ETRI-Knowledge Sharing Plaform

ENGLISH

성과물

논문 검색
구분 SCI
연도 ~ 키워드

상세정보

학술지 Uncertainty-Aware Knowledge Distillation for Collision Identification of Collaborative Robots
Cited 0 time in scopus Download 12 time Share share facebook twitter linkedin kakaostory
저자
권우경, 진용식, 이상준
발행일
202110
출처
Sensors, v.21 no.19, pp.1-16
ISSN
1424-8220
출판사
MDPI
DOI
https://dx.doi.org/10.3390/s21196674
협약과제
21ZD1100, 대경권 지역산업 기반 ICT 융합기술 고도화 지원사업, 문기영
초록
Human-robot interaction has received a lot of attention as collaborative robots became widely utilized in many industrial fields. Among techniques for human-robot interaction, collision identification is an indispensable element in collaborative robots to prevent fatal accidents. This paper proposes a deep learning method for identifying external collisions in 6-DoF articulated robots. The proposed method expands the idea of CollisionNet, which was previously proposed for collision detection, to identify the locations of external forces. The key contribution of this paper is uncertainty-aware knowledge distillation for improving the accuracy of a deep neural network. Sample-level uncertainties are estimated from a teacher network, and larger penalties are imposed for uncertain samples during the training of a student network. Experiments demonstrate that the proposed method is effective for improving the performance of collision identification.
키워드
Collaborative robot, Collision identification, Deep learning, Knowledge distillation, Uncertainty estimation
KSP 제안 키워드
Articulated robots, Collaborative robot, Collision detection, Deep learning method, Deep neural network(DNN), External force, Fatal accidents, Human-Robot Interaction(HRI), Six degrees of freedom(6-DoF), Uncertainty estimation, deep learning(DL)
본 저작물은 크리에이티브 커먼즈 저작자 표시 (CC BY) 조건에 따라 이용할 수 있습니다.
저작자 표시 (CC BY)