E(2)同変グラフプランニングによるナビゲーション
E(2)-Equivariant Graph Planning for Navigation
2Dナビゲーションを幾何グラフ上のプランニングとして定式化し、E(2)同変なメッセージパッシングで価値反復を行う手法を提案。多カメラ入力にも対応し、学習効率・安定性・汎化性能の向上を実証した。
著者: Linfeng Zhao, Hongyu Li, Taskin Padir, Huaizu Jiang, Lawson L. S. Wong
分類: cs.RO, cs.AI
原文アブストラクト
Learning for robot navigation presents a critical and challenging task. The scarcity and costliness of real-world datasets necessitate efficient learning approaches. In this letter, we exploit Euclidean symmetry in planning for 2D navigation, which originates from Euclidean transformations between reference frames and enables parameter sharing. To address the challenges of unstructured environments, we formulate the navigation problem as planning on a geometric graph and develop an equivariant message passing network to perform value iteration. Furthermore, to handle multi-camera input, we propose a learnable equivariant layer to lift features to a desired space. We conduct comprehensive evaluations across five diverse tasks encompassing structured and unstructured environments, along with maps of known and unknown, given point goals or semantic goals. Our experiments confirm the substantial benefits on training efficiency, stability, and generalization. More details can be found at the project website: https://lhy.xyz/e2-planning/.
