Eixão-UAM: LLM-assisted iterative design of a low-altitude urban air mobility corridor in Brasilia

Abstract

The development of urban air mobility (UAM) systems requires scalable, regulation-aware planning of low-altitude airspace and supporting infrastructure. This study proposes an end-to-end framework for the design, simulation, and iterative optimization of a structured UAM corridor over Brasilia's central road axis (Eixão-UAM), aligned with the Brazilian unmanned aircraft traffic management (BR-UTM) ecosystem. In addition, this study proposes a multilayered aerial configuration stratified by unmanned aerial vehicle class, supported by a modular ground infrastructure composed of vertihubs, vertiports, and vertistops. A takeoff-scheduling simulator is developed to evaluate platform allocation strategies under realistic traffic and weather conditions. Initial experiments compare a round-robin (RR) baseline with a genetic algorithm (GA), and results reveal that RR outperforms GA v1 in terms of the average waiting time. To address this gap, a large language model (LLM) assisted optimization loop is implemented using GPT-4o Mini and Gemini 2.5 Pro. The LLMs act as reasoning partners, supporting the root-cause diagnoses, fitness function redesign, and rapid prototyping of five GA variants. Among these, GA v5 achieves a 59.62% reduction in maximum waiting time and an approximately 10% reduction in average waiting time over GA v1, thereby approaching the robustness of RR. In contrast, GA v2–v4 and GA v6 perform less consistently, showing an importance of fitness function design. These results underscore the role of an iterative, LLM-guided development in enhancing classical optimization, demonstrating that generative artificial intelligence (AI) can contribute to simulation acceleration and the cocreation of operational logic. The proposed method provides a replicable blueprint for integrating LLMs into early-stage UAM planning, offering both theoretical insights and architectural guidance for future low-altitude airspace systems.

1. Introduction

Urban air mobility (UAM) is rapidly emerging as a transformative component of smart urban infrastructure, enabling the use of unmanned aerial vehicles (UAVs) for logistics, emergency response, inspections, and passenger transport (Cohen et al., 2021; Jiang et al., 2023; Liao et al., 2024; Liu S and Liu, 2025; Xu et al., 2025). As cities grow denser and ground infrastructure becomes increasingly saturated, low-altitude air corridors offer a promising solution to mitigate surface congestion and improve service responsiveness (Verma et al., 2022; Moon et al., 2024). However, implementing scalable and safe UAM systems poses significant challenges in terms of airspace structuring, flight intent management, and ground infrastructure deployment.

Brazil is uniquely positioned to lead UAM integration through its national unmanned aircraft traffic management (UTM) framework, known as BR-UTM (da Silva et al., 2020; Jasper and Nunes, 2022; Turchetti and Murça, 2024). This framework features key platforms such as unmanned aircraft system for UAV registration and authorization (SISANT), which is managed by the National Civil Aviation Agency of Brazil (ANAC), and the request for access for remotely piloted aircraft-next generation (SARPAS NG), which is operated by the Brazilian Department of Airspace Control (DECEA). Both the ANAC and DECEA are interoperable with the national Gov.br digital infrastructure. Leveraging this regulatory and technological foundation, Brasilia, the capital of Brazil and a city designed with geometric precision, offers an ideal testing ground for structured low-altitude airspace models.

To explore these opportunities, this study concentrates on the modeling and validation of a UAM corridor along Brasilia's iconic south–north road axis (Eixão Rodoviário, also known as Eixão, Portuguese for “big axis”). The primary objective is to design and assess the BSB Eixão-UAM corridor within Brazil's BR-UTM framework, ensuring alignment with national airspace management initiatives. While the corridor design remains the central focus, the analysis of takeoff scheduling at a vertiport is presented as the first operational demonstration. This addition grounds the corridor concept in practical feasibility.