Investigation of Factorized Optical Flows as Mid-Level Representations

TL;DR

This paper explores the use of factorized optical flow maps as mid-level representations to improve the integration of perception and control in robotic learning, demonstrating their benefits through a configurable framework and real-world validation.

Contribution

It introduces a novel factorized flow map representation and a flexible framework for analyzing their impact on deep reinforcement learning in robotics.

Findings

Factorized optical flows enhance RL agent performance.

Framework effectively compares different mid-level representations.

Real-world validation confirms the practical benefits.

Abstract

In this paper, we introduce a new concept of incorporating factorized flow maps as mid-level representations, for bridging the perception and the control modules in modular learning based robotic frameworks. To investigate the advantages of factorized flow maps and examine their interplay with the other types of mid-level representations, we further develop a configurable framework, along with four different environments that contain both static and dynamic objects, for analyzing the impacts of factorized optical flow maps on the performance of deep reinforcement learning agents. Based on this framework, we report our experimental results on various scenarios, and offer a set of analyses to justify our hypothesis. Finally, we validate flow factorization in real world scenarios.