expiration by up to 36.6% compared to other baselines.

Lidar can provide three-dimensional detection of the subsurface layer of the global ocean and represents the future direction of ocean remote sensing. One equation and two unknowns in elastic scattering lidar greatly affect the inversion accuracy. Based on airborne oceanographic lidar and in-situ measurements, an error analysis of widely used profile parameter inversion methods was performed, and a high-precision inversion method was developed. After careful data preprocessing, error analysis was performed on bio-optical parameters obtained from inversion methods including Collis Slope method, Klett backscatter iteration method, Churnside perturbation retrieval method and empirical models. In response to the inversion accuracy problems, a parameters optimization method has been proposed that does not require the assumption of a lidar ratio or a constant of lidar attenuation coefficient. Inversion results from the airborne lidar under the conditions of Case I waters in the South China Sea show that the method effectively improves the inversion accuracy of the β(π) profile to about 10% at depths over 50 m where the bio-optical parameters underwater change rapidly, and it can avoid misjudging the subsurface phytoplankton layer. Through detailed data preprocessing, profile parameter inversion and error analysis, this study provides a theoretical foundation for the development of future spaceborne oceanographic lidar data products.

The large signal dynamic range affecting the profile recognition of refined structures is one of the major challenges for future spaceborne oceanographic light detection and ranging (lidar) systems. Reduce the intensity of the sea surface signal and ensure that the detector operates in a linear response region, which helps reduce the subsurface signal error and improves the capability to detect weak signals in deep water. As a solution, the off-nadir pointing could reduce the photon counts from the sea surface but leads to depth aliasing. This reduces the vertical resolution and makes it difficult to determine the sea surface’s position and retrieve the thin chlorophyll layer. The lidar signal’s dynamic range is simulated to improve the detection accuracy. Based on the oceanographic lidar simulator, the laser transmission characteristics are analyzed, taking into account various different environmental parameters (including wind speed, sea surface roughness, concentration of whitecaps and bubbles) and lidar specifications (including laser off-nadir angle, divergence angle, and pulsewidth). The results show that increasing the off-nadir angle to 7°–15° can effectively reduce the dynamic range of the sea surface signal by about one order of magnitude, while increasing the aliasing depth by about 4–8 m. Reducing the beam divergence angle is beneficial for accurate inversion of profiles within the limits of engineering realization. Other parameters, such as pulsewidth, wind speed, and sea surface roughness, have little influence on depth aliasing and depth estimation errors.

With the intelligence and automation of industrial Internet of Things, a new collaborative Cloud-Fog Automation paradigm has emerged. The emergence of federated learning (FL) has further enhanced the capabilities of Cloud-Fog Automation, making it possible to develop more secure and versatile collaborative industrial models. However, FL faces various security risks. More importantly, the security risks faced by FL when applied in Cloud-Fog Automation, along with corresponding security measures, have not yet been explored. To address this issue, we make an initial attempt to analyze the security of FL within the context of Cloud-Fog Automation, with the aim of facilitating the design of a more secure FL framework for this paradigm. Specifically, we first analyze the security risks that may be encountered at different phases, then analyze the challenges that need to be faced to resolve these risks. Subsequently, we conduct a systematic review of the state-of-the-art security solutions, and finally summarize the future research directions.

