An Open-World Simulated Environment for Developmental Robotics
This provides a foundational tool for developmental robotics research, addressing a gap in self-supervised learning environments.
The paper tackles the need for environments supporting self-supervised learning in AI by introducing SEDRo, a simulated environment that mimics human infant experiences from fetus to 12 months, with evaluation based on developmental psychology tests.
As the current trend of artificial intelligence is shifting towards self-supervised learning, conventional norms such as highly curated domain-specific data, application-specific learning models, extrinsic reward based learning policies etc. might not provide with the suitable ground for such developments. In this paper, we introduce SEDRo, a Simulated Environment for Developmental Robotics which allows a learning agent to have similar experiences that a human infant goes through from the fetus stage up to 12 months. A series of simulated tests based on developmental psychology will be used to evaluate the progress of a learning model.