Are Learning-Based Approaches Ready for Real-World Indoor Navigation? A Case for Imitation Learning

TL;DR

This paper investigates the use of imitation learning for indoor robot navigation, comparing it to traditional methods, and finds that IL can be effective but faces challenges in dynamic environments.

Contribution

It demonstrates the viability of imitation learning for indoor navigation using multimodal sensor data and compares it to traditional methods in real-world settings.

Findings

Multimodal IL outperforms traditional methods in most static scenarios.

Challenges remain in dynamic environments due to limited demonstration diversity.

IL shows potential as a practical and adaptable navigation approach.

Abstract

Traditional indoor robot navigation methods provide a reliable solution when adapted to constrained scenarios, but lack flexibility or require manual re-tuning when deployed in more complex settings. In contrast, learning-based approaches learn directly from sensor data and environmental interactions, enabling easier adaptability. While significant work has been presented in the context of learning navigation policies, learning-based methods are rarely compared to traditional navigation methods directly, which is a problem for their ultimate acceptance in general navigation contexts. In this work, we explore the viability of imitation learning (IL) for indoor navigation, using expert (joystick) demonstrations to train various navigation policy networks based on RGB images, LiDAR, and a combination of both, and we compare our IL approach to a traditional potential field-based navigation method. We evaluate the approach on a physical mobile robot platform equipped with a 2D LiDAR and a camera in an indoor university environment. Our multimodal model demonstrates superior navigation capabilities in most scenarios, but faces challenges in dynamic environments, likely due to limited diversity in the demonstrations. Nevertheless, the ability to learn directly from data and generalise across layouts suggests that IL can be a practical navigation approach, and potentially a useful initialisation strategy for subsequent lifelong learning.