Effective multi-intersection collaboration is crucial for mitigating urban traffic congestion through reinforcement learning (RL)-based traffic signal control (TSC). Existing work mainly considers scenarios involving a single vehicle type, where cooperation is typically limited to neighboring intersections. However, in urban traffic scenarios where high priority vehicles coexist with ordinary vehicles, considering only a limited number of neighboring nodes may be insufficient to ensure the swift passage of high priority vehicles while minimizing the impact on overall traffic efficiency. Therefore, we formulate the multiple intersections’ decision-making process in urban scenarios as a Markov game and propose a novel centralized cooperative RL framework called MATLIT to solve the game. Specifically, we adopt a multi-agent transformer (MAT)-based architecture that facilitates efficient global cooperation among intersections. The attention mechanism and auto-regressive process of the MAT effectively mitigate the curse of the dimensionality problem, which guarantees MATLIT to tackle large-scale traffic scenarios. Meanwhile, the stability and sequence action generation capacity of the MAT-based architecture is further enhanced by incorporating MAT with a gated mechanism. Furthermore, considering the inherent topological constraints in urban traffic scenarios, we utilize graph attention networks (GATs) to capture graph-structured mutual influences. Additionally, in response to the urban traffic scenarios with various types of high priority vehicles that have time-varying priorities, we integrate the soft actor-critic (SAC) algorithm to enhance the exploration capabilities of our framework, allowing it to learn robust strategies in heterogeneous traffic conditions. Extensive experiments demonstrate that our proposed MATLIT framework outperforms all baselines and can reduce high priority vehicles’ waiting time by 24.57% while reducing the average waiting time of all vehicles by 18.51% in realistic urban scenarios.

The development of surround-view cameras is crucial for the advancement of autonomous driving. Utilizing depth information and image features to simulate LiDAR bird’s-eye-view (BEV) features can accomplish efficient 3D object detection tasks. Existing dense BEV generation methods heavily rely on the use of depth features, however, the suboptimal exploitation of these features often results in ambiguity in object location and feature representation during the BEV generation process. To address this, we have designed a hybrid encode and distillation method to enhance 3D object detection performance, termed HybridBEV. Initially, we designed the HybridEncode module, which employs a resampling strategy of depth features in voxel space to obtain BEV features that more accurately reflect the distribution of objects. Subsequently, we introduced multiple distillation methods to supervise the network’s voxel features and BEV feature representations, assisting the student network in learning critical features from the teacher model and ensuring that BEV features can more distinctly represent object distribution. Furthermore, during network training, we loaded pre-trained weights from the teacher network to guide network optimization and accelerate training. Extensive experiments on the nuScenes benchmark demonstrate that HybridBEV can effectively improve the performance of the student network and outperform previous state-of-the-art methods based on surround-view cameras. The code will be published athttps://github.com/wjyxx/HybridBEV

In the landscape of automated warehousing, automated guided vehicles (AGVs) play a key role in enhancing operational accuracy and reducing labor costs. Optimizing the movement path of a group of AGVs, especially in time-sensitive environments, is crucial for maintaining efficiency and reducing costs. We describe this problem as timeliness-constrained multi-agent path finding (TC-MAPF). Although current multi-agent path finding (MAPF) approaches primarily focus on maximizing throughput, they typically overlook the critical need for adhering to stringent timeliness constraints associated with transportation tasks. Moreover, these methods depend heavily on carefully designed reward functions that are specific to particular environmental settings. This reliance constrains their flexibility and broad applicability. In this paper, we propose a novel MAPF algorithm to address the TC-MAPF problem, formulating it as a constrained Markov game. In this formulation, timeliness constraints are integrated into the reward function, ensuring that the agents prioritize meeting these critical deadlines during their operations. Additionally, we introduce a cooperative multi-agent intrinsic reward (CMAIR) framework to enhance the adaptability and generalization of the reward mechanisms across various dynamic environments. The CMAIR framework comprehensively considers the positions and actions of all agents, ensuring efficient exploration and policy optimization. We evaluate our method in various simulated warehouse scenarios and demonstrate that it significantly improves throughput, particularly under stringent time constraints compared to several existing MAPF methods.

Coordinating traffic lights and vehicles in urban environments is essential for optimizing road traffic efficiency and reducing congestion. Traditional traffic control strategies, which typically schedule traffic lights and vehicles independently, limit the potential for enhanced overall travel efficiency. In this study, we jointly control traffic lights and vehicle management to optimize traffic flow. Specifically, a dedicated signal control process at each intersection is managed by specialized agents, while vehicles within the vicinity are modeled as agents contributing to a unified optimization process. To facilitate effective collaboration among these heterogeneous agents, we propose a novel Heterogeneous Attention Reinforcement Learning (HARL) algorithm, where Graph Attention Networks (GAT) are designed to generate weighted vectors from traffic observations across intersections, accurately capturing and utilizing real-time traffic conditions. The integration of GAT improves the representation of traffic states in complex urban scenarios, thereby enhancing traffic management efficiency. We conduct extensive experiments on various real-world urban scenarios and the simulation results show that the proposed HARL significantly reduces the queue length by more than 16.13% compared to other methods.

With the capability of intelligent control and communicating with surrounding vehicles and infrastructures, connected and automated vehicles (CAVs) can drive cooperatively and have more positive effects on traffic efficiency. Cooperative and real-time path planning for CAVs stands as a pivotal solution to mitigate traffic congestion and augment travel efficiency. However, most of the existing path planning schemes predominantly concentrate on minimizing the travel times of vehicles, sidelining the broader issue of alleviating traffic congestion in urban settings. Therefore, in this paper, we propose a novel collaborative vehicle path planning scheme, leveraging the intelligent control and the communicating ability of CAVs. The primary objective is to reduce traffic congestion within the overall transportation system and improve traffic efficiency. Specifically, we focus on a general urban scenario with various types of vehicles, including CAVs, connected vehicles (CVs), and traditional human-driven vehicles (TVs), To enhance traffic efficiency in such a scenario, we design a collaborative path planning scheme to discover the efficient paths for both CAVs as well as CVs. In this scheme, we treat each CAV as an agent and formulate the multiple CAVs' path-planning problem as a Markov game. To solve the above Markov game, we design a multi-agent convolutional attention reinforcement learning (MACA) framework to generate paths with minimal travel time for CAVs. More concretely, the proposed MACA framework incorporates a convolutional neural network (CNN) layer to capture spatial correlation behind traffic conditions. Additionally, a graph attention network (GAT) layer is employed to integrate the influence of neighboring agents during the path-planning process. To further reduce traffic congestion, we extend the MACA framework into a collaborative MACA (C-MACA) scheme in vehicular networks, where CAVs are empowered to periodically broadcast their path information to surrounding CVs, providing valuable insights for their path planning. Subsequently, to prevent new congestion caused by the aggregation of CVs, we design a heuristic algorithm for CVs to make informed path decisions. We build up a simulator based on a real-world city road map and conduct extensive experiments. The experimental results demonstrate that the proposed scheme can decrease CVs' travel time by up to 10.9 % and reduce the average queue length around junctions by up to 6.5 % over several state-of-the-art approaches, without sacrificing the travel efficiency of CAVs.

In order to support diverse applications in intelligent transportation, intelligent connected vehicles (ICVs) need to send multiple types of messages, such as periodic messages and event-driven messages with different frame specifications. However, existing researches often concentrate on the transmission of single-message types, overlooking hybrid communication scenarios where multiple types of messages coexist, posing challenges in meeting the diverse transmission needs of different message types. To optimize the Quality of Service (QoS) in such scenarios, we take the perspective of ICVs and formulate their decision making as a multi-agent reinforcement learning problem. More specifically, we propose a cooperative individual rewards assisted multi-agent reinforcement learning (CIRA) framework. The transformer structure in CIRA is used to avoid mutual interference during the transmission of different vehicles. Besides, the introduction of individual rewards and the dual-layer architecture of CIRA contribute to providing ICVs with more forward-looking message dissemination scheme. Finally, we set up a simulator to create dynamic traffic scenarios reflecting different real-world conditions. We conduct extensive experiments to evaluate the proposed CIRA framework’s performance. The results show that CIRA can significantly improve the packet reception rates and ensure low communication delays in various scenarios.

