A Comprehensive Survey on Conflict Detection and Resolution in Unmanned Aircraft System Traffic Management

The anticipated proliferation of Unmanned Aerial Vehicles (UAVs) in the airspace in the coming years has raised concerns about how to manage their flights to avoid collisions and crashes at various stages of flight. To this end, many Unmanned Aircraft Traffic Management systems (UTM) have been designed. These systems use various methods for managing UAV conflicts. Several surveys have reviewed conflict resolution methods for UAVs. However, to the best of our knowledge, there is no survey specifically addressing conflict detection and resolution methods in UTM, particularly those using AI-based methods. Therefore, this article serves as a comprehensive survey of all UAVs conflicts detection and resolution methods proposed in the literature and their use in the UTM systems. This survey classifies the methods into two categories: classical (non-learning) methods and learning-based methods. Classical methods typically rely on pre-defined algorithms or rules for UAVs to avoid collisions, whereas Artificial Intelligence-based methods, including Machine Learning (ML) and especially Reinforcement Learning (RL), enable UAVs to adapt to their environment, autonomously resolve conflicts, and exhibit intelligent behavior based on their experiences. It also presents their application in the conflict resolution service for UTMs. Additionally, the challenges and issues associated with each type of methods are discussed. This article can serve as a foundational resource for researchers in guiding their selection of methods for conflict resolution, particularly those relevant to UTM systems.

I. Introduction

The proliferation of the UAV utilization spans across various civilian sectors, including agriculture, energy, public safety and security, e-commerce and delivery, mobility, and transportation. UAVs and Unmanned Ground Vehicles (UGVs) can be integrated to enhance operational efficiency in various sectors, including disaster response and environmental monitoring [1]. As indicated in the European Drones Outlook Study [2], the European drone market is anticipated to witness substantial expansion, with projections indicating that demand will surpass 10 billion € annually by 2035 and exceed 15 billion € annually by 2050. Civil applications alone are forecast to contribute over 5 billion € annually by 2035. Furthermore, the defense and leisure sectors are expected to collectively generate nearly 2 billion € in annual revenue from product sales in Europe over the long term. Several nations and commercial enterprises have begun developing UTM systems to manage UAV activities in airspace while coordinating with the Air Traffic Management (ATM) system, responsible for the management of manned aircraft activities, to ensure safe and efficient operations within airspace. Integrating UAVs into the airspace requires collaboration between ATM and UTM, addressing regulatory and technological aspects like airspace design, infrastructure, standards, and new regulations to accommodate UAVs as airspace users.Projects like NASA UTM in the USA [3] and U-Space in Europe [4] represent two of the largest governmental initiatives in this domain. Additionally, companies like Airmap,1 Unifly,2 Airbus,3 and U-Flyte 4 have invested in UTM development projects. However, these UTMs face numerous challenges, particularly concerning safety. Despite the benefits of UAV utilization, the potential dangers arise from collisions with static and dynamic obstacles. As defined in [5], static obstacles are fixed entities located at known positions, such as buildings and trees. Whereas, dynamic obstacles are objects capable of appearing suddenly along the UAV’s trajectory, with their locations subject to change. The risk of collision and accidents with the UAVs escalates with increased air traffic. A report by [6] has documented a significant number of UAV-related incidents worldwide in recent years. In this report, it is shown that the consequences of UAV collision can range from minor injuries and limited financial losses, which may disrupt the operation of essential infrastructure, to more severe outcomes involving significant injuries and extensive damage to critical assets, with profound economic repercussions. Various Collision Avoidance Systems (CAS) have been proposed in the literature. Most of these systems use classical conflict resolution methods that require prior knowledge of all obstacles and pre-mission path planning, such as optimized trajectory or force field methods. Clearly, these methods have notable limitations particularly in dynamic environments characterized by unpredictable or constantly evolving spatial conditions and flight paths changes due to mission alterations or meteorological variations. To overcome these limitations, many researchers have started exploring new conflict resolution methods using Machine Learning (ML) and more particularly Reinforcement Learning (RL). This is because RL methods excel in dynamic environments due to their ability to learn from interactions with the environment adapting and optimize their strategies over time to handle complex and evolving scenarios.¹

https://www.airmap.com/manage/ansp/utm-center-ansp

²

https://www.unifly.aero/solutions/unmanned-traffic-management

³

https://www.airbus.com/en/innovation/autonomous-connected/airbus-unmanned-traffic-management

⁴

www.u-flyte.com