Interactively shaping robot behaviour with unlabeled human instructions - Sorbonne Université
Journal Articles Autonomous Agents and Multi-Agent Systems Year : 2020

Interactively shaping robot behaviour with unlabeled human instructions

Abstract

In this paper, we propose a framework that enables a human teacher to shape a robot behaviour by interactively providing it with unlabeled instructions. We ground the meaning of instruction signals in the task-learning process, and use them simultaneously for guiding the latter. We implement our framework as a modular architecture, named TICS (Task-Instruction-Contingency-Shaping) that combines different information sources: a predefined reward function, human evaluative feedback and unlabeled instructions. This approach provides a novel perspective for robotic task learning that lies between Reinforcement Learning and Supervised Learning paradigms. We evaluate our framework both in simulation and with a real robot. The experimental results demonstrate the effectiveness of our framework in accelerating the task-learning process and in reducing the number of required teaching signals.
Fichier principal
Vignette du fichier
Najar et al. - 2020 - Interactively shaping robot behaviour with unlabel.pdf (1.39 Mo) Télécharger le fichier
Origin Files produced by the author(s)
Loading...

Dates and versions

hal-02996137 , version 1 (09-11-2020)

Identifiers

Cite

Anis Najar, Olivier Sigaud, Mohamed Chetouani. Interactively shaping robot behaviour with unlabeled human instructions. Autonomous Agents and Multi-Agent Systems, 2020, 34 (2), ⟨10.1007/s10458-020-09459-6⟩. ⟨hal-02996137⟩
64 View
155 Download

Altmetric

Share

More