Mobile edge computing (MEC) is expected to support real-time services at wireless networks, where task replication is applied to guarantee job completion within a strict deadline through replicating multiple copies to different edge servers. Most of previous works focused on guaranteeing the reliability of individual task in MEC-based networks with the assumption of homogeneous task execution distribution. Further, these algorithms cannot suit dynamic network scales, due to overhigh communication or retraining overhead. Therefore, this paper formulates the problem of heterogeneous task replication in a finer level by modeling outage probability of individual replication, where the decisions of all tasks are jointly optimized within the constraints of both mobile users and MEC servers for minimizing job outage probability. To adapt to varying network scales, we develop centralized and distributed algorithms, respectively. The centralized algorithm is developed based on Interior Point Method, which obtains the optimal solution of relaxed model and then approximates to the solution of original problem. Further, the distributed algorithm decomposes the HTR into multiple subproblems and parallelly compute each local solution based on Distributed ADMM. Finally, we build a simulation model and conduct comprehensive results, which demonstrates that the proposed algorithms can achieve high-accuracy solution with fast convergence.

A group ofconnected and autonomous vehicleswith common interests can drive in a cooperative manner, namely cooperative driving. In such a networked control system, an efficient message dissemination scheme is critical for cooperative drivings to periodically broadcast their kinetic status, i.e.,beacon. However, most existing researches are designed for a simple or specific scenario, e.g., ignoring the impacts of the complex communication environment and emerging hybrid traffic scenarios. Worse still, the inevitable message transmission interference and the limited interaction among vehicles in harsh communication environments seriously hinder cooperation among cooperative drivings and deteriorate the beaconing performance. In this paper, we formulate the decision-making process of cooperative drivings as a Markov game. Furthermore, we propose acooperative hierarchical attention reinforcement learning (CHA)framework to solve this Markov game. Specifically, the hierarchical structure of CHA leads cooperative drivings to be foresighted. Besides, we integrate each hierarchical level of CHA separately with graph attention networks to incorporate agents' mutual influences in the decision-making process. Moreover, each hierarchical level learns a cooperative reward function to motivate each agent to cooperate with others under harsh communication conditions. Finally, we set up a simulator and conduct extensive experiments to validate the effectiveness of CHA.

In real-world urban environments, hybrid and disorder traffic brings new challenges for the intelligent traffic light control system (ITLCS). Apart from coordinating traffic flows around intersections, the ITLCS is responsive to ensuring high priority vehicles pass through intersections quickly. To this end, we formulate the multiple intersections’ decision-making problem as a Semi-Markov game and propose amulti-agent attention double actor-critic (MAADAC)framework to solve this game, integrating theoptions frameworkwithgraph attention networks (GATs). Specifically, the options framework empowers agents to learn to make a long sequence of satisfactory decisions, such as keeping a reasonable phase for a short period to ensure high priority vehicles pass through intersections quickly. Besides, we adopt GATs to capture graph-structure mutual influences among agents. We set up a simulator based on real-world city road networks and conduct extensive experiments to evaluate the performance of MAADAC. The experimental results show that MAADAC can reduce high priority vehicles’ waiting time in the interval of 18.16%-38.14% versus the density of vehicles in real-world urban scenarios over several state-of-the-art approaches. Also, our framework can guarantee the passing efficiency of high priority vehicles under various traffic conditions with the change in the proportion of high priority vehicles.

Since emergency vehicles (EVs) are essential for urban emergency response, it is essential to help EVs arrive faster. Existing work has investigated route optimization or traffic signal preemption, but they are insufficient because most studies consider the two areas separately and lack a deep understanding of their relationship. For instance, traffic signal preemption can cause changes in traffic flow, affecting the optimal route for EVs. Moreover, previous work does not distinguish between EV types or consider the negative impact on ordinary vehicles (OVs). To address these issues, we propose a framework that jointly considers priority allocation, routing optimization, and traffic signal preemption (EVPRT) in conjunction with the Vehicle to Everything (V2X) environment. To this end, we design an emergency vehicle priority system (EVPS) to assign priorities to different EV types. Then, we design a dynamic route optimization method to update the optimal routes for EVs. Finally, we design a multi-agent reinforcement learning (MARL) based traffic signal preemption algorithm and use a Graph Attention Network (GAT) to extract potential features of different intersections. Furthermore, to provide communication conditions for EVPRT, we adopt a V2X-based communication technology for information interaction. The simulation findings show that our proposed method significantly decreases EV travel time and enhances the capacity of urban emergency service management.

