TACOS: Task Agnostic COordinator of a multi-drone System
Alessandro Nazzari, Roberto Rubinacci, Marco Lovera
TL;DR
The paper addresses scalable, flexible control of a multi-UAV swarm by leveraging a hierarchical, language-model–driven architecture (TACOS) that combines a Coordinator for high-level planning with a Supervisor for execution, all interfacing through a library of executable APIs under the ReAct paradigm. It demonstrates that separating reasoning from action improves planning quality and execution reliability, validated via extensive simulations with drone failures and real-world indoor flights. An ablation study shows the benefits of explicit reasoning and task-plan decomposition, while case studies on target search and moving targets illustrate dynamic replanning and robust reallocation. The work advances intuitive, language-based swarm control with practical integration to low-level planners like ATOMICA and ROS Noetic, highlighting significant potential for flexible, resilient multi-robot missions.
Abstract
When a single pilot is responsible for managing a multi-drone system, the task may demand varying levels of autonomy, from direct control of individual UAVs, to group-level coordination, to fully autonomous swarm behaviors for accomplishing high-level tasks. Enabling such flexible interaction requires a framework that supports multiple modes of shared autonomy. As language models continue to improve in reasoning and planning, they provide a natural foundation for such systems, reducing pilot workload by enabling high-level task delegation through intuitive, language-based interfaces. In this paper we present TACOS (Task-Agnostic COordinator of a multi-drone System), a unified framework that enables high-level natural language control of multi-UAV systems through Large Language Models (LLMs). TACOS integrates three key capabilities into a single architecture: a one-to-many natural language interface for intuitive user interaction, an intelligent coordinator for translating user intent into structured task plans, and an autonomous agent that executes plans interacting with the real world. TACOS allows a LLM to interact with a library of executable APIs, bridging semantic reasoning with real-time multi-robot coordination. We demonstrate the system on a real-world multi-drone system, and conduct an ablation study to assess the contribution of each module.