Search Results (855)

Signed networks, which incorporate both cooperative and antagonistic interactions, naturally give rise to symmetric behaviors in multi-agent systems. One such behavior is bipartite containment tracking, where follower agents converge to a symmetric configuration determined by multiple groups of leaders with opposing influence. Moreover, a timely response is critical to ensuring high performance in containment tracking tasks, particularly for high-order multi-agent systems operating in dynamic and uncertain environments. To this end, this paper investigates the predefined-time bipartite containment tracking problem for high-order multi-agent systems affected by external disturbances. A robust tracking control scheme is developed based on the backstepping method to ensure that the tracking errors converge to a predefined residual set within a user-specified time. The convergence time is explicitly adjustable through a design parameter, and the proposed scheme effectively avoids the singularities often encountered in conventional predefined-time control approaches. The stability and robustness of the proposed scheme are rigorously established through Lyapunov-based analysis, and extensive simulation results are provided to validate our theoretical findings. Full article

Distributed target tracking in multi-agent networks plays a critical role in cooperative sensing and autonomous navigation. However, it faces significant challenges in highly dynamic and adversarial setups. This study aims to enhance the resilience of decentralized target tracking algorithms against measurement faults and cyber–physical threats, especially false data injection attacks. We propose a consensus-based estimation algorithm that integrates a nearly constant velocity model with saturation-based filtering to suppress impulsive measurement variations and promote robust, distributed state estimation. To counteract adversarial conditions, we incorporate a dynamic false data injection detection and isolation mechanism that uses innovation thresholds to identify and disregard suspicious measurements before they can degrade the global estimate. The effectiveness of the proposed algorithms is demonstrated through a series of simulation-based case studies under both benign and adversarial conditions. The results show that increased network connectivity and higher consensus iteration rates improve estimation accuracy and convergence speed, while properly tuned saturation filters achieve a practical balance between fault suppression and accurate estimation. Furthermore, under localized, coordinated, and transient false data injection attacks, the detection mechanism successfully identifies compromised agents and prevents their data from corrupting the distributed global estimate. Overall, this study illustrates that the proposed algorithm provides a simplified fault-tolerant solution that significantly enhances the accuracy and resilience of distributed target tracking without imposing excessive communication or computational burdens. Full article

This paper examines the evolving role of social economy organisations (SEOs) in advancing sustainability and contributing to the green transition. While traditionally focused on social inclusion and local development, SEOs are increasingly integrating environmental objectives into their operations, particularly through circular economy (CE) practices, decarbonisation strategies, and short food supply chains (SFSCs). Based on qualitative research and the analysis of 16 good practices from five European countries, the study demonstrates how SEOs create blended social and environmental value by combining economic, social, and ecological goals. The findings show that SEOs foster environmental sustainability by reducing resource consumption and carbon emissions, creating green jobs, strengthening local cooperation, and raising environmental awareness within communities. Importantly, SEOs emerge not only as service providers but also as innovators and agents of change in local ecosystems. The paper concludes with policy recommendations to enhance the role of SEOs in the green transition and identifies directions for future research, particularly regarding the measurement of their long-term environmental impact and the conditions enabling effective collaboration with public and private sector actors. Full article

The strategic development of the low-altitude economy requires efficient urban logistics solutions. The existing Unmanned Aerial Vehicle (UAV) truck delivery system faces severe challenges in dealing with dynamic weather constraints and multi-agent coordination. This article proposes a neural network-based optimisation framework that integrates constrained K-means clustering and a three-stage neural architecture. In this work, a mathematical model for heterogeneous vehicle constraints considering time windows and UAV energy consumption is developed, and it is validated through reference to the Solomon benchmark’s arithmetic examples. Experimental results show that the Truck–Drone Cooperative Traveling Salesman Problem (TDCTSP) model reduces the cost by 21.3% and the delivery time variance by 18.7% compared to the truck-only solution (Truck Traveling Salesman Problem (TTSP)). Improved neural network (INN) algorithms are also superior to the traditional genetic algorithm (GA) and Adaptive Large Neighborhood Search (ALNS) methods in terms of the quality of computed solutions. This research provides an adaptive solution for intelligent low-altitude logistics, which provides a theoretical basis and practical tools for the development of urban air traffic under environmental uncertainty. Full article

