Comparing the Performance of Traffic Coordination Methods for Advanced Aerial Mobility
Traffic Management in Advanced Aerial Mobility (AAM) inherits many elements of conventional Air Traffic Management (ATM), but brings new complexities and challenges of its own. One of its ways of guaranteeing separation is the use of airborne, stand-alone Detect-And-Avoid, an operational concept where each aircraft decides its avoidance maneuvers independently, observing right-of-way rules and, in specific implementations, some form of pairwise coordination. This is a fundamental safety element for autonomous aircraft but, according to our research, is not sufficient for high-density airspaces as envisioned for urban environments. In these environments, some way of explicit and strategic traffic coordination must be in place, as done for conventional ATM. For efficiency reasons, ATM is evolving to more flexible uses of the airspace, such that the use of dynamically allocated corridors is a rising concept for AAM. These strategic forms of traffic coordination are potentially highly efficient if the aircraft adhere to their trajectory contracts and there are no significant perturbations to the traffic. However, if significant perturbations occur, such as loss of data communication, or the sudden appearance of an intruder, a centralized system may not react appropriately in due time. In busy scenarios, even small deviations from plans may compound so rapidly as to result in large differences in the overall achieved scenario, resulting in congestions and convoluted conflicts. Therefore, it is worth studying traffic coordination techniques that work locally with shorter look-ahead times. To that end, we explore an airborne collaborative method for traffic coordination, which is capable of safely solving conflicts with multiple aircraft, stressing its capabilities throughout a large number of scenarios and comparing its performance with established methods.
Code (0)
등록된 구현이 없습니다.
Tasks
ManagementSimilar Papers 제목 키워드 기반
Decentralized Air Traffic Management for Advanced Air Mobility
Leading proposals for Advanced Air Mobility (AAM) recognize the need for strategic and tactical airspace deconfliction, where the need for coordination appears in different forms and nuances. One recurring element is the…
DiversityManagementCoordLight: Learning Decentralized Coordination for Network-Wide Traffic Signal Control
Adaptive traffic signal control (ATSC) is crucial in alleviating congestion, maximizing throughput and promoting sustainable mobility in ever-expanding cities. Multi-Agent Reinforcement Learning (MARL) has recently shown…
Multi-agent Reinforcement LearningAdaptive Coordination Offsets for Signalized Arterial Intersections using Deep Reinforcement Learning
Coordinating intersections in arterial networks is critical to the performance of urban transportation systems. Deep reinforcement learning (RL) has gained traction in traffic control research along with data-driven appr…
Deep Reinforcement Learningreinforcement-learningReinforcement Learning (RL)The Performance Impact of Combining Agent Factorization with Different Learning Algorithms for Multiagent Coordination
Factorizing a multiagent system refers to partitioning the state- action space to individual agents and defining the interactions be- tween those agents. This so-called agent factorization is of much im- portance in r…
ManagementPolicy Gradient MethodsGraphTrafficGPT: Enhancing Traffic Management Through Graph-Based AI Agent Coordination
Large Language Models (LLMs) offer significant promise for intelligent traffic management; however, current chain-based systems like TrafficGPT are hindered by sequential task execution, high token usage, and poor scalab…