Dhruv Metha Ramesh, Tri-An Nguyen, Junshan Huang, Jingjin Yu, Kostas E. Bekris, Abdeslam Boularias · 2026 · arXiv
DOI: 10.48550/arXiv.2610.28362 · open-access PDF
A navegação entre obstáculos móveis (NAMO) requer que um robô reorganize objetos que bloqueiam o acesso ao seu objetivo. Para um robô empurrando, o progresso pode depender de um empurrão de configuração que permite uma abertura posterior sem limpar uma passagem. O número de seqüências candidatas pode crescer exponencialmente com o comprimento da seqüência, tornando a pesquisa baseada na física dispendiosa. Introduzimos o Push Action Ranker (SPAR), treinado em pesquisa, que aprende a priorizar os primitivos empurrados dos resultados de pesquisa simulados.
Machine translation (Meta M2M100); the English original follows.
Navigation Among Movable Obstacles (NAMO) requires a robot to rearrange objects that block access to its goal. For a pushing robot, progress may depend on a setup push that enables a later opening without itself clearing a passage. The number of candidate sequences can grow exponentially with sequence length, making physics-based search costly. We introduce the Search-trained Push Action Ranker (SPAR), which learns to prioritize push primitives from simulated search outcomes. Searches over sequences of up to two pushes per object provide supervision that distinguishes immediate openings from s
Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese
GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.28362, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txtSources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange