Supervised learning and reinforcement learning of feedback models for reactive behaviors: Tactile feedback testbed. (November 2022)

• Record Type: Journal Article
• Authors: Sutanto, Giovanni; Rombach, Katharina; Chebotar, Yevgen; Su, Zhe; Schaal, Stefan; Sukhatme, Gaurav S.; Meier, Franziska
• Is Part Of: International journal of robotics research. Volume 41:Number 13/14(2022)
• Page Start: 1121
• Page End: 1145
• Publication Date: 2022-11
• Subjects: supervised learning -- feedback models -- reactive behaviors -- tactile feedback -- dynamic movement primitives -- phase-modulated neural networks -- reinforcement learning -- weighted least square -- demonstration -- alignment -- segmentation; Robots -- Periodicals; Robots, Industrial -- Periodicals; 629.89205
• DOI: 10.1177/02783649221143399
• Languages: English
• ISSNs: 0278-3649

Abstract

Robots need to be able to adapt to unexpected changes in the environment such that they can autonomously succeed in their tasks. However, hand-designing feedback models for adaptation is tedious, if at all possible, making data-driven methods a promising alternative. In this paper, we introduce a full framework for learning feedback models for reactive motion planning. Our pipeline starts by segmenting demonstrations of a complete task into motion primitives via a semi-automated segmentation algorithm. Then, given additional demonstrations of successful adaptation behaviors, we learn initial feedback models through learning-from-demonstrations. In the final phase, a sample-efficient reinforcement learning algorithm fine-tunes these feedback models for novel task settings through few real system interactions. We evaluate our approach on a real anthropomorphic robot in learning a tactile feedback task.