Mobile Robots through Task-Based Human Instructions using Incremental Curriculum Learning

TL;DR

This paper presents a novel approach combining incremental curriculum learning with deep reinforcement learning to improve mobile robot navigation based on human instructions, leading to better generalization and training efficiency.

Contribution

It introduces a curriculum-based training framework that systematically enhances robot understanding of complex instructions, advancing the integration of ICL and DRL for robotic navigation.

Findings

Robots trained with ICL outperform those without in navigation tasks.

The approach improves training efficiency and generalization in dynamic environments.

Structured learning progressions benefit robotic instruction interpretation.

Abstract

This paper explores the integration of incremental curriculum learning (ICL) with deep reinforcement learning (DRL) techniques to facilitate mobile robot navigation through task-based human instruction. By adopting a curriculum that mirrors the progressive complexity encountered in human learning, our approach systematically enhances robots' ability to interpret and execute complex instructions over time. We explore the principles of DRL and its synergy with ICL, demonstrating how this combination not only improves training efficiency but also equips mobile robots with the generalization capability required for navigating through dynamic indoor environments. Empirical results indicate that robots trained with our ICL-enhanced DRL framework outperform those trained without curriculum learning, highlighting the benefits of structured learning progressions in robotic training.