Robust and Versatile Bipedal Jumping Control through Reinforcement Learning

์ €์ž: Zhongyu Li, Xue Bin Peng, Pieter Abbeel, Sergey Levine, Glen Berseth, Koushil Sreenath | ๋‚ ์งœ: 2023-02-19 | URL: https://arxiv.org/abs/2302.09450 📄 PDF


Essence

Figure 1

Fig. 1: Representative dynamic jumping maneuvers performed by a bipedal robot Cassie using the proposed goal-conditioned

Reinforcement learning๊ณผ ์ƒˆ๋กœ์šด ์ •์ฑ… ๊ตฌ์กฐ๋ฅผ ํ™œ์šฉํ•˜์—ฌ ์ด์กฑ ๋กœ๋ด‡ Cassie๊ฐ€ ๋‹ค์–‘ํ•œ ์ฐฉ์ง€ ์œ„์น˜์™€ ๋ฐฉํ–ฅ์œผ๋กœ ์ ํ”„ํ•˜๋Š” ๊ฐ•๊ฑดํ•˜๊ณ  ๋‹ค๋ชฉ์ ์ธ ๋™์  ์ ํ”„ ์ œ์–ด๋ฅผ ์‹คํ˜„ํ–ˆ๋‹ค.

Motivation

Achievement

Figure 1

Fig. 1: Representative dynamic jumping maneuvers performed by a bipedal robot Cassie using the proposed goal-conditioned

How

Figure 2

Fig. 2: The schematic to train the robot to perform versatile jumping

Originality

Limitation & Further Study

Evaluation

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

์ดํ‰: ์ด์กฑ ๋กœ๋ด‡์˜ ๋™์  ์ ํ”„ ์ œ์–ด์—์„œ RL๊ณผ ์ƒˆ๋กœ์šด ์ •์ฑ… ๊ตฌ์กฐ๋ฅผ ๊ฒฐํ•ฉํ•˜์—ฌ ๊ธฐ์กด ๋ฐฉ๋ฒ•์„ ํฌ๊ฒŒ ๋›ฐ์–ด๋„˜๋Š” ์‹ค์ œ ์„ธ๊ณ„ ์„ฑ๊ณผ๋ฅผ ๋‹ฌ์„ฑํ•œ ์šฐ์ˆ˜ํ•œ ์—ฐ๊ตฌ์ด๋ฉฐ, ๋‹ค๋ชฉ์  ๊ฐ•๊ฑดํ•œ ๋กœ๋ด‡ ์ œ์–ด์˜ ์ƒˆ๋กœ์šด ๊ฐ€๋Šฅ์„ฑ์„ ๋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˜์ •