Obstacle Avoidance for UAS in Continuous Action Space Using Deep Reinforcement Learning

Hu, Jueming; Yang, Xuxi; Wang, Weichang; Wei, Ping; Lü, Ying; Liu, Yongming

doi:10.48550/arxiv.2111.07037

Cited by 1 publication

(1 citation statement)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Zhou & Kwan developed a comprehensive contingency planning framework for loss of communication cases by considering all the aspects of loss of link contingency [9]. With the use of deep reinforcement learning, improvement of collision avoidance system using deep Q-network and its adaptation into the unmanned traffic is explored by Li et al [10] and obstacle avoidance during operations using proximal policy optimization is provided by Hu et al [11]. Grüter et al worked on emergency flight planning of UAVs to a safe landing zone during an emergency situation by using Voronoi diagrams and selecting the most suitable path with dynamic programming [12].…”

Section: Introductionmentioning

confidence: 99%

Contingency Management Concept Generation for U-Space System

Altun

Inalhan

2022

2022 Integrated Communication, Navigation and Surveillance Conference (ICNS)

View full text Add to dashboard Cite

Contingency management in aviation is a vital concept that ensures safety, security, and efficiency in operations. To fully benefit from the envisioned Advanced Air Mobility system, the need of a structured and system-wide contingency planning will be even more important since the air transportation system paradigm will shift into a highly automated system that includes high-density traffic. The complexity will increase considerably by enlarging the operations to the underserved urban areas. Therefore, the new system needs to provide a more agile, accessible, and flexible environment. In this paper, the need of a contingency management from a holistic approach is described and the base requirements to build such a system are defined by considering the roles and responsibilities of each stakeholder that are defined for the U-space system. Alongside the defined requirements, the tasks of the stakeholders and the expected main contingency sources are explained to have a better understanding of the system. The objective of this work is to provide the base guidelines that help to set appropriate actions by relevant stakeholder under an adverse condition which might compromise the safety and security of the operations within the air traffic network.

show abstract