Switch: Learning Agile Skills Switching for Humanoid Robots

์ €์ž: Yuen-Fui Lau, Qihan Zhao, Yinhuai Wang, Runyi Yu, Hok Wai Tsui, Qifeng Chen, Ping Tan | ๋‚ ์งœ: 2026-04-16 | URL: https://arxiv.org/abs/2604.14834 📄 PDF


Essence

Figure 2

Fig. 2: The Switch system: (a) We retarget human motion capture skills onto the robot. We then construct a skill graph w

Switch๋Š” Skill Graph๋ฅผ ๊ธฐ๋ฐ˜์œผ๋กœ humanoid robot์ด ์ž„์˜์˜ ์‹œ์ ์—์„œ ๋‹ค์–‘ํ•œ ๋™์ž‘ ๊ธฐ์ˆ ๋“ค ์‚ฌ์ด๋ฅผ ์ž์œ ๋กญ๊ฒŒ ์ „ํ™˜ํ•  ์ˆ˜ ์žˆ๋Š” ๊ณ„์ธต์  ์ „์‹  ์ œ์–ด ์‹œ์Šคํ…œ์„ ์ œ์‹œํ•œ๋‹ค.

Motivation

Achievement

Figure 4

Fig. 4: Quantitative Evaluation of Skill Execution Performance. We analyzes the full-body tracking performance of the pr

How

Figure 2

Fig. 2: The Switch system: (a) We retarget human motion capture skills onto the robot. We then construct a skill graph w

Originality

Limitation & Further Study

Evaluation

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

์ดํ‰: Switch๋Š” Skill Graph๋ผ๋Š” ๋‹จ์ˆœํ•˜๋ฉด์„œ๋„ ํšจ๊ณผ์ ์ธ ๊ตฌ์กฐ์™€ online graph search ๊ธฐ๋ฐ˜์˜ ๋™์  ์žฌ๊ณ„ํš์„ ํ†ตํ•ด humanoid robot์˜ skill switching ๋ฌธ์ œ๋ฅผ ์‹ค์šฉ์ ์œผ๋กœ ํ•ด๊ฒฐํ•œ ์˜๋ฏธ ์žˆ๋Š” ์—ฐ๊ตฌ์ด๋ฉฐ, ์‹ค์ œ ๋กœ๋ด‡ ํ”Œ๋žซํผ์—์„œ์˜ ๊ฒ€์ฆ์œผ๋กœ ๋†’์€ ์ ์šฉ ๊ฐ€์น˜๋ฅผ ๋ณด์—ฌ์ค€๋‹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˜์ •