09 - Crewed and Uncrewed Aircraft Operations and Traffic Management

LOW-NOISE VERTIPORT PROCEDURE PLANNING USING DEEP REINFORCEMENT LEARNING

Y. Wang¹, J. Guo¹, Y. He¹, F. Wu¹, Z. Gao¹; ¹The Hong Kong University of Science and Technology, Hong Kong SAR of China

This study presents a Deep RL framework that transforms vertiport airspace design. By optimizing for acoustic fairness, we generate dynamic flight corridors that reduce community noise by max 5.8 dB. This offers a vital policy-in-the-loop tool for regulators to standardize low-noise procedures and quantify vertiport acoustic capacity.

View Paper

DOI: 10.71945/icas2026_0580