In Green Vehicular Ad-hoc Networks (VANET), how to establish and maintain stable routes to ensure fast and efficient transmission of messages is quite an important and demanding task because of the intrinsic characteristics of VANET, such as uneven distribution and high mobility of vehicles. In this paper, we propose a novel vehicle density prediction-based routing protocol assisted by Green Vehicular Fog Computing (VFC) architecture named VVDP. Since buses have specific trajectories and departure intervals, they are considered fog nodes to improve fog coverage and transmission efficiency. Specifically, we first introduce a vehicle density prediction model to capture the spatio-temporal features of vehicle density effectively. Moreover, we divide the map into streets. The cloud layer of VFC performs precise vehicle density prediction and comprehensively considers the density of buses and common vehicles to assign weights to each street, which can reduce transmission failures due to uneven distribution of vehicles. Besides, buses are used to provide a street-based global path for messages assisted by the weights of each street. To enhance the transmission efficiency, we propose a novel concept: the link trust factor, which combines the mobility factor and direction factor as metrics for selecting the optimal relay. Overall, VVDP not only fully utilizes the rule of vehicle density but also allows for flexible adjustment in accordance with the current circumstance. Simulation results reveal that our proposed routing scheme outperforms others in terms of delivery ratio and end-to-end delay.

The development of autonomous driving has led to the proposal of vehicle-to-everything (V2X) to improve the reliability of autonomous driving systems through information sharing among vehicles and infrastructure. However, the high-speed mobility of autonomous vehicles and the dynamic surrounding environment make the V2X network unstable and unreliable. To address this issue, empirically studying the characteristics and predicting the signal intensity of the V2X network is crucial, which can provide more information for further optimizing communication strategies and enhancing driving safety. In this paper, we collect the real-world vehicle-to-infrastructure (V2I) performance under different driving scenarios and build the quantitative relationship between several environmental factors and the V2I performance. We also develop deep learning models to predict the V2I signal intensity based on external environmental conditions. Experimental results in real-world data demonstrate the effectiveness of our model on classification and regression tasks compared to other models. Our study aims to enhance the applications of V2X on autonomous driving and improve driving safety and traffic efficiency.

To unleash the power of buses, we propose a bus-pooling system that keeps the notion of bus stops and terminals but discards the concept of fixed bus lines by enabling buses to choose the next stop or terminal based on orders submitted by passengers. Each bus, unlike a taxi, must consider the additional delays experienced by the passengers already on board when deciding how to adapt its route to serve new orders. This paper treats each bus as an agent and formulates the buses’ re-routing decision-making process as a Semi-Markov game. Then, we propose a novel double graph attention actor-critic (DGAAC) framework by integrating high-level and low-level actor-critics separately with graph attention networks (GATs) to solve the game. Specifically, GATs embedded in high-level and low-level critics take a large-scale graph covering a city-scale area as input and capture graph-structured mutual influences among buses. In contrast, the high-level and low-level actors equipped with GATs only take the n-hop sub-graph with local information as the input and are employed as the distributed decision module of each bus. We conduct extensive experiments on one of the largest real-world datasets in Shenzhen, China, and validate that the proposed DGAAC framework greatly outperforms all baselines.

