This paper presents LEVIOSA, a novel framework for text- and speech-based uncrewed aerial vehicle (UAV) trajectory generation. By leveraging multimodal large language models (LLMs) to interpret natural language commands, the system converts text and audio inputs into executable flight paths for UAV swarms. The approach aims to simplify the complex task of multi-UAV trajectory generation, which has significant applications in fields such as search and rescue, agriculture, infrastructure inspection, and entertainment. The framework involves two key innovations: a multi-critic consensus mechanism to evaluate trajectory quality and a hierarchical prompt structuring for improved task execution. The innovations ensure fidelity to user goals. The framework integrates several multimodal LLMs for high-level planning, converting natural language inputs into 3D waypoints that guide UAV movements and per-UAV low-level controllers to control each UAV in executing its assigned 3D waypoint path based on the high-level plan. The methodology was tested on various trajectory types with promising accuracy, synchronization, and collision avoidance results. The findings pave the way for more intuitive human–robot interactions and advanced multi-UAV coordination.

Active inference, grounded in the free energy principle, provides a powerful lens for understanding how agents balance exploration and goal-directed behavior in uncertain environments. Here, we propose a new planning framework that integrates Monte Carlo tree search (MCTS) with active inference objectives to systematically reduce epistemic uncertainty while pursuing extrinsic rewards. Our key insight is that MCTS, already renowned for its search efficiency, can be naturally extended to incorporate free energy minimization by blending expected rewards with information gain. Concretely, the cross-entropy method (CEM) is used to optimize action proposals at the root node, while tree expansions leverage reward modeling alongside intrinsic exploration bonuses. This synergy allows our planner to maintain coherent estimates of value and uncertainty throughout planning, without sacrificing computational tractability. Empirically, we benchmark our planner on a diverse set of continuous control tasks, where it demonstrates performance gains over both stand-alone CEM and MCTS with random rollouts.

Ad-hoc networks facilitate localized communication in environments lacking global internet connectivity. However, traditional Wi-Fi Direct (WFD) methods face challenges in forming intergroup networks during field deployments, due to manual password entry, single-path traffic between Group Owners (GOs), and performance bottlenecks from UDP multicast. To address these issues, we propose a Simple TCP-enabled, REsilient Android Mesh (STREAM); a multigroup bridging strategy where WFD Group Members (GMs) join two groups and bridge the connection between them exclusively using TCP sockets --an accomplishment yet to be realized in WFD ad-hoc networks on stock Android devices. This approach maximizes throughput and resilience for intergroup connections. STREAM also features self-healing capabilities to detect and rectify network disruptions automatically, removing the need for user intervention. Our architecture leverages Wi-Fi bandwidth, achieving average file transfer speeds of 91.2 Mbps in intergroup networks, demonstrating a significant performance improvement. Our results show that STREAM offers a solution for developing scalable and reliable multigroup Wi-Fi Direct networks.