ロボット政策模倣のためのベイズガウス混合モデル
Bayesian Gaussian mixture model for robotic policy imitation
各状態での行動の不確実性をベイズ手法で定量化し、模倣政策と追加政策を融合することで未知状態でも適切に動作するロボットスキル学習を実現した。
著者: Emmanuel Pignat, Sylvain Calinon
分類: cs.RO
原文アブストラクト
A common approach to learn robotic skills is to imitate a demonstrated policy. Due to the compounding of small errors and perturbations, this approach may let the robot leave the states in which the demonstrations were provided. This requires the consideration of additional strategies to guarantee that the robot will behave appropriately when facing unknown states. We propose to use a Bayesian method to quantify the action uncertainty at each state. The proposed Bayesian method is simple to set up, computationally efficient, and can adapt to a wide range of problems. Our approach exploits the estimated uncertainty to fuse the imitation policy with additional policies. It is validated on a Panda robot with the imitation of three manipulation tasks in the continuous domain using different control input/state pairs.
