Less Is More: Scalable Visual Navigation from Limited Data

TL;DR

This paper introduces LiMo, a transformer-based visual navigation system that effectively learns goal-conditioned trajectories from limited data by augmenting human demonstrations with synthetic planner-generated trajectories, enhancing data efficiency and robustness.

Contribution

The work demonstrates how classical geometric planners can generate synthetic data to improve visual navigation policies trained on limited demonstrations, reducing the need for extensive data collection.

Findings

Augmenting limited demonstrations with planner-generated trajectories improves performance.

Dataset diversity and scale significantly impact navigation success.

Robust navigation can be achieved with strategic data curation rather than more data.

Abstract

Imitation learning provides a powerful framework for goal-conditioned visual navigation in mobile robots, enabling obstacle avoidance while respecting human preferences and social norms. However, its effectiveness depends critically on the quality and diversity of training data. In this work, we show how classical geometric planners can be leveraged to generate synthetic trajectories that complement costly human demonstrations. We train Less is More (LiMo), a transformer-based visual navigation policy that predicts goal-conditioned SE(2) trajectories from a single RGB observation, and find that augmenting limited expert demonstrations with planner-generated supervision yields substantial performance gains. Through ablations and complementary qualitative and quantitative analyses, we characterize how dataset scale and diversity affect planning performance. We demonstrate real-robot deployment and argue that robust visual navigation is enabled not by simply collecting more demonstrations, but by strategically curating diverse, high-quality datasets. Our results suggest that scalable, embodiment-specific geometric supervision is a practical path toward data-efficient visual navigation.