A Survey of Embodied Learning for Object-Centric Robotic Manipulation

์ €์ž: Ying Zheng, Lei Yao, Yuejiao Su, Yi Zhang, Yi Wang, Sicheng Zhao, Yiyi Zhang, Lap-Pui Chau | ๋‚ ์งœ: 2024-08-21 | URL: https://arxiv.org/abs/2408.11537 📄 PDF


Essence

Figure 1

Fig. 1. An illustration of robotic manipulation system (left) and the typology of embodied learning methods for object-c

๋ณธ ๋…ผ๋ฌธ์€ object-centric robotic manipulation์„ ์œ„ํ•œ embodied learning์˜ ์ตœ์‹  ๋™ํ–ฅ์„ ์ฒด๊ณ„์ ์œผ๋กœ ์กฐ์‚ฌํ•˜๋ฉฐ, embodied perceptual learning, embodied policy learning, embodied task-oriented learning์˜ ์„ธ ๊ฐ€์ง€ ์ฃผ์š” ๋ถ„์•ผ๋กœ ๋ถ„๋ฅ˜ํ•˜์—ฌ ์ข…ํ•ฉ์ ์ธ ์„œ๋ฒ ์ด๋ฅผ ์ œ๊ณตํ•œ๋‹ค.

Motivation

Achievement

Figure 1

Fig. 1. An illustration of robotic manipulation system (left) and the typology of embodied learning methods for object-c

How

Figure 1

Fig. 1. An illustration of robotic manipulation system (left) and the typology of embodied learning methods for object-c

Originality

Limitation & Further Study

Evaluation

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

์ดํ‰: ๋ณธ ๋…ผ๋ฌธ์€ object-centric robotic manipulation์„ ์œ„ํ•œ embodied learning์˜ ์ตœ์‹  ๋™ํ–ฅ์„ ์ฒด๊ณ„์ ์ด๊ณ  ํฌ๊ด„์ ์œผ๋กœ ์ •๋ฆฌํ•œ ์šฐ์ˆ˜ํ•œ ์„œ๋ฒ ์ด์ด๋ฉฐ, ๊ธฐ์กด ์—ฐ๊ตฌ์™€ ๋‹ฌ๋ฆฌ ์ตœ์‹  generative/foundation models์„ ํฌํ•จํ•˜๊ณ  perception-policy-task์˜ ํ†ตํ•ฉ์  ๊ด€์ ์„ ์ œ์‹œํ•จ์œผ๋กœ์จ ๋กœ๋ด‡ ์กฐ์ž‘ ๋ถ„์•ผ ์—ฐ๊ตฌ์ž๋“ค์—๊ฒŒ ๋งค์šฐ ์œ ์šฉํ•œ ์ฐธ๊ณ ์ž๋ฃŒ๊ฐ€ ๋  ๊ฒƒ์œผ๋กœ ํŒ๋‹จ๋œ๋‹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˜์ •