To improve the traffic efficiency of connected and automated vehicles (CAVs) in on-ramp merging areas, this study proposes a novel region-level multi-agent reinforcement learning framework, Regionally Coordinated Nash-Advantage Decomposition Deep Q-Network with Conflict-Aware Q Fusion (RC-NashAD-DQN). Unlike existing vehicle-level control methods, which suffer from high computational overhead and poor scalability, our approach abstracts on-ramp and main road areas as region-level control agents, achieving coordinated yet independent decision-making while maintaining control precision and merging efficiency comparable to fine-grained vehicle-level approaches. Each agent adopts a value–advantage decomposition architecture to enhance policy stability and distinguish action values, while sharing state–action information to improve inter-agent awareness. A Nash equilibrium solver is applied to derive joint strategies, and a conflict-aware Q-fusion mechanism is introduced as a regularization term rather than a direct action-selection tool, enabling the system to resolve local conflicts—particularly at region boundaries—without compromising global coordination. This design reduces training complexity, accelerates convergence, and improves robustness against communication imperfections. The framework is evaluated using the SUMO simulator at the Taishan Road interchange on the S1 Yongtaiwen Expressway under heterogeneous traffic conditions involving both passenger cars and container trucks, and is compared with baseline models including C-DRL-VSL and MADDPG. Extensive simulations demonstrate that RC-NashAD-DQN significantly improves average traffic speed by 17.07% and reduces average delay by 12.68 s, outperforming all baselines in efficiency metrics while maintaining robust convergence performance. These improvements enhance cooperation and merging efficiency among vehicles, contributing to sustainable urban mobility and the advancement of intelligent transportation systems. Full article

As network environments become increasingly dynamic and users’ Quality of Service (QoS) demands grow more diverse, efficient and adaptive routing strategies are urgently needed. However, traditional routing strategies suffer from limitations such as poor adaptability to fluctuating traffic, lack of differentiated service handling, and slow convergence in complex network scenarios. To this end, we propose a routing strategy based on multi-agent deep deterministic policy gradient for differentiated QoS services (RS-MADDPG) in a software-defined networking (SDN) environment. First, network state information is collected in real time and transmitted to the control layer for processing. Then, the processed information is forwarded to the intelligent layer. In this layer, multiple agents cooperate during training to learn routing policies that adapt to dynamic network conditions. Finally, the learned policies enable agents to perform adaptive routing decisions that explicitly address differentiated QoS requirements by incorporating a custom reward structure that dynamically balances throughput, delay, and packet loss according to traffic type. Simulation results demonstrate that RS-MADDPG achieves convergence approximately 30 training cycles earlier than baseline methods, while improving average throughput by 3%, reducing latency by 7%, and lowering packet loss rate by 2%. Full article

This theoretical article explores the contrasting ontologies, axiologies, and political economies of transhumanism and posthumanism. Transhumanism envisions the human as an enhanced, autonomous agent shaped by neoliberal and Enlightenment ideals. Posthumanism challenges this by emphasizing relationality, ecological entanglement, and critiques of commodification. Both engage with technology’s role in reshaping humanity. Drawing on Braidotti’s posthumanism, Haraway’s cyborg figuration, Ahmed’s politics of emotion, Berlant’s cruel optimism, Massumi’s affective modulation, Seigworth and Gregg’s affective intensities, Zuboff’s surveillance capitalism, Fisher’s capitalist realism, Cooper’s surplus life, Sadowski’s digital capitalism, Lupton’s quantified self, Schafheitle et al.’s datafied subject, Pasquale’s black box society, Terranova’s network culture, Bratton’s platform sovereignty, Dean’s communicative capitalism, and Morozov’s technological solutionism, the article elucidates how subjectivity, data, and infrastructure are reorganized by corporate systems. Introducing technogensis as the co-creation of human and technological subjectivities, it links corporate-platform practices to future trajectories governed by Apple, Meta, and Google. These branded technologies function not only as enhancements but as infrastructures of governance that commodify subjectivity, regulate affect and behavior, and reproduce socio-economic stratification. A future is extrapolated where humans are not liberated by technology but incubated, intubated, and ventilated by techno-conglomerate governments. These attention-monopolizing, affective-capturing, behavior-modulating, and profit-extracting platforms do more than enhance; they brand subjectivity, rendering existence subscription-based under the guise of personal optimization and freedom. This reframes transhumanism as a cybernetic intensification of liberal subjectivity, offering tools to interrogate governance, equity, agency, and democratic participation, and resist techno-utopian narratives. Building on this, a posthumanist alternative emphasizes relational, multispecies subjectivities, collective agency, and ecological accountability, outlining pathways for ethical design and participatory governance to resist neoliberal commodification and foster emergent, open-ended techno-social futures. Full article

