複数座標系のコストバランスによるスキル獲得
Skill Acquisition via Automated Multi-Coordinate Cost Balancing
複数の微分座標系でデモンストレーションを表現し、凸最適化で点間移動スキルを再現する学習フレームワークMCCBを提案。各座標系の最適な重み付けも自動学習する。
著者: Harish Ravichandar, S. Reza Ahmadzadeh, M. Asif Rana, Sonia Chernova
分類: cs.RO
原文アブストラクト
We propose a learning framework, named Multi-Coordinate Cost Balancing (MCCB), to address the problem of acquiring point-to-point movement skills from demonstrations. MCCB encodes demonstrations simultaneously in multiple differential coordinates that specify local geometric properties. MCCB generates reproductions by solving a convex optimization problem with a multi-coordinate cost function and linear constraints on the reproductions, such as initial, target, and via points. Further, since the relative importance of each coordinate system in the cost function might be unknown for a given skill, MCCB learns optimal weighting factors that balance the cost function. We demonstrate the effectiveness of MCCB via detailed experiments conducted on one handwriting dataset and three complex skill datasets.
