LUCID: An Agentic AI Framework on Digital-Twin in the Loop for QoS-Guaranteeing Robotic Control

Multi-Agent Autonomous 6G Service Control With Intelligent Reconfiguration

Future 6G services will require strict performance guarantees, especially in terms of delay, end-to-end (e2e) across multiple network domains including packet and radio segments. While deterministic transport and slice-based capacity allocation can improve segment-level performance, ensuring e2e Network Service (NS) performance remains challenging as it requires making decisions Near–Real-Time (Near-RT) on a per-service basis, which does not fit well within the typical centralized control and orchestration hierarchy. Multi-agent systems (MAS), where a number of distributed agents collaborate, has demonstrated its capabilities for such Near-RT control. Agents equipped with Deep Reinforcement Learning (DRL) engines autonomously made traffic routing decisions based on e2e telemetry measurements. In this paper, we extend such MAS solutions for NS traffic routing focused on covering several issues that appear under frequent NS reconfiguration, e.g., caused by end device mobility. In addition, we define a lifecycle for NS operation that includes the initial MAS deployment, model reconfiguration during operation, and NS reconfiguration. The proposed lifecycle requires the definition of DRL training and validation procedures to produce models ready to be deployed with guaranteed performance under certain network conditions. In addition, model selection algorithms are defined for the lifecycle scenarios. In case of NS reconfiguration, a procedure for probe testing the actual network conditions is proposed to improve model selection. Evaluation across a meaningful set of network and traffic scenarios shows that the MAS is able to maintain e2e delay guarantees under all the lifecycle scenarios.

Intelligent Multi-UAV Navigation in ITNTNs: A Hierarchical LLM Approach

The deployment of high-speed Uncrewed Aerial Vehicles (UAVs) in 3D aerial highways necessitates robust coordination of physical flight kinematics and multi-tier network handovers. While Deep Reinforcement Learning (DRL) offers rapid tactical control, it lacks the zero-shot strategic reasoning required to quickly adapt to dynamic Integrated Terrestrial and Non-Terrestrial Networks (ITNTNs). Conversely, Large Language Models (LLMs) excel at semantic reasoning but suffer from high inference latency, rendering them unsuitable for real-time aerodynamic control. To bridge this gap, we propose a novel Hierarchical LLM-driven control framework. A massive cloud-based LLM deployed on a High-Altitude Platform Station (HAPS) manages slow-timescale global load balancing, while lightweight edge-LLMs on individual UAVs translate local observations into tactical sub-goals. These sub-goals guide a fast-timescale physical DRL controller to execute collision-free, handover-aware trajectories. Simulation results demonstrate that our agentic architecture significantly reduces collision rates and improves aggregate system throughput compared to existing baselines.

Intent-Driven Task Coordination in Low-Altitude IoT Networks: Autonomous Cognitive Agents via LLMs and Distributed Active Inference

Low-altitude Internet of Things (IoT) networks are emerging as an important platform for real-time monitoring, aerial logistics, and other distributed intelligent services. However, as missions become more complex and less structured, manual task decomposition and predefined coordination strategies no longer scale, leading to inefficiencies, long delays, and limited real-time adaptability. At the same time, existing deep reinforcement learning (DRL) methods rely on fixed reward formulations, which often incur prohibitive retraining costs when mission objectives or network conditions change. Addressing these challenges requires a paradigm that can transform high-level human intent into efficient multiagent task coordination and offloading under dynamic, resource-constrained conditions. To this end, we propose a generative AI framework that integrates large language models (LLMs) with distributed active inference (AIF) for intent-driven task graph generation and online resource scheduling. Specifically, a scenario-based iterative stream generation mechanism converts natural-language instructions into executable task graphs while mitigating context window exhaustion and structural hallucinations. Each uncrewed aerial vehicle (UAV) and edge server operates as an autonomous AIF agent that maintains a variational belief over hidden states coupled with the task graph, and minimizes expected free energy (EFE) from local noisy observations to optimize task offloading and mobility without centralized coordination or global retraining. Extensive experiments show that the proposed method consistently outperforms mainstream DRL benchmarks in convergence, robustness, and adaptability, demonstrating its effectiveness for dynamic low-altitude edge intelligence.

Generalizable Multi-Agent Planning From Signal Temporal Logic Specifications Via Diffusion

Multi-agent systems in the real-world (e.g., drone swarms, autonomous cars, warehouse robots) must satisfy rich, temporal tasks while avoiding collisions. Signal Temporal Logic (STL) elegantly encodes such objectives, but current STL planning methods face critical limitations. State-of-the-art optimization-based approaches can handle arbitrary STL specifications but struggle with scalability, becoming computationally impractical as the number of agents grows. Learning-based methods efficiently handle a large number of agents with rapid planning times but fare poorly when deployment-time objectives differ from those used during training, and do not support planning tasks that require different specifications to be ascribed to different agents (i.e., heterogeneity) or team-level specifications requiring coordination of multiple agents. This fundamental trade-off between generalizability and scalability presents a challenge for realizing multi-agent STL planning algorithms in practice. To overcome this challenge, we introduce a new diffusion method for multi-agent planning with STL specifications. Using a differentiable approximation of STL, we integrate the STL gradient in the denoising process, making our approach generalizable to novel formulas whose predicates are placed anywhere within the goal region covered during training, while achieving the same scalability as existing learning-based methods. Our method supports heterogeneous specifications, and by using diffusion models, naturally enhances plan diversity, thereby significantly reducing safety-related violations (e.g., collisions) among agents. A detailed evaluation study justifies the utility of STL-guided diffusion-based multi-agent planners for constructing generalizable, scalable, and diverse plans. Videos and code are available at https://www.jeappen.com/diff-ma-stl/ and https://github.com/jeappen/diff-ma-stl .

WARA: A Closed-Loop Multi-Agent Framework for Wireless Optimization Autoresearch

Comparative results show that WARA substantially outperforms one-shot LLM generation and approaches the quality profile of recently accepted peer-reviewed papers, demonstrating the potential of closed-loop artifact control for end-to-end LLM-assisted wireless optimization research.

Dynamic Traffic Steering for Networked Robotics Using 3GPP-Compliant Application Functions