DreamWaQ: legged robot walks in harsh environments with its imagination скачать в хорошем качестве

DreamWaQ: legged robot walks in harsh environments with its imagination

KAIST has developed a quadrupedal robot locomotion technology that moves up and down stairs without the aid of visual or tactile sensors in a disaster situation where it is impossible to see due to smoke and moves without falling over bumpy environments such as tree roots. Professor Hyun Myung's research team (Urban Robotics Lab) in the School of Electrical Engineering developed a walking robot locomotion technology that enables robust 'blind locomotion' in various atypical environments. See also:    • DreamWaQ: legged robot walks in harsh envi...   For details: https://urobot.kaist.ac.kr/media-rele... Reference: I Made Aswin Nahrendra, Byeongho Yu, and Hyun Myung, "DreamWaQ: Learning Robust Quadrupedal Locomotion With Implicit Terrain Imagination via Deep Reinforcement Learning," Accepted to ICRA (IEEE Int'l Conf. Robotics and Automation) 2023, May 2023. https://arxiv.org/abs/2301.10602 Abstract: Quadrupedal robots resemble the physical ability of legged animals to walk through unstructured terrains. However, designing a controller for quadrupedal robots poses a significant challenge due to their functional complexity and requires adaptation to various terrains. Recently, deep reinforcement learning, inspired by how legged animals learn to walk from their experiences, has been utilized to synthesize natural quadrupedal locomotion. However, state-of-the-art methods strongly depend on a complex and reliable sensing framework. Furthermore, prior works that rely only on proprioception have shown a limited demonstration for overcoming challenging terrains, especially for a long distance. This work proposes a novel quadrupedal locomotion learning framework that allows quadrupedal robots to walk through challenging terrains, even with limited sensing modalities. The proposed framework was validated in real-world outdoor environments with varying conditions within a single run for a long distance.