Learning Aerodynamics for the Control of Flying Humanoid Robots

์ €์ž: Antonello Paolino, Gabriele Nava, Fabio Di Natale, Fabio Bergonti, Punith Reddy Vanteddu, Donato Grassi, Luca Riccobene, Alex Zanotti, Renato Tognaccini, Gianluca Iaccarino, Daniele Pucci | ๋‚ ์งœ: 2025-05-30 | URL: https://arxiv.org/abs/2506.00305 📄 PDF


Essence

Figure 1

Fig. 1: Design of the iRonCub-Mk1 physical prototype. Front (a) and rear (b) pictures of the

๋น„ํ–‰ ์ธ๊ฐ„ํ˜• ๋กœ๋ด‡์˜ ๊ณต๊ธฐ์—ญํ•™ ๋ชจ๋ธ๋ง์„ ์œ„ํ•ด CFD ์‹œ๋ฎฌ๋ ˆ์ด์…˜, ํ’๋™ ์‹คํ—˜, ๋”ฅ๋Ÿฌ๋‹์„ ๊ฒฐํ•ฉํ•œ ํฌ๊ด„์  ์ ‘๊ทผ ๋ฐฉ์‹์„ ์ œ์‹œํ•˜๊ณ , ์ œํŠธ ์—”์ง„์„ ์žฅ์ฐฉํ•œ iRonCub-Mk1 ๋กœ๋ด‡์„ ์„ค๊ณ„ยท์ œ์ž‘ํ•˜์—ฌ ๋น„ํ–‰ ์ œ์–ด๋ฅผ ๊ตฌํ˜„ํ•œ๋‹ค.

Motivation

Achievement

Figure 4

Fig. 4: Aerodynamic models for simulation and control. Deep Neural Network (DNN): (a)

How

Figure 3

Fig. 3: iRonCub aerodynamics CFD simulations. Validation of RANS simulations on iRon-

Originality

Limitation & Further Study

Evaluation

Novelty: 4/5 Technical Soundness: 3/5 Significance: 4/5 Clarity: 4/5 Overall: 4/5

์ดํ‰: ์ธ๊ฐ„ํ˜• ๋กœ๋ด‡์˜ ๋น„ํ–‰ ๋Šฅ๋ ฅ ํ™•๋ณด๋ฅผ ์œ„ํ•ด ๊ณต๊ธฐ์—ญํ•™ ๋ชจ๋ธ๋ง๊ณผ ์ œ์–ด๋ฅผ ์ข…ํ•ฉ์ ์œผ๋กœ ๋‹ค๋ฃฌ ๊ธฐ์ˆ ์ ยท๊ณผํ•™์ ์œผ๋กœ ์˜๋ฏธ ์žˆ๋Š” ์—ฐ๊ตฌ์ด๋ฉฐ, ๋‹ค์ค‘ ๋ชจ๋“œ ๋กœ๋ด‡์˜ ๋ฏธ๋ž˜ ์„ค๊ณ„์— ์ค‘์š”ํ•œ ๊ธฐ์—ฌ๋ฅผ ์ œ์‹œํ•œ๋‹ค. ๋‹ค๋งŒ ์‹ค์ œ ๋น„ํ–‰ ์‹คํ—˜ ๊ฒ€์ฆ๊ณผ ํ•™์Šต ๋ชจ๋ธ์˜ ์ผ๋ฐ˜ํ™” ์„ฑ๋Šฅ ํ‰๊ฐ€๊ฐ€ ํ›„์† ๊ณผ์ œ์ด๋‹ค.

← ๋ชฉ๋ก์œผ๋กœ ๋Œ์•„๊ฐ€๊ธฐ

๐ŸŽง Audio Overview

์ด ๋…ผ๋ฌธ ๋ฆฌ๋ทฐ๋ฅผ ํŒŸ์บ์ŠคํŠธํ˜• ์˜ค๋””์˜ค๋กœ ์ƒ์„ฑํ•ฉ๋‹ˆ๋‹ค. (Gemini ยท ํ‚ค๋Š” ๋ธŒ๋ผ์šฐ์ €์—๋งŒ ์ €์žฅ ยท ์™„์„ฑ๋ณธ์€ ์ด๋ฉ”์ผ๋กœ๋„ ์ „์†ก)
โ–ธ ๊ณ ๊ธ‰: ๊ตฌ์„ฑ ๋ฐฉํ–ฅ(๋Œ€๋ณธ ์ž‘์„ฑ ์ง€์นจ) ์ง์ ‘ ์ˆ˜์ •