FetchMan: Learning Visual Humanoid Loco-Manipulation Policies from Simulated Experiences

Source: arXiv cs.RO·2026-09-01 12:00Publish time unverified

arXiv:2608.17027v2 Announce Type: replace Abstract: Visual loco-manipulation policies that can generalize to novel scenes and objects have long been a goal of robotics research. However, today's data-hungry algorithms make collecting sufficient demonstrations a struggle for tabletop manipulation, and even more so for humanoids that must also walk and balance.