Vehicular Ad-Hoc Network (VANET) is currently experiencing a critical technological transformation as more and more vehicles move to a higher level of automation. To cope with increasingly complex traffic conditions, automated vehicles need to maintain regular communication with each other. This highly dynamic topology structure poses significant challenges to routing protocols. This paper proposes a vehicle distribution prediction-based routing protocol called VDP. The protocol divides the map into grids, analyzes the role of different areas in a single grid by simulating the communication process between adjacent grids, and uses the neural network model to predict the distribution of vehicles in the grid. We combine the prediction results with the complexity of the urban environment to arrive at the optimal inter-grid path, which is then used for grid selection. Moreover, a grid-based routing method is proposed to select the optimal relay node according to real-time traffic information. All in all, VDP not only makes full use of the law of vehicle distribution but also can be flexibly adjusted according to the current actual situation. We have conducted extensive simulation evaluations to evaluate the performance of VDP under different prediction models. The experimental results on an accurate road map show that our method is superior to the existing position-based routing protocols.

Vehicular communication networks (VCNs) have been widely recognized as promising solutions to support safety-related applications in urban transportation systems. However, constructing and maintaining such networks is quite challenging due to the complex traffic and communication environment. Substantial studies have focused on the design of the networking schemes and message dissemination protocols. Nonetheless, most existing designs only consider the connectivity and rapid end-to-end transmission, regardless of the network coverage and duration. In this article, we propose a novel temporary warning network (TWN) for safety message dissemination in the urban traffic environment, in which both the spatial distribution and temporal duration of the networking scheme are taken into account. Specifically, TWN is constructed by the selection of relay vehicles based on the spatiotemporal correlation of vehicle trajectory so that the safety message can be quickly disseminated within the Regions of Interest (RoIs). To maintain TWN during an accident, a reselection mechanism is also proposed, which enables newly come vehicles in the RoI to receive the messages in time. Finally, we conduct extensive numerical experiments to validate the effectiveness of our method in various traffic scenarios.

A group of connected and autonomous vehicles (CAVs) with common interests can drive in a cooperative manner, namely cooperative driving, which has been verified to significantly improve road safety, traffic efficiency, and environmental sustainability. A more general scenario with various types of cooperative driving applications such as truck platooning and vehicle clustering will coexist on roads in the foreseeable future. To support such multiple cooperative drivings, it is critical to design an efficient message dissemination scheduling for vehicles to broadcast their kinetic status, i.e., beacon periodically. Most ongoing researches suggest designing the communication protocols via traffic and communication modeling on top of dedicated short range communications (DSRC) or cellular-based vehicle-to-vehicle (C-V2V) communications as a potential remedy. However, most of the existing researches are designed for a simple or specific traffic scenario, e.g., ignoring the impacts of the complex communication environment and emerging hybrid traffic scenarios. Moreover, some studies design beaconing strategies based on the implication of channel and traffic conditions in the beacons of other vehicles. However, the delayed perception of these information may seriously deteriorate the beaconing performance. In this paper, we take the perspective of cooperative drivings and formulate their decision-making process as a Markov game. Furthermore, we propose a multi-agent hierarchical attention reinforcement learning (MAHA) framework to solve the Markov game. More concretely, the hierarchical structure of the proposed MAHA can lead cooperative drivings to be foresightful. Hence, even without immediate incentives, the well-trained agents can still take favorable actions that benefit their long-term rewards. Besides, we integrate each hierarchical level of MAHA separately with the graph attention network (GAT) to incorporate agents' mutual influences in the decision-making process. Besides, we set up a simulator and adopt this simulator to generate dynamic traffic scenarios, which reflect the different real-world scenarios faced by cooperative drivings. We conduct extensive experiments to evaluate the proposed MAHA framework's performance. The results show that MAHA can significantly improve the beacon reception rate and guarantee low communication delay in all of these scenarios.

