In this paper, we present a learning platform for robotic grasping in real world, in which actor-critic deep reinforcement learning is employed to directly learn the grasping skill from raw image pixels and rarely observed rewards. This is a challenging task because existing algorithms based on deep reinforcement learning require an extensive number of training data or massive computational cost so that they cannot be affordable in real world settings. To address this problems, the proposed learning platform basically consists of two training phases; a learning phase in simulator and subsequent learning in real world. Here, main processing blocks in the platform are extraction of latent vector based on state representation learning and disentanglement of a raw image, generation of adapted synthetic image using generative adversarial networks, and object detection and arm segmentation for the disentanglement. We demonstrate the effectiveness of this approach in a real environment.

Abstract
1. 서 론
2. 관련연구
3. 제안하는 학습 플랫폼
4. 시뮬레이터에서의 학습
    4.1 상태표현 학습
    4.2 생성적 적대 신경망 학습
5. 현실 세계에서의 학습
    5.1 영상에서의 물체 검출
    5.2 영상에서의 로봇 암 분할
6. 실 험
    6.1 로봇 파지실험을 위한 설정
    6.2 로봇 파지 실험 결과
7. 결 론
References

• 김태원(Department of Electronics and Computer Engineering, Hanyang University) | Taewon Kim
• 박예성(Department of Electronics and Computer Engineering, Hanyang University) | Yeseong Park
• 김종복(Department of Electronics and Computer Engineering, Hanyang University) | Jong Bok Kim
• 박영빈(Department of Electronics and Computer Engineering, Hanyang University) | Youngbin Park
• 서일홍(Department of Electronics and Computer Engineering, Hanyang University) | Il Hong Suh Corresponding author