Learning Transferable Policies for Monocular Reactive MAV Control
This addresses the challenge of adapting autonomous behavior in robotics, though it is incremental as a small step in this direction.
The paper tackles the problem of transferring learned motion policies from a source to a target domain for autonomous MAV flight, demonstrating efficacy through real-world outdoor experiments.
The ability to transfer knowledge gained in previous tasks into new contexts is one of the most important mechanisms of human learning. Despite this, adapting autonomous behavior to be reused in partially similar settings is still an open problem in current robotics research. In this paper, we take a small step in this direction and propose a generic framework for learning transferable motion policies. Our goal is to solve a learning problem in a target domain by utilizing the training data in a different but related source domain. We present this in the context of an autonomous MAV flight using monocular reactive control, and demonstrate the efficacy of our proposed approach through extensive real-world flight experiments in outdoor cluttered environments.