Vehicular ad hoc networks (VANETs) have been widely recognized as a promising solution to improve traffic safety and efficiency for the ability to provide situation awareness even though the potential dangers and traffic anomalies are out of the visual range. In VANETs, time-division multiple access (TDMA) based overlay protocols can prevent transmission collisions, and play an important role in providing an efficient communication channel. However, due to high vehicle mobility and time-varying traffic flow, the existing TDMA-based slot allocation approaches cannot fully utilize the channel resources, which may result in high transmission delay and packet collision. To overcome these shortcomings, we propose a traffic-aware TDMA-based MAC (TA-MAC) protocol which utilizes the capability of mobile edge computing (MEC) in this paper. Specifically, based on MEC and vehicle-to-road-side-units (V2R) communications, a traffic-aware mechanism is first proposed to estimate the traffic condition on the road segment. Then, we propose a new slot assignment method that aims at guaranteeing the high channel utilization and low delay of safety message under dynamic traffic conditions. Finally, we conduct extensive experiments to demonstrate the effectiveness of the proposed protocol.

A group of connected and autonomous vehicles (CAVs) with common interests can drive in a cooperative manner, namely cooperative driving, which has been verified to significantly improve road safety, traffic efficiency and environmental sustainability. A more general scenario that various types of cooperative driving applications such as truck platooning and vehicle clustering, will coexist on roads in the foreseeable future. To support such multiple cooperative drivings, it is critical to design an efficient message dissemination scheduling in a shared communication channel. Most ongoing research suggests using the time-division multiple access (TDMA) method on top of IEEE 802.11p as a potential remedy. However, TDMA requires time synchronization and is not flexible, especially in the multiple cooperative drivings scenario where the beacon frequency needs to be updated and the number of cooperative drivings changes to meet the time-varying traffic conditions. In this paper, we focus on the study of the message dissemination protocol for platooning, a typical and well-known cooperative driving pattern. Specifically, we proposed a hybrid message dissemination protocol which aims at guaranteeing the reliable delivery of beacon messages for a multi-platooning system. We first adopt a TDMA-based medium access method for intra-platoon communication to improve the reliability and efficiency of beacon dissemination. We then present a token-passing medium access method for inter-platoon communication, which maps platoons into a token ring to schedule their beacon transmission time. We conduct extensive numerical experiments to validate the effectiveness of our protocol.

Vehicular communication networking (VCN) has been widely recognized as a promising solution to support safety-related applications in urban transportation systems. In VCN, efficient message dissemination can let vehicles be better aware of the potential risks and traffic anomalies, which is critical to road safety and traffic efficiency. Substantial studies have focused on the design of inter-vehicle message dissemination protocols. Nonetheless, most existing designs only consider the rapid end-to-end transmission, few of which take into account the broadcast coverage. In this paper, we propose a new message dissemination scheme in the urban traffic scenario by considering both the time constraint and the spatial distribution of data dissemination. Specifically, based on the temporal and spatial correlation of vehicle trajectory, relay vehicles are selected to construct a temporary warning network (TWN) for a rapid safety message dissemination in the regions of interest (ROI). Finally, we conduct extensive numerical experiments to validate the effectiveness of our method in various traffic scenarios.

With the advances of Internet of Things technologies, individual vehicles can now exchange information to improve traffic safety, and some vehicles can further improve safety and efficiency by coordinating their mobility via cooperative driving. To facilitate these applications, many studies have been focused on the design of inter-vehicle message dissemination protocols. However, most existing designs either assume individual driving pattern or consider cooperative driving only. Moreover, few of them fully exploit infrastructures, such as cameras, sensors, and road-side units. In this paper, we address the design of message dissemination that supports heterogeneous driving patterns. Specifically, we first propose an infrastructure-assisted message dissemination framework that can utilize the capability of infrastructures. We then present a novel beacon scheduling algorithm that aims at guaranteeing the timely and reliable delivery of both periodic beacon messages for cooperative driving and event-triggered safety messages for individual driving. To evaluate the performance of the protocol, we develop both theoretical analysis and simulation experiments. Extensive numerical results confirm the effectiveness of the proposed protocol.