事前学習済み視覚言語埋め込みだけでロボットナビゲーションは可能か?
Can Pretrained Vision-Language Embeddings Alone Guide Robot Navigation?
凍結した視覚言語モデルの埋め込みのみを用いて模倣学習でナビゲーションポリシーを訓練し、言語指定目標への到達成功率74%を達成したが、長期計画や空間推論に課題があることを示した。
詳しい要約
1. どんなもの?
2. 先行研究と比べてどこがすごい?
3. 技術・手法の肝は?
4. どうやって有効だと検証した?
5. 議論はある?
6. 次に読むべき論文は?
※ AIが要旨から生成した要約です。正確性は原文をご確認ください。
著者: Nitesh Subedi, Adam Haroon, Shreyan Ganguly, Samuel T. K. Tetteh, Prajwal Koirala, Cody Fleming, Soumik Sarkar
分類: cs.RO
原文アブストラクト
Foundation models have revolutionized robotics by providing rich semantic representations without task-specific training. While many approaches integrate pretrained vision-language models (VLMs) with specialized navigation architectures, the fundamental question remains: can these pretrained embeddings alone successfully guide navigation without additional fine-tuning or specialized modules? We present a minimalist framework that decouples this question by training a behavior cloning policy directly on frozen vision-language embeddings from demonstrations collected by a privileged expert. Our approach achieves a 74% success rate in navigation to language-specified targets, compared to 100% for the state-aware expert, though requiring 3.2 times more steps on average. This performance gap reveals that pretrained embeddings effectively support basic language grounding but struggle with long-horizon planning and spatial reasoning. By providing this empirical baseline, we highlight both the capabilities and limitations of using foundation models as drop-in representations for embodied tasks, offering critical insights for robotics researchers facing practical design tradeoffs between system complexity and performance in resource-constrained scenarios. Our code is available at https://github.com/oadamharoon/text2nav
