Reproduce “BiPOP: Partial-Order Planning for Long-Horizon Bimanual Robotic Manipulation” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Kun Chu, Xufeng Zhao, Jan-Gerrit Habekost, Cornelius Weber, Stefan Wermter
BiPOP coordinates long-horizon bimanual tasks by using an LLM to interpret instructions and scenes and a temporal symbolic planner to check action constraints and schedule actions. In simulated ServeWater, it succeeds on 100% of 100 instances, compared with 87% for open-loop o3 and 45% for GPT-4.1 ReAct. Its guarantee is at task-planning level; it does not establish collision-free trajectories.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.