For multiple unmanned underwater vehicles (UUVs) systems, obstacle avoidance during cooperative operation in complex marine environments remains a challenging issue. Recent studies demonstrate the effectiveness of deep reinforcement learning (DRL) for obstacle avoidance in unknown marine environments. However, existing methods struggle in marine environments with complex non-convex obstacles, especially during multi-UUV cooperative operation, as they typically simplify environmental obstacles to convex shapes with sparse distributions and ignore the dynamic coupling between cooperative operation and collision avoidance. To address these limitations, we propose a centralized training with decentralized execution framework with a novel multi-agent dynamic encoder based on an efficient self-attention mechanism. The framework, to our knowledge, is the first to dynamically process observations from an arbitrary number of neighbors that effectively addresses multi-UUV collision avoidance in marine environments with complex non-convex obstacles while satisfying additional constraints derived from cooperative operation. Experimental results show that the proposed method effectively avoids obstacles and satisfies cooperative constraints in both simulated and real-world scenarios with complex non-convex obstacles. Our method outperforms typical collision avoidance baselines and enables policy transfer from simulation to real-world scenarios without additional training, demonstrating practical application potential. Full article

Background: The treatment landscape for advanced hepatocellular carcinoma (HCC) has evolved significantly recently; however, access to novel agents remains limited because of high costs. This study aimed to evaluate the systemic treatment patterns and survival outcomes for advanced HCC across different systemic treatment sequences under real-world resource constraints. Methods: This retrospective study was conducted at a tertiary center in Southern Thailand. The medical records of patients (n = 330) with advanced HCC treated with systemic therapy between 2010 and 2024 were reviewed. Outcomes included overall survival (OS), progression-free survival (PFS), and objective response rate (ORR). Prognostic factors for OS were investigated. Results: First-line therapies included tyrosine kinase inhibitor (TKI; 69.7%), chemotherapy (23.3%), immunotherapy (IO)/targeted therapy (3.6%), dual IO (1.8%), and IO monotherapy (1.5%). The median OS, PFS, and ORR for each cohort were 7.2, 5.2, 10.9, 8.5, and 8.6 months; 3.94, 3.22, 3.48, 6.19, and 2.69 months; and 9.6%, 10.4%, 16.7%, 0%, and 20.0%, respectively. OS improved with increasing lines of therapy (4.5, 12.2, 19.4, and 40.7 months for one to four lines, respectively). Portal vein tumor thrombus, ascites, elevated bilirubin level, high alpha-fetoprotein level, and poor Eastern Cooperative Oncology Group performance status were associated with poor prognosis; multiple treatment lines and overweight status were associated with improved OS. Conclusions: In this large real-world cohort, TKIs remained the mainstay effective treatment option because of limited access to IO-based regimens. Sequential systemic therapy significantly improved survival, emphasizing the importance of preserving treatment eligibility and multidisciplinary team involvement. Chemotherapy could be considered a viable option in resource-limited settings. Full article

Coordinating Multi-Unmanned Surface Vehicle (USV) swarms in complex, adversarial maritime environments is a significant challenge, as existing multi-agent reinforcement learning (MARL) methods often fail to capture intricate spatiotemporal dependencies, leading to suboptimal policies. To address this, we propose Adv-TransAC, a novel Spatio-Temporal Meta-Reinforcement Learning framework. Its core innovation is a hybrid GAT-transformer architecture that decouples spatial and temporal reasoning: a Graph Attention Network (GAT) models instantaneous tactical formations, while a transformer analyzes their temporal evolution to infer intent. This is combined with an adversarial meta-learning mechanism to enable rapid adaptation to opponent tactics. In high-fidelity escort and defense simulations, Adv-TransAC significantly outperforms state-of-the-art MARL baselines in task success rate and policy robustness. The learned policies demonstrate the emergence of complex cooperative behaviors, such as intelligent risk-aware coordination and proactive interception maneuvers. The framework’s practicality is further validated by a communication-efficient federated optimization architecture. By effectively modeling spatiotemporal dynamics and enabling rapid adaptation, Adv-TransAC provides a powerful solution that moves beyond reactive decision-making, establishing a strong foundation for next-generation, intelligent maritime platforms. Full article

Against the backdrop of global low-carbon transition, the integrated development of electricity and carbon markets demands higher efficiency in the optimal operation of virtual power plants (VPPs) and distribution networks, yet conventional trading mechanisms face limitations such as inadequate recognition of differentiated contributions and inequitable benefit allocation. To address these challenges, this paper proposes a collaborative optimal trading mechanism for VPP clusters and distribution networks in an electricity–carbon coupled market environment by first establishing a joint operation framework to systematically coordinate multi-agent interactions, then developing a bi-level optimization model where the upper level formulates peer-to-peer (P2P) trading plans for electrical energy and carbon allowances through cooperative gaming among VPPs while the lower level optimizes distribution network power flow and feeds back the electro-carbon comprehensive price (EACP). By introducing an asymmetric Nash bargaining model for fair benefit distribution and employing the Alternating Direction Method of Multipliers (ADMM) for efficient computation, case studies demonstrate that the proposed method overcomes traditional models’ shortcomings in contribution evaluation and profit allocation, achieving 2794.8 units in cost savings for VPP clusters while enhancing cooperation stability and ensuring secure, economical distribution network operation, thereby providing a universal technical pathway for the synergistic advancement of global electricity and carbon markets. Full article

As a crucial vehicle for green technological innovation, cooperative networks significantly promote resource integration and knowledge sharing. Yet, their dynamic evolution and micro-mechanism remain underexplored. Drawing on data from the joint applications of green invention patents between 2006 and 2021, this study constructed a multi-agent GTCIN involving multiple stakeholders, such as enterprises, universities, and research institutions, and analyzed the topological structure and evolutionary characteristics of this network; an exponential random graph model (ERGM) was introduced to elucidate its endogenous and exogenous driving mechanisms. The results indicate that while innovation connections increased significantly, the connection density decreased. The network evolved from a “loose homogeneity” to “core aggregation” and then to “outward diffusion”. State-owned enterprises in the power industry and well-known universities are located at the core of the network. Preferential attachment and transitive closure as endogenous mechanisms exert strong and continuous positive effects by reinforcing local clustering and cumulative growth. The effects of exogenous forces exhibit stage-specific characteristics. State ownership and regional location become significant positive drivers only in the mid-to-late stages. The impact of green innovation capability is nonlinear, initially promoting but later exhibiting a significant inhibitory effect. In contrast, green knowledge diversity exerts an opposite pattern, having a negative effect in the early stage due to integration difficulties that turns positive as technical standards mature. Geographical, technological, social, and institutional proximity all have a positive promoting effect on network evolution, with technological proximity being the most influential. However, organizational proximity exerts a significant inhibitory effect in the later stages of GTCIN evolution. This study reveals the shifting influence of endogenous and exogenous mechanisms across different evolutionary phases, providing theoretical and empirical insights into the formation and development of green innovation networks. Full article

Green product innovation in aluminum building material supply chains is critical for sustainability, particularly amid growing economic and environmental pressures. However, effective coordination is challenged by the presence of multiple agents with divergent interests and heterogeneous innovation capacities. This study proposes coordination mechanisms based on a biform game that integrates both non-cooperative and cooperative elements. Key findings include the following: (1) Greater innovation capability heterogeneity promotes green innovation investment by the stronger manufacturer and enhances overall welfare, but reduce the supplier’s profit. (2) Biform game-based decision making supports the triple bottom line more effectively than decentralized models and offers greater flexibility than centralized ones. (3) A multi-perspective compensation contract, incorporating three decision-making modes, is developed within the biform game. Exogenous decision making helps resolve the endogenous game dilemma, improving coordination outcomes. (4) The coordination framework allows firms to dynamically adjust compensation parameters in response to environmental changes, thereby enhancing supply chain resilience. Our main contribution lies in applying a novel biform game approach to address coordination challenges in green product innovation under innovation capability heterogeneity. In addition, a multi-perspective contract coordination paradigm is proposed to support triple bottom line sustainability. Full article

In swarm robotics, collective transport refers to the cooperative movement of a large object by multiple small robots, each with limited individual capabilities such as sensing, mobility, and communication. When working together, however, these simple agents can achieve complex tasks. This study explores a collective transport method based on the caging approach, which involves surrounding the object in a way that restricts its movement while still allowing limited motion, effectively preventing escape from the robot formation. The proposed approach is structured into four main phases: locating the object, recruiting additional robots, forming an initial cage around the object, and finally, performing the transportation. The method is tested using simulations in the CoppeliaSim environment, employing a team of Khepera-III robots. Performance metrics include execution time for the search and recruitment phases, and both execution time and trajectory accuracy, via a normalized error, for the transport phase. To further validate the method, a comparison is made between the caging-based strategy and a traditional pushing strategy. Full article

The paradigm of reinforcement learning (RL) refers to agents that learn iteratively through continuous interactions with their environment. However, when the value function is unknown, a neural network is used, which is typically encoded into an unknown temporal difference equation. When RL is implemented in physical systems, explicit convergence and stability analyses are required to guarantee the worst-case operations for any trial, even when the initial conditions are set to zero. In this paper, physical RL (p-RL) refers to the application of RL in dynamical systems that interact with their environments, such as robot manipulators in contact tasks and humanoid robots in cooperation or interaction tasks. Unfortunately, most p-RL schemes lack stability properties, which can even be dangerous for specific robot applications, such as those involving contact (constrained) tasks or interaction tasks. Considering an unknown and disturbed DAE2 robot, in this paper a p-RL approach is developed to guaranteeing robust stability throughout a continuous-time-adaptive actor–critic, with local exponential convergence of force–position tracking error. The novel adaptive mechanisms lead to robustness, while an integral sliding mode enforces tracking. Simulations are presented and discussed to show our proposal’s effectiveness, and some final remarks are addressed concerning the structural